April 2010 Volume 90 Number 4
Research Reports 476
Perturbation-Based Balance-Training Program for Older Adults
493
Health-Related Quality of Life and RoboticAssisted Therapy for Hand Function
592
509
Supervised Exercise Program for People With Bleeding Disorders and Hemophilic Arthritis
Perspective
527
Intensive Physical Training After Stroke
602
538
Electromyographic Comparison of Exercises With Dumbbells and Elastic Tubing
CARE V Conference Series
550
Assessment of Risk of Recurrent Falls in Elderly People
561
Kinematics of Rising From a Chair
572
Construct Validity of Muscle Force Tests
cover_04.10.indd 1
581
Cold Modalities and Nerve Conduction
Case Report Therapeutic Exercise Program for Patients With Osteoarthritis of the Hip
Management of Chronic Fatigue Syndrome/ Myalgic Encephalomyelitis
615
Qualitative Research Ethics
629
Continuing Professional Development and Physical Therapists’ Roles in Arthritis Management
3/10/10 10:54 AM
in the battle for hard-surface disinfection, you’ll need
BROAD-SPECTRUM POWER
A powerful disinfectant safe for sensitive equipment... That’s Healthcare facility disinfectants need to be aggressive agents against a broad spectrum of pathogens. Yet, many times over, that aggression extends beyond the realm of microorganisms and brutalizes sensitive equipment, often voiding warranties and rendering your costly investment useless. That’s where PROTEX Disinfectant Spray is different. Protex is an EPA approved, one-step spray that effectively kills a wide range of bacteria, viruses, and fungi including: • MRSA
• HIV
• Coronavirus • E-coli • VISA
NOW AVAILABLE IN 32 OZ. TRIGGER-SPRAY BOTTLE Protex Disinfectant Spray is available in 12 oz. spray and 32 oz. trigger-spray bottles and is currently available for sale only in USA and Puerto Rico.
• VRE
• HBV
• Salmonella • HCV
• Herpes • Influenza
• Norwalk Virus
And since it’s alcohol-free, PROTEX can be used on a variety of non-porous surfaces such as ultrasound transducers and probes, mammography compressor plates, other non-surgical surfaces.
one step disinfection and sensitive equipment safety with disinfectant spray. Visit www.parkerlabs.com/protex for a complete list of uses. ISO 13485:2003
Ads_4.10.indd 2
3/9/10 11:33 AM
Recommend TEMPUR-PEDIC and grow your practice with more profit! ®
The
Pro-HealthCore Bed
D[ Tempur-Pedic®
RECOMMEND our products...
Pro-Pak includes: full product information and patient catalogs!
FREE Info 2.75 FREE Pillow with your first order! Up to a $100.00 value.
PROVIDE solutions to your patients...
INCREASE YOUR PROFIT... Offer your patients the clinically proven† benefits of Tempur-Pedic products and increase your bottom line at the same time. ,QKPVJG6GORWT2GFKE 2TQHGUUKQPCN2TQITCOVQFC[Visit TempurMed.com to read our extensive research studies.
NO minimums, NO inventory, NO reimbursement—sell what you want, when you want. Call for your FREE PRO PAK today!
800-510-8623
Source: 2009 Mattress A&U Study © 2010 Tempur-Pedic Management, Inc
03
Ads_4.10.indd 469
3/9/10 11:33 AM
Find the Things That You Need. Virtually Any Time. Virtually Any Place!
www.apta.org/store See what’s new from APTA!
APTA’s online store features almost all the things that PTs, PTAs, and students of physical therapy need: practice essentials, guidelines and resources, home-study materials, patient education information, clinicians’ kits, self-assessment tools, community outreach materials, labcoats, personalized logowear, great gifts, and so much more! 9
Case
Bookmark this very important address right now. Then pay a quick visit whenever you need to make a purchase for yourself, your staff, your colleagues, or your students. ap y Th er
rg ic al Physww w. be ap t.o
A ss oc
ia ti on
ic an
Ads_4.10.indd 474
cover2
5/20/0
2:15
PM
Page
1
More at www.apta.org/ crmanual
A HO
MAN W- T O
UAL
for
CLIN
ICIAN
S
Edition Third
by Edited A D, FAPT
, PT, Ph
wen e McE
Iren
™
MR D D %HS 3 in mot A m er
book
Guidelin
es
Provi for the sion Physi cal Th of in the H erapy ome Se cond Ed ition
3/15/10 1:49 PM
Physical Therapy Journal of the American Physical Therapy Association
Editorial Office Managing Editor / Associate Director of Publications: Jan P. Reynolds,
[email protected] PTJ Online Editor / Assistant Managing Editor: Steven Glaros Associate Editor: Stephen Brooks, ELS Production Manager: Liz Haberkorn
Editor in Chief Rebecca L. Craik, PT, PhD, FAPTA, Philadelphia, PA
[email protected] Deputy Editor in Chief Daniel L. Riddle, PT, PhD, FAPTA, Richmond, VA
Editor in Chief Emeritus Jules M. Rothstein, PT, PhD, FAPTA (1947–2005)
Steering Committee
Permissions / Reprint Coordinator: Michele Tillson
Anthony Delitto, PT, PhD, FAPTA (Chair), Pittsburgh, PA; J. Haxby Abbott, PhD, MScPT, DipGrad, FNZCP, Dunedin, New Zealand; Joanell Bohmert, PT, MS, Mahtomedi, MN; Alan M. Jette, PT, PhD, FAPTA, Boston, MA; Charles Magistro, PT, FAPTA, Claremont, CA; Ruth B. Purtilo, PT, PhD, FAPTA, Boston, MA; Julie Whitman, PT, DSc, OCS, Westminster, CO
Advertising Manager: Julie Hilgenberg
Editorial Board
Manuscripts Coordinator: Karen Darley
Director of Publications: Lois Douthitt
APTA Executive Staff Senior Vice President for Communications: Felicity Feather Clancy Chief Financial Officer: Rob Batarla Chief Executive Officer: John D. Barnes
Advertising Sales Ad Marketing Group, Inc 2200 Wilson Blvd, Suite 102-333 Arlington, VA 22201 703/243-9046, ext 102 President / Advertising Account Manager: Jane Dees Richardson
Board of Directors President: R. Scott Ward, PT, PhD Vice President: Paul A. Rockar Jr, PT, DPT, MS Secretary: Babette S. Sanders, PT, MS Treasurer: Connie D. Hauser, PT, DPT, ATC Speaker of the House: Shawne E. Soper, PT, DPT, MBA Vice Speaker of the House: Laurita M. Hack, PT, DPT, MBA, PhD, FAPTA Directors: Sharon L. Dunn, PT, PhD, OCS; Kevin L. Hulsey, PT, DPT, MA; Dianne V. Jewell, PT, DPT, PhD, CCS, FAACVPR; Aimee B. Klein, PT, DPT, DSc, OCS; Kathleen K. Mairella, PT, DPT, MA; Stephen C.F. McDavitt, PT, DPT, MS, FAAOMPT; Lisa K. Saladin, PT, PhD; Mary C. Sinnott, PT, DPT, MEd; Nicole L. Stout, PT, MPT, CLT-LANA
Rachelle Buchbinder, MBBS(Hons), MSc, PhD, FRACP, Malvern, Victoria, Australia; W. Todd Cade, PT, PhD, St. Louis, MO; James Carey, PT, PhD, Minneapolis, MN; John Childs, PT, PhD, Schertz, TX; Charles Ciccone, PT, PhD, FAPTA (Continuing Education), Ithaca, NY; Joshua Cleland, PT, DPT, PhD, OCS, FAAOMPT, Concord, NH; Janice J. Eng, PT/OT, PhD, Vancouver, BC, Canada; James C. (Cole) Galloway, PT, PhD, Newark, DE; Steven Z. George, PT, PhD, Gainesville, FL; Kathleen Gill-Body, PT, DPT, NCS, Boston, MA; Paul J.M. Helders, PT, PhD, PCS, Utrecht, The Netherlands; Maura D. Iversen, PT, ScD, MPH, Boston, MA; Diane U. Jette, PT, DSc, Burlington, VT; Christopher Maher, PT, PhD, Lidcombe, NSW, Australia; Christopher J. Main, PhD, FBPsS, Keele, United Kingdom; Kathleen Kline Mangione, PT, PhD, GCS, Philadelphia, PA; Patricia Ohtake, PT, PhD, Buffalo, NY; Carolynn Patten, PT, PhD, Gainesville, FL; Linda Resnik, PT, PhD, OCS, Providence, RI; Kathleen Sluka, PT, PhD, Iowa City, IA; Patty Solomon, PT, PhD, Hamilton, Ont, Canada
Statistical Consultants Steven E. Hanna, PhD, Hamilton, Ont, Canada; John E. Hewett, PhD, Columbia, MO; Hang Lee, PhD, Boston, MA; Xiangrong Kong, PhD, Baltimore, MD; Paul Stratford, PT, MSc, Hamilton, Ont, Canada; Samuel Wu, PhD, Gainesville, FL
Committee on Health Policy and Ethics Linda Resnik, PT, PhD, OCS (Chair), Providence, RI; Janet Freburger, PT, PhD, Chapel Hill, NC; Alan Jette, PT, PhD, FAPTA, Boston, MA; Michael Johnson, PT, PhD, OCS, Philadelphia, PA; Justin Moore, PT, DPT, Alexandria, VA; Ruth Purtilo, PT, PhD, FAPTA, Boston, MA
Linking Evidence And Practice Advisory Group Rachelle Buchbinder, MBBS(Hons), MSc, PhD, FRACP, Malvern, Victoria, Australia (Co-Chair); Diane U. Jette, PT, DSc, Burlington, VT (Co-Chair); W. Todd Cade, PT, PhD, St. Louis, MO; Christopher Maher, PT, PhD, Lidcombe, NSW, Australia; Kathleen Kline Mangione, PT, PhD, GCS, Philadelphia, PA; David Scalzitti, PT, DPT, PhD, Alexandria, VA
The Bottom Line Committee Eric Robertson, PT, DPT, OCS; Joanell Bohmert, PT, MS; Lara Boyd, PT, PhD; James Cavanaugh IV, PT, PhD, NCS; Todd Davenport, PT, DPT, OCS; Ann Dennison, PT, DPT, OCS; William Egan, PT, DPT, OCS; Helen Host, PT, PhD; Evan Johnson, PT, DPT, MS, OCS, MTC; M. Kathleen Kelly, PT, PhD; Catherine Lang, PT, PhD; Tara Jo Manal, PT, MPT, OCS, SCS; Kristin Parlman, PT, DPT, NCS; Susan Perry, PT, DPT, NCS; Maj Nicole H. Raney, PT, DSc, OCS, FAAOMPT; Rick Ritter, PT; Kathleen Rockefeller, PT, MPH, ScD; Michael Ross, PT, DHS, OCS; Katherine Sullivan, PT, PhD; Mary Thigpen, PT, PhD; Jamie Tomlinson, PT, MS; Brian Tovin, DPT, MMSc, SCS, ATC, FAAOMPT; Nancy White, PT, MS, OCS; Julie Whitman, PT, DSc, OCS
472 ■ Physical Therapy Volume 90 Number 4
Masthead_4.10.indd 472
April 2010
3/10/10 9:51 AM
Subscriptions
Physical Therapy (PTJ) (ISSN 00319023) is published monthly by the American Physical Therapy Association (APTA), 1111 North Fairfax Street, Alexandria, VA 22314-1488, at an annual subscription rate of $12 for members, included in dues. Nonmember rates are as follows: Individual (inside USA)— $99; individual (outside USA)—$119 surface mail, $179 air mail. Institutional (inside USA)—$129; institutional (outside USA)—$149 surface mail, $209 air mail. Periodical postage is paid at Alexandria, VA, and at additional mailing offices. Postmaster: Send address changes to Physical Therapy, 1111 North Fairfax Street, Alexandria, VA 22314-1488. Single copies: $15 USA, $15 outside USA; with the exception of January 2001: $50 USA, $70 outside USA. All orders payable in US currency. No replacements for nonreceipt after a 3-month period has elapsed. Canada Post International Publications Mail Product Sales Agreement No. 0055832.
Members and Subscribers Send changes of address to: APTA, Attn: Membership Dept, 1111 North Fairfax St, Alexandria, VA 22314-1488. Subscription inquiries: 703/684-2782, ext 3124. PTJ is available in a special format for readers who are visually impaired. For information, contact APTA’s Membership Department at 703/684-2782, ext 3124.
Mission Statement
Physical Therapy (PTJ) engages and inspires an international readership on topics related to physical therapy. As the leading international journal for research in physical therapy and related fields, PTJ publishes innovative and highly relevant content for both clinicians and scientists and uses a variety of interactive approaches to communicate that content, with the expressed purpose of improving patient care.
Readers are invited to submit manuscripts to PTJ. PTJ’s content—including editorials, commentaries, letters, and book reviews—represents the opinions of the authors and should not be attributed to PTJ or its Editorial Board. Content does not reflect the official policy of APTA or the institution with which the author is affiliated, unless expressly stated.
Masthead_4.10.indd 473
Full-text articles are available for free at ptjournal.apta.org 12 months after the publication date. Full text also is provided through DataStar, Dialog, EBSCOHost Academic Search, Factiva, InfoTrac, ProFound, and ProQuest.
Reprints
PTJ Online at ptjournal.apta.org PTJ Online is available via RSS feeds. PTJ posts articles ahead of print and rapid reader responses to articles. Articles, letters to the editor, and editorials are available in full text starting with Volume 79 (1999) and in searchable PDF format starting with Volume 60 (1980). Entire issues are available online beginning with Volume 86 (2006) and include additional data, video clips, and podcasts.
Indexing and Document Delivery
Readers should direct requests for reprints to the corresponding author of the article. Students and other academic customers may receive permission to reprint copyrighted material from this publication by contacting the Copyright Clearance Center Inc, 222 Rosewood Dr, Danvers, MA 01923. Authors who want reprints should contact June Billman, Cadmus Communications, at 800/4875625, or
[email protected]. Nonacademic institutions needing reprint permission information should go to ptjournal.apta.org/misc/terms.dtl.
Advertising
PTJ is indexed and/or abstracted by Abridged Index Medicus, Abstracts of Health Care Management Studies, AgeLine, Allied and Complementary Medicine Database (AMED), Bibliography of Developmental Medicine and Child Neurology, Current Contents, Cumulative Index to Nursing and Allied Health Literature (CINAHL), EMBASE/Exerpta Medica, Exceptional Child Education Resources, Focus on: Sports Science and Medicine, General Science Index (GSI), Health Index, Hospital and Health Administration Index, Index Medicus, Inpharma Weekly, International Nursing Index, ISR, Medical & Surgical Dermatology, MEDLINE, Neuroscience Citation Index, Personal Alert: Automatic Subject Citation Alert (ASCA), Pharmacoeconomics and Outcomes News, Physical Education Index, Reactions Weekly, RECAL Bibliographic Database, Science Citation Index (SCI), Social Sciences Citation Index (SSCI), and SportsS. Article abstracts are available online at ptjournal.apta.org (1980 through present) and via DataStar, Dialog, FirstSearch, Information Access, Ovid
April 2010
Technologies. Ingenta provides online document delivery for articles published since September 1988.
Advertisements are accepted by PTJ when they conform to the ethical standards of the American Physical Therapy Association. PTJ does not verify the accuracy of claims made in advertisements, and acceptance does not imply endorsement by PTJ or the Association. Acceptance of advertisements for professional development courses addressing advanced-level competencies in clinical specialty areas does not imply review or endorsement by the American Board of Physical Therapy Specialties.
Statement of Nondiscrimination APTA prohibits preferential or adverse discrimination on the basis of race, creed, color, gender, age, national or ethnic origin, sexual orientation, disability, or health status in all areas including, but not limited to, its qualifications for membership, rights of members, policies, programs, activities, and employment practices. APTA is committed to promoting cultural diversity throughout the profession.
Volume 90 Number 4 Physical Therapy ■ 473
3/10/10 9:57 AM
Editorial Submission Guidelines Get to the Point
D
uring PTJ’s Editorial Board meeting at APTA’s Combined Sections Meeting in San Diego, we discussed both profound and pragmatic topics—from our policy regarding research on healthy subjects (authors should provide a clear rationale for why they conducted research on healthy subjects and why it’s relevant to patients), to the usefulness of abstracts (we will add relevance statements), to a critique of a report on a randomized controlled trial published in PTJ by one of our own Editorial Board members (just because you’re an Editorial Board member doesn’t mean that you spare yourself the exquisite pain of constructive criticism). As always, I learned a great deal at this meeting, and I am in awe of our spectacular, international team. We made at least one decision that authors will be glad to know: we reduced the length and complexity of PTJ’s author guidelines. We’ve stripped down the instructions to make priorities and requirements clear. As a board, we’ve agreed to be rigorous in requiring compliance with CONSORT, PRISM, STARD, STROBE, and QUALRES guidelines. We’ve also created checklists for each category of manuscript so that authors and reviewers will have a ready reference as they write and as they review. Authors are now asked to submit the completed checklist along with their manuscript. This will help make the review process more efficient. Please visit http://ptjournal.apta.org/misc/ifora.dtl to see the changes for yourself, and don’t hesitate to contact the managing editor at
[email protected] if you have any questions or suggestions. Rebecca L. Craik, PT, PhD, FAPTA Editor in Chief [DOI: 10.2522/ptj.2010.90.4.475]
April 2010
editorial_Craik_4.10.indd 475
Volume 90 Number 4 Physical Therapy ■ 475
3/16/10 11:00 AM
Research Report
A. Mansfield, PhD, is Postdoctoral Fellow, Heart and Stroke Foundation Centre for Stroke Recovery, Sunnybrook Health Sciences Centre, Toronto, Ontario, Canada, and Toronto Rehabilitation Institute, Toronto, Ontario, Canada. She was affiliated with the Institute of Medical Science, University of Toronto, Toronto, Ontario, Canada, and the Centre for Studies in Aging, Sunnybrook Health Sciences Centre, at the time the study was conducted. A.L. Peters, MHK, is Research Assistant, Sobell Department of Motor Neuroscience and Movement Disorders, University College London Institute of Neurology, London, United Kingdom. She was affiliated with the Centre for Studies in Aging, Sunnybrook Health Sciences Centre, at the time the study was conducted. B.A. Liu, MD, is Associate Scientist, Clinical Integrative Biology–Brain Sciences Program, Sunnybrook Health Science Centre, and Assistant Professor, Department of Medicine, University of Toronto. B.E. Maki, PhD, is Senior Scientist and Director, Centre for Studies in Aging, Sunnybrook Health Sciences Centre, 2075 Bayview Ave, Toronto, Ontario, M4N 3M5 Canada. He also is Professor, Department of Surgery, University of Toronto, and is affiliated with the Toronto Rehabilitation Institute. Address all correspondence to Dr Maki at:
[email protected]. [Mansfield A, Peters AL, Liu BA, Maki BE. Effect of a perturbationbased balance training program on compensatory stepping and grasping reactions in older adults: a randomized controlled trial. Phys Ther. 2010;90:476 – 491.] © 2010 American Physical Therapy Association Post a Rapid Response or find The Bottom Line: www.ptjournal.org 476
f
Physical Therapy
Effect of a Perturbation-Based Balance Training Program on Compensatory Stepping and Grasping Reactions in Older Adults: A Randomized Controlled Trial Avril Mansfield, Amy L. Peters, Barbara A. Liu, Brian E. Maki
Background. Compensatory stepping and grasping reactions are prevalent responses to sudden loss of balance and play a critical role in preventing falls. The ability to execute these reactions effectively is impaired in older adults.
Objective. The purpose of this study was to evaluate a perturbation-based balance training program designed to target specific age-related impairments in compensatory stepping and grasping balance recovery reactions.
Design. This was a double-blind randomized controlled trial. Setting. The study was conducted at research laboratories in a large urban hospital. Participants. Thirty community-dwelling older adults (aged 64 – 80 years) with a recent history of falls or self-reported instability participated in the study.
Intervention. Participants were randomly assigned to receive either a 6-week perturbation-based (motion platform) balance training program or a 6-week control program involving flexibility and relaxation training.
Measurements. Features of balance reactions targeted by the perturbationbased program were: (1) multi-step reactions, (2) extra lateral steps following anteroposterior perturbations, (3) foot collisions following lateral perturbations, and (4) time to complete grasping reactions. The reactions were evoked during testing by highly unpredictable surface translation and cable pull perturbations, both of which differed from the perturbations used during training.
Results. Compared with the control program, the perturbation-based training led to greater reductions in frequency of multi-step reactions and foot collisions that were statistically significant for surface translations but not cable pulls. The perturbation group also showed significantly greater reduction in handrail contact time compared with the control group for cable pulls and a possible trend in this direction for surface translations. Limitations. Further work is needed to determine whether a maintenance program is needed to retain the training benefits and to assess whether these benefits reduce fall risk in daily life. Conclusion. Perturbation-based training shows promise as an effective intervention to improve the ability of older adults to prevent themselves from falling when they lose their balance.
Volume 90
Number 4
April 2010
Perturbation-Based Balance Training Program for Older Adults
T
he devastating consequences of falls in older adults have been well documented, and much work has aimed to develop effective fall prevention interventions.1,2 Exercise involving a balance component appears to be a promising intervention,3 but results to date have been mixed.4 – 6 It may be possible for balance training programs to achieve greater reductions in falls by targeting specific aspects of balance control that are more directly related to the demands of preventing falls in daily life.7,8 Although the causes of falling are complex, it is ultimately the ability to respond effectively to a sudden, unexpected balance perturbation (either “external,” such as slips or trips, or “internal,” due to selfinitiated movement) that determines whether a fall occurs.9 Of particular importance are balance recovery reactions involving rapid stepping or grasping movements.10 These change-in-support reactions are the only defense against large balance perturbations11 and are common responses to smaller perturbations.12,13 Numerous studies have documented age-related impairments in change-insupport reactions, even in young and relatively healthy elderly individuals (see review10). Many of these impairments have been associated with increased fall risk, determined retrospectively14 –16 and prospectively.17 In one study,17 participants who fell forward or backward during a 1-year fall-monitoring period required more steps than “nonfallers” to recover balance during perturbation tests and were more likely to experience a collision between the swing and stance limbs when responding to mediolateral (ML) perturbations during testing. These “fallers” were more reliant than nonfallers on using arm reactions to recover balance during testing, yet their reach-tograsp reactions were slower. Participants who fell sideways in daily life April 2010
were more likely to show evidence of lateral instability, taking one or more extra lateral steps subsequent to the initial forward or backward step evoked by anteroposterior (AP) perturbations. These results raise the possibility that training targeting these specific aspects of change-insupport reactions may improve the ability to recover balance and help to reduce fall risk. The neural control of volitional movements and perturbation-evoked reactions differs in some fundamental ways.18 –20 These differences suggest that effective training of change-insupport reactions requires administration of balance perturbations, rather than simply training volitional movements. Some researchers have begun to investigate perturbationbased training of change-in-support reactions in people with impaired postural control due to aging or pathology.21–24 However, only one of these studies examined the effects of perturbation-based training on the control of compensatory stepping reactions, and that study was limited to people with Parkinson disease.22 No previous study has addressed the potential to use perturbation-based training to counter specific impairments in compensatory stepping or grasping in a nonclinical older population. This study aimed to evaluate the efficacy of perturbation-based balance training targeting specific age-related impairments in change-in-support balance reactions. The essential feature of any balance perturbation is that it induces relative motion between the center of mass (COM) and the base of support (BOS).20 Our training involved a BOS perturbation method (support-surface translation) that allows perturbation direction and magnitude to be varied unpredictably while avoiding the complications associated with COM perturbation methods (which typically require cables attached to the indi-
vidual).10,19,20 Training effects were evaluated using both surface translation and cable pull perturbations to determine whether training had the desired effects and whether training effects transferred to COM perturbation. Four hypotheses were tested, corresponding to the 4 training targets, which were to reduce: (1) frequency of multi-step reactions, (2) frequency of extra lateral steps during step reactions evoked by AP perturbations, (3) frequency of foot collisions during step reactions evoked by ML perturbations, and (4) time required to complete grasping reactions. We hypothesized that perturbation-based training would lead to greater improvements in these outcomes than a control flexibility and relaxation program.
Method Design Overview The protocol for this double-blind randomized controlled trial has been described previously.7 All participants provided written informed consent prior to participation. Setting and Participants Interventions and assessments were conducted in 2 research laboratories Available With This Article at ptjournal.apta.org • eTable 1: Training Program Overview • eTable 2: Secondary Outcome Measures • Video: “Pretraining to Posttraining Differences in Step Patterns in Response to Anteroposterior (AP) Surface Translations” • Audio Abstracts Podcast This article was published ahead of print on February 18, 2010, at ptjournal.apta.org.
Volume 90
Number 4
Physical Therapy f
477
Perturbation-Based Balance Training Program for Older Adults Table 1. Cohort Descriptors for Participants Who Completed the Studya Variable
Control Group
No. of participants Age (y) Sex (men:women) Height (m) Weight (kg)
14
16
69.1 (3.8)
70.3 (4.7)
Fear of falling (no. of participants)
8:8
.28
1.66 (0.07)
.89
72.7 (9.8)
.66
12/16
.33
9/14
ABCd (% score)
.47
7:7
11/14
c
P
1.66 (0.08) 75 (17.2)
Fall in the past 5 years (no. of participants)b
Perturbation-Based Training Group
92.4 (9.8)
7/15 91.9 (13)
.19 .92
FallScreen fall risk scoree
0.45 (0.47)
0.53 (0.83)
.86
Timed “Up & Go” Test68 score (s)
5.92 (1.34)
6.30 (1.51)
.52
a Values shown are means with standard deviations in parentheses unless otherwise indicated. The P values are for the comparison between the groups based on Fisher exact test for categorical data and one-way analysis of variance for continuous measures. b A fall was defined as an event in which the participant came to rest unintentionally on the ground, floor, or other lower level. Falls involving an altered base of support (eg, while wearing skis or on a bicycle) were excluded. The number reported is the number of participants who fell at least once in the 5 years prior to screening. c Participants were asked “Are you afraid of falling?” The number reported is the number of participants who responded “yes.” d Activities-specific Balance Confidence Scale.69 e The FallScreen fall risk score combines results from the edge contrast sensitivity, proprioception (at the knee), isometric knee extension strength, hand reaction time, and standing on foam with eyes closed tests. A score greater than 2 indicates an increased risk for falling.70
in a large urban hospital. Right-limb– dominant, community-dwelling older adults (aged 64 – 80 years) were recruited via newspaper advertisements, posters in the community, and word of mouth. To target an at-risk population, we included volunteers who reported having a decline in balance, a fall within the last 5 years, or a recent near fall. Volunteers with medical conditions that limited daily activities, that were likely to cause sensory or motor impairments (eg, stroke, diabetes), or that could increase injury risk during balance tests (eg, osteoporosis) were excluded. We also excluded volunteers taking medications affecting balance or causing dizziness (eg, psychotropics) and those who were unable to provide consent or to understand instructions due to cognitive impairment or poor English-language comprehension. Volunteers with possible contraindications to exercise25 obtained permission from their physician prior to participating. To pre478
f
Physical Therapy
Volume 90
the 2 groups on basic demographic factors (age and sex) and to minimize baseline inter-group differences in ability to execute effective stepping and grasping reactions (cutoffs based on previously reported differences between fallers and nonfallers17). The individual who administered both training programs (A.M.) was the only member of the research team aware of training group allocation. A blinded research assistant administered the balance tests and processed any data involving personal judgment. Participants also were blinded, in that they were unaware of which training program was expected to improve balance.
vent potential confounding effects of other exercise programs, we excluded volunteers participating in supervised exercise one or more times per week. Cohort baseline characteristics are listed in Table 1.
The perturbation-based and control training sessions lasted 30 minutes, 3 times per week for 6 weeks. This regimen was based on previous balance training studies that demonstrated improvements in balance control with similar training duration and frequency.27–30 Both training programs involved one-on-one training in the same laboratory. Participants were asked to refrain from initiating other exercise programs, or to otherwise change their activity levels, during the 6-week program.
Randomization and Interventions Participants were assigned, by computerized random-number generation, to either a perturbation-based training group or a control group using stratified-blocked randomization,26 with 4 binary stratification variables (16 strata): (1) age (⬍73 years versus ⱖ73 years), (2) sex, (3) compensatory stepping (⬍1 versus ⱖ1 extra steps), and (4) compensatory grasping (reaction time of ⬍120 milliseconds versus ⱖ120 milliseconds). The stepping and grasping values were determined from baseline testing. Participants were classified into 1 of the 16 strata and were randomly assigned to either group within each stratum. Stratification aimed to match
The perturbation-based program was developed in accordance with principles of motor learning7 (eTab. 1, available at ptjournal.apta.org). Stepping and grasping reactions were evoked during training using a custombuilt motion platform, which translated unpredictably in 4 directions (Fig. 1A). Half of each session (12–15 minutes, ⱖ24 perturbations) was devoted to stepping, and the other half was devoted to grasping. Handrails mounted on the platform were removed during step training, and foot movement was restricted with foam blocks during grasp training. Specific age-related impairments noted earlier were targeted via: (1) complex instructions referring to the whole skill31 (eg, “take as few steps as possible”),
Number 4
April 2010
Perturbation-Based Balance Training Program for Older Adults
Figure 1. Equipment used (A) during balance training and (B) during pretraining and posttraining balance assessment. Both systems delivered perturbations unpredictably in forward, backward, left, and right directions. The training platform (panel A) was driven by pneumatic cylinders, and the magnitude of the platform acceleration was varied (1– 4 m/s2) by adjusting the compressor air pressure. Handrails that were adjustable in height (87–101 cm) and position (37– 42 cm from midline) were installed on the platform during grasping training, and foam blocks (40 cm high) were placed around the feet to deter stepping. The motor-driven motion platform and associated instrumentation used during the assessments (panel B) are described in more detail elsewhere.34,35 A custom-built cable pull perturbation system was mounted on this platform so that the nature of the perturbation (support surface motion versus cable pull) could be varied unpredictably.7 The cable pulls were applied by dropping a weight attached to a belt worn around the pelvis (at the level of the anterior superior iliac spines) via a cable-and-pulley system. Four separate cables were attached to the belt so that the participant was pulled unpredictably forward, backward, left, or right when the weight was dropped, depending on which cable was attached to the weight. The weight-drop apparatus could not be seen by the participants, and a locking mechanism provided an equal amount of slack (⬃2– 4 cm) in each cable prior to the weight drop so that participants could not detect which (if any) of the cables were attached to the weight. Similar to panel A (but not shown in panel B), a cylindrical handrail (height⫽55% of participant’s height, diameter⫽38 mm, length⫽1.63 m) was mounted on the platform (to the right of the participant, 25% of participant’s height from the midline of the body), and foam blocks (40 cm high) were placed around the feet to deter stepping during grasping trials. To deter arm reactions in stepping trials, participants held a lightweight rod behind their back37 and were instructed not to move their arms or release the rod. At the start of each trial, participants either stood or walked in place at the center of the platform, with the feet in a comfortable, standardized position.72 Reprinted with permission of Elsevier from: Maki BE, Cheng KC-C, Mansfield A, et al. Preventing falls in older adults: new interventions to promote more effective change-in-support balance reactions. J Electromyogr Kinesiol. 2008;18:243–254.
April 2010
Volume 90
Number 4
Physical Therapy f
479
Perturbation-Based Balance Training Program for Older Adults (2) physical constraints (foam-rubber blocks preventing lateral crossover steps, thereby reducing risk of foot collisions), and (3) external feedback (handrail contact times) that was gradually faded as training progressed.31,32 Perturbation magnitude was increased progressively according to the capabilities of each individual. In later sessions, participants performed concurrent cognitive and motor tasks (eg, listing words belonging to a category, walking in place; eTab. 1, available at ptjournal.apta.org) prior to the perturbation to better prepare them for reallife balance loss situations where they may be engaged in another task prior to a postural perturbation. Participants wore a safety harness throughout perturbation-based training.
with specific equipment used during training, a different multi-axis motion platform delivered surface-translation perturbations during balance testing.34,35 This platform differed in spatial features (semi-enclosed by walls, larger floor surface, location of handrail) and allowed distinctly different patterns of motion to be delivered (with variation in acceleration, velocity, displacement, and duration; Fig. 1). We also performed tests involving multi-axis cable pulls for reasons explained earlier. The cable pull apparatus was mounted on the motion platform so perturbation type, direction, and timing could be varied unpredictably from trial to trial7,36 (Fig. 1B). A safety harness was worn throughout balance testing.
The control program involved limited physical activity to avoid physiological changes that could affect balance control. Participants completed passive muscular relaxation exercises33 1 day per week and flexibility exercises on the other 2 days. Each flexibility session involved a 10minute, low-intensity warm-up (heart rate ⬍60% of maximum) followed by 15 minutes of stretching exercises, half of which were performed seated (7–9 different exercises, held for 15 seconds twice on each side). The warm-up involved walking in place, reaching, and trunk twists. Muscles and muscle groups targeted during stretching were: quadriceps, calves, hip abductors and adductors, hamstrings, latissimus dorsi, pectorals, abdominals, triceps, deltoids, wrist flexors and extensors, trapezius, and sternocleidomastoid.
The same protocol was used during preintervention and postintervention test sessions. After 12 initial familiarization trials, participants performed 3 trial blocks focusing on: (1) stepping evoked by AP perturbation of stance, (2) stepping evoked by ML perturbation while walking in place, and (3) grasping evoked by backward perturbation of stance. The walk-in-place task was used to increase the likelihood that collisions between the step and stance limbs would occur during lateral step reactions.36,37 Participants completed a cognitive task (counting backward by 3’s) during trial blocks 1 and 3; these tasks were included to distract participants from consciously attending to perturbations, and training effects on dual tasking were not assessed. Perturbation magnitudes and waveforms used during testing were similar to those used in previous studies that investigated age-related differences in change-in-support reactions.38,39 Details of perturbations, task conditions, and instructions are provided in Table 2 and Figure 1.
Outcome Measures Perturbation-evoked balance reactions were evaluated within 1 week prior to the start of training and within 1 week after program completion. To avoid the possibility that participants in the perturbation-based training group would perform better after intervention due to familiarity 480
f
Physical Therapy
Volume 90
Number 4
During each trial, 3 forceplates (1 Kistler force sensor [model 9281]* and 2 AMTI force platforms [model OR6 –5]†) recorded ground reaction forces, a 3-dimensional motion capture system (Peak Motus, version 8)‡ recorded coordinates of markers on the feet and arms, and 4 video cameras recorded gross motor behaviors. Load cells recorded cable pull force and safety harness loading. Accelerometers and a linear potentiometer measured motion platform acceleration and displacement. The accelerometer signals were used to correct forceplate measurements for inertial artifacts arising from platform motion.40 Surface electromyographic (EMG) activity was recorded (TeleMyo 900)§ from the biceps brachii muscle. Signals were low-pass filtered (10 Hz) prior to sampling at 200 Hz, except for the video data (sampled at 60 Hz), EMG data (bandpass filtered at 10 –500 Hz prior to sampling at 1,000 Hz), and motion analysis data (sampled at 200 Hz, then digitally low-pass filtered with cutoff frequency determined using residual analysis41). The outcome measures required to test the 4 primary hypotheses were: (1) frequency of multi-step reactions, (2) frequency of extra lateral steps during step reactions evoked by AP perturbation, (3) frequency of foot collisions during lateral step reactions, and (4) handrail contact time. To help understand training effects in the primary outcome measures, we performed exploratory analyses of other characteristics of change-insupport reactions. For step reaction trials, these secondary measures included step pattern, timing, and dis* Kistler-Instrumente AG, Eulachstrasse 22, Winterthur, CH-8408 Switzerland. † Advanced Medical Technology Inc, 176 Waltham St, Watertown, MA 02472-4800 ‡ Peak Performance Technologies Inc, 7388 S Revere Pkwy, Suite 901, Centennial, CO 80112. § Noraxon USA Inc, 13430 N Scottsdale Rd, Suite 104, Scottsdale, AZ 85254.
April 2010
Perturbation-Based Balance Training Program for Older Adults Table 2. Details of Protocol Used in the Preintervention and Postintervention Balance Test Sessionsa Perturbations and Numbers of Trials Analyzed Trials
Additional Trials
Trial block 1: stance, arm motion restricted, instructed to react naturally but minimize number of steps if need to step
Task Condition
Stepping evoked by AP perturbation
Focus of Analysis
5 backward translationsb 3 forward translations 5 forward cable pullsc 3 backward cable pulls Total⫽16 trials
2 ML translations (L, R) 4 translations, second waveform (F, B, L, R)d 2 ML cable pulls (L, R) Total⫽8 trials
Trial block 2: walking in place,e arm motion restricted, instructed to react naturally but minimize number of steps if need to step
Stepping evoked by ML perturbation
5 leftward translations 3 rightward translations 5 rightward cable pulls 3 leftward cable pulls Total⫽16 trials
2 AP translations (F, B) 4 translations, second waveform (F, B, L, R)d 2 AP cable pulls (F, B) Total⫽8 trials
Trial block 3: stance, foot motion restricted, instructed to recover balance by grasping handrail at right of participant as quickly as possible
Grasping evoked by backward perturbation
5 forward translations 5 backward cable pulls Total⫽10 trials
4 backward translations 4 forward cable pulls Total⫽8 trials
a During each trial block, the listed surface translation and cable pull perturbations were delivered in an unpredictable randomized sequence, in the directions indicated (F⫽forward, B⫽backward, L⫽left, R⫽right; AP⫽anteroposterior, and ML⫽mediolateral). The additional trials were included to increase unpredictability and were not analyzed. b The surface translation perturbations comprised a 300-millisecond acceleration pulse followed immediately by a 300-millisecond deceleration pulse. Each pulse was approximately square, with an amplitude of 2.0 m/s2 for forward translations (evoking backward falling motion) and 3.0 m/s2 for other translation directions (backward, left, right). The displacement and peak velocity were 0.18 m and 0.6 m/s, respectively, for forward translations and 0.27 m and 0.9 m/s, respectively, for other directions. c The cable pull perturbations were applied by dropping a weight equal to 20% of body weight. The drop height was 40 cm for stepping trials and 30 cm for grasping trials. d The second waveform surface translations were included to deter participants from learning to use the platform deceleration to aid in recovering balance.71 This waveform comprised a 200-millisecond acceleration pulse, a 400-millisecond constant velocity interval, and a 200-millisecond deceleration. e Perturbations for the walking-in-place trials were timed to occur at foot-lift of the foot contralateral to the fall direction evoked by the perturbation (eg, right foot-lift for a leftward fall) in order to increase the probability of observing a collision between the step foot and the stance leg.37 The perturbation was delivered after 3 to 8 steps and was triggered when the stance-leg forceplate loading exceeded 90% of body weight.
tance (ML and AP components). Secondary measures for grasp reaction trials included frequency of grasping errors (hand undershoots, overshoots, or collides with the rail) and onset latency of right biceps muscle EMG activity. The behavioral outcome data (step patterns, foot collisions, grasping errors) were obtained from a computer algorithm that combined forceplate and kinematic data and were verified by a blinded investigator. Foot-off and foot-contact times were determined from vertical ground reaction forces (⬍ or ⬎1% of body weight). Kinematic data were used to determine step distance (toe marker displacement) and handrail contact time (wrist marker displacement). Onset latencies for muscle activation were determined by a computer algorithm42 and confirmed by visual inspection. All timing measures were defined relative to perturbation onset (surface acceleration ⬎0.1 m/s2 or cable force ⬎5 N). To April 2010
ensure quality of data processing, data sets were submitted to a postprocessing algorithm that checked for inconsistencies and outliers (⬎3 standard deviations from the mean). Outliers were checked against the raw data and corrected, as necessary. Data Analysis Sample size calculations for repeated-measures analysis of variance (ANOVA)43 were performed using the variance from the initial 10 subjects (5 per training group). Separate calculations were performed to determine the sample size requirements based on stepping reactions (average number of extra steps taken to recover balance) and grasping reactions (handrail contact time) evoked by surface translations. For both calculations, probability of type I error was .05, and probability of type II error was .2. In a previous study, fallers required 0.5 more steps to recover balance and took 50 milliseconds longer
to grasp a handrail, on average, in comparison with nonfallers,17 suggesting that the perturbation-based training may reduce fall risk if it yields net reductions that exceed any reductions in the control group by these amounts. Using these net reduction values (0.5 steps, 50 milliseconds), in combination with the initial variance estimates (standard deviations of 0.44 steps and 39 milliseconds), it was determined that 15 and 12 participants per group were required, respectively. Therefore, we aimed to recruit 15 participants per group. Baseline differences between the perturbation-based training and control groups in primary and secondary outcome measures were assessed via ANOVA and Fisher exact test. Training effects in these measures were determined by using an ANOVA to analyze differences between pretraining and posttraining scores. For frequency variables (which involved
Volume 90
Number 4
Physical Therapy f
481
Perturbation-Based Balance Training Program for Older Adults only one pretraining observation and one posttraining observation per participant), training group (perturbationbased versus control) was the only factor in the ANOVA. For variables involving multiple observations within participant (eg, handrail contact time), participant was included as a random factor in the ANOVA, and pretrainingposttraining difference scores were computed by subtracting the participant’s mean pretraining score from each of the posttraining observations. The pretraining-posttraining difference scores were rank transformed prior to analysis to avoid errors arising from violations of assumptions underlying the ANOVA.44 Separate analyses were completed for each perturbation type (ie, surface translation or cable pull). In the absence of evidence that training effects were dependent upon perturbation direction, trials involving different directions were pooled (ie, forward and backward perturbations in trial block 1, leftward and rightward perturbations in trial block 2); otherwise, each direction was analyzed separately. For exploratory purposes, we analyzed training effects on measures of physical functioning (strength [forcegenerating capacity], reaction time, and standing balance measures from the FallScreen battery45; lower-limb power; mobility; and flexibility), balance confidence, and state anxiety. The criterion level of significance for all analyses was alpha⫽.05. Ninetyfive percent confidence intervals were calculated for all point estimates. All statistical analyses were performed using SAS software, version 9.1.㛳 Role of the Funding Source This study was supported by the Canadian Institutes of Health Research (grant NET-54025), the Ontario Neurotrauma Foundation (summer internship grant ONF2007-PREV-INT452), and the Canadian Foundation 㛳
SAS Institute Inc, PO Box 8000, Cary, NC 27513.
482
f
Physical Therapy
Volume 90
for Innovation. The views contained in this publication are those of the grantees and do not necessarily reflect those of the funding agencies.
Results Participant Recruitment, Screening, and Retention Of 186 volunteers who met the inclusion criteria (between December 2005 and January 2007), 37 were eligible, agreed to participate, and completed baseline balance tests (Fig. 2). Analysis of training effects was based on 30 participants (16 in the perturbation training group, 14 in the control group) who completed the study. The 2 groups showed no differences in baseline descriptors (Tab. 1) and were well-matched (P⬎.25) on the randomization-stratification factors (age and sex [Tab. 1], number of steps and grasping reaction time [Tab. 3]). Baseline between-group differences were not evident in any primary or secondary outcome measure (P⬎.065). Program Adherence and Adverse Effects Twenty-one participants (out of 30) attended all 18 sessions, 8 participants attended 17 sessions, and 1 participant attended 16 sessions. There was no difference in mean number of sessions attended between the control group (17.6) and the perturbationbased training group (17.8). One participant in the perturbation-based training group withdrew due to concerns that training would exacerbate a prior injury; however, no injury was reported. In 6 trials (4 participants), the safety harness was used to prevent a fall to the floor (out of 17,552 training trials completed by 16 participants); no injuries resulted from these incidents. Missing Data One participant in the control group completed only trial block 1 (Tab. 2) during the preintervention session
Number 4
due to technical problems; thus, analyses of ML step reactions were based on 29 participants. Two additional participants (1 in each group) did not complete trial block 3, and technical problems during cable pull trials prevented collection of grasp timing data in 2 other participants in the perturbation-based training group; therefore, analyses of compensatory grasping were based on 27 participants (surface translations) or 25 participants (cable pulls). Training Effects: Primary Outcome Measures Training effects on the 4 features of stepping and grasping reactions targeted by the perturbation-based training program are detailed below. Values presented are means of the pretraining-posttraining changes in each measure, with a negative value indicating a training-related reduction (95% confidence interval in brackets). Additional descriptive statistics are presented in Table 3. Multi-step reactions (trial blocks 1 and 2). Following training, the perturbation-based training group showed a greater reduction in frequency of multi-step reactions compared with the control group; however, this effect was significant only for surface translations (perturbationbased training group: ⫺31% of trials [⫺40, ⫺23], control group: ⫺19% of trials [⫺28, ⫺10]; F1,28⫽4.95, P⫽ .034), not cable pulls (perturbationbased training group: ⫺12% of trials [⫺20, ⫺3], control group: ⫺16% of trials [⫺25, ⫺6]; F1,28⫽0.08, P⫽.78) (Fig. 3A). Extra lateral steps during AP step reactions (trial block 1). There was no training effect for frequency of AP perturbation trials in which additional lateral steps occurred for surface translations (perturbationbased training group: ⫺1% of trials [⫺11, 9], control group: ⫺5% of trials [⫺14, 5]; F1,28⫽0.47, P⫽.50) or April 2010
Perturbation-Based Balance Training Program for Older Adults
Figure 2. Flowchart summarizing participant recruitment, screening, and retention. Reasons for declining to participate (n⫽21) were: time requirements of the study (n⫽13), unwilling to be randomly assigned to group (n⫽3), not enough compensation (n⫽2), health concerns (n⫽1), unwilling to be videotaped (n⫽1), and no reason provided (n⫽1). Medical screening led to exclusion of 74 volunteers on the basis of either a single factor (n⫽70) or multiple factors (n⫽4). These exclusion factors were: osteoporosis (n⫽29), joint replacement or fusion (n⫽14), diabetes (n⫽6), stroke (n⫽4), medications (n⫽4), poor vision (n⫽3), regular use of mobility aids (n⫽3), depression (n⫽3), poor mobility (n⫽3), Parkinson disease (n⫽2), peripheral neuropathy (n⫽2), acute illness (n⫽2), vertigo or dizziness (n⫽2), narcolepsy (n⫽1), self-reported memory problems (n⫽1), standardized Mini-Mental State Examination score of 24 or less (n⫽1), and physician’s recommendation (n⫽1). An additional 17 individuals (not included in the total of 186 participants) contacted us regarding study participation but did not complete the medical screening, as they did not meet the instability inclusion criteria (ie, fall in the last 5 years, recent near fall, or self-reported decline in balance). Analysis of training effects was based on the 30 participants (16 in the perturbation-based training group, 14 in the control group) who completed the study. April 2010
Volume 90
Number 4
Physical Therapy f
483
Perturbation-Based Balance Training Program for Older Adults Table 3. Preintervention and Postintervention Characteristics of Stepping and Grasping Reactions Evoked by Surface Translation and Cable Pull Perturbations for Primary and Secondary Outcome Measuresa Control Group Measure
Preintervention
Postintervention
Perturbation-Based Training Group Effect Size
Preintervention
Postintervention
Effect Size
P
Primary measures All stepping reactions (trial blocks 1 and 2) Frequency of multi-step reactions (% of trials) Surface translation
70 (63, 75)
52 (43, 57)
⫺0.39
68 (62, 74)
37 (31, 42)
⫺0.66
.034
Cable pull
58 (49, 63)
43 (34, 47)
⫺0.32
45 (39, 51)
33 (27, 39)
⫺0.25
.78
18 (10, 25)
13 (7, 20)
⫺0.13
20 (13, 27)
20 (13, 26)
⫺0.02
.50
3 (0, 6)
4 (0, 7)
0 (0, 0)
1 (0, 2)
0.13
.93
Anteroposterior stepping reactions (trial block 1) Frequency of extra lateral steps (% of trials) Surface translation Cable pull
0.05
Mediolateral stepping reactions (trial block 2) Frequency of foot collisions (% of trials) Surface translation Cable pull
54 (44, 63)
40 (31, 50)
⫺0.27
57 (48, 66)
19 (12, 26)
⫺0.86
.0046
9 (3, 14)
5 (1, 9)
⫺0.16
3 (0, 6)
0 (0, 0)
⫺0.26
.70
Grasping reactions (trial block 3) Handrail contact time (ms) Surface translation
539 (517, 561)
525 (509, 541)
⫺0.18
552 (532, 573)
504 (489, 519)
⫺0.62
.088
Cable pull
669 (621, 716)
663 (583, 744)
⫺0.02
643 (614, 673)
523 (510, 536)
⫺1.21
.004
21 (17, 26)
15 (10, 20)
⫺0.25
27 (22, 31)
0 (0)
⫺1.51
.018
7 (5, 9)
6 (4, 8)
⫺0.08
12 (10, 14)
6 (4, 7)
⫺0.55
.27
Secondary measures Anteroposterior stepping reactions (trial block 1) ⬎2 anteroposterior stepsb (% of trials) Surface translation Cable pull Foot-off time (ms) Surface translation
357 (349, 365)
348 (339, 356)
⫺0.21
349 (343, 355)
331 (325, 337)
⫺0.49
.22
Cable pull
471 (457, 485)
443 (427, 458)
⫺0.36
475 (464, 487)
416 (407, 424)
⫺1.02
.029
Surface translation
489 (478, 499)
488 (477, 500)
⫺0.01
495 (486, 505)
483 (474, 592)
⫺0.23
.12
Cable pull
602 (587, 617)
582 (564, 599)
⫺0.23
624 (612, 636)
581 (570, 591)
⫺0.66
.11
Foot-contact time (ms)
Forward fall stepping reactions (trial block 1) Forward step displacement (cm) Surface translation
33.9 (32.2, 35.7)
32.7 (32.3, 34.1)
⫺0.19
37.6 (35.7, 39.4)
34.1 (32.5, 35.6)
⫺0.46
.71
Cable pull
28.9 (27.0, 30.7)
29.4 (27.2, 31.7)
0.06
34.8 (32.6, 37.0)
36.8 (35.0, 38.6)
0.23
.50
Surface translation
2.3 (1.7, 2.9)
2.4 (1.7, 3.1)
0.06
2.5 (1.8, 3.2)
2.7 (2.1, 3.3)
0.07
.58
Cable pull
1.2 (0.6, 1.8)
1.2 (0.5, 2.0)
⬍0.01
1.3 (0.6, 1.9)
1.8 (1.2, 2.4)
0.22
.044
Lateral step displacement (cm)
Backward fall stepping reactions (trial block 1) Backward step displacement (cm) Surface translation
17.0 (14.4, 19.6)
17.7 (15.3, 20.1)
0.09
17.4 (15.1, 19.6)
24.0 (22.0, 26.0)
0.92
.0017
Cable pull
21.2 (18.4, 24.1)
21.1 (18.2, 24.0)
⫺0.01
21.0 (18.3, 23.7)
25.9 (23.2, 28.6)
0.54
.043 (Continued)
484
f
Physical Therapy
Volume 90
Number 4
April 2010
Perturbation-Based Balance Training Program for Older Adults Table 3. Continued Control Group Measure
Preintervention
Postintervention
Perturbation-Based Training Group Effect Size
Preintervention
Postintervention
Effect Size
P
Lateral step displacement (cm) Surface translation
0.3 (⫺0.7, 1.4)
1.2 (0.2, 2.3)
0.27
Cable pull
1.3 (0.5, 2.0)
1.3 (0.5, 2.0)
⬍0.01
0.6 (⫺0.1, 1.4)
2.2 (1.4, 3.0)
0.60
.16
1.1 (0.4, 1.7)
4.6 (3.4, 5.8)
1.09
.0042
Mediolateral stepping reactions (trial block 2) Crossover steps (% of trials)c Surface translation
36 (26, 45)
34 (25, 43)
⫺0.04
28 (20, 35)
9 (4, 15)
⫺0.48
.16
Cable pull
39 (29, 48)
30 (21, 39)
⫺0.18
27 (20, 35)
10 (4, 15)
⫺0.47
.46
Surface translation
153 (146, 162)
142 (135, 149)
⫺0.40
142 (135, 149)
130 (122, 138)
⫺0.38
.99
Cable pull
177 (158, 197)
150 (136, 163)
⫺0.45
177 (165, 189)
138 (128, 147)
⫺0.85
.80
24 (13, 35)
3 (⫺1, 8)
⫺0.62
11 (4, 18)
9 (3, 16)
⫺0.05
.018
0 (0, 0)
2 (⫺2, 6)
0.20
0 (0, 0)
0 (0, 0)
0
.63
Grasping reactions (trial block 3) Biceps muscle latency (ms)
Frequency of grasping errors (% of trials) Surface translation Cable pull
d
a
Values shown are means with 95% confidence intervals in parentheses. The P value is for the analysis of the between-group pretraining to posttraining differences for each measure. b This step pattern involves at least 2 extra steps after the first step, none of which are placed laterally (ie, 3 or more forward- and backward-directed steps; see parts A and C of the video, available at ptjournal.apta.org). c This step pattern involves a step with the swing foot that is placed lateral to the stance foot. d A grasping error is defined by a failure in the initial attempt to contact the handrail due to undershoot, overshoot, or a collision with the handrail.
cable pulls (perturbation-based training group: 1% of trials [⫺1, 2], control group: 1% of trials [⫺4, 5]; F1,28⫽0.01, P⫽.95) (Fig. 3B). Foot collisions during ML step reactions (trial block 2). The perturbation-based training group showed a greater reduction in frequency of trials in which collisions between the swing and stance limbs occurred, but only for surface translations (perturbation-based training group: ⫺38% of trials [⫺49, ⫺27], control group: ⫺13% of trials [⫺27, 0], F1,27⫽9.554, P⫽.0046) and not for cable pulls (perturbation-based training group: ⫺3% of trials [⫺6, 0], control group: ⫺4% of trials [⫺11, 3]; F1,27⫽0.15, P⫽.70) (Fig. 3C). Handrail contact time (trial block 3). The perturbation-based training group showed a greater trainingrelated reduction in handrail contact time compared with the control April 2010
group. This effect was statistically significant for cable pulls (perturbationbased training group: ⫺120 milliseconds [⫺154, ⫺87], control group: ⫺5 milliseconds [⫺97, 87]; F1,23⫽10.25, P⫽.0040), but not for surface translations (perturbationbased training group: ⫺49 milliseconds [⫺74, ⫺23], control group: ⫺14 milliseconds [⫺41, 14]; F1,25⫽3.15, P⫽.088) (Fig. 3D). The latter analysis was confounded by the occurrence of grasping errors (Tab. 3), which delayed handrail contact. When such trials were excluded, the training effect for surface translations was significant, showing greater improvement in the perturbation-based training group compared with the control group (perturbation-based training group: ⫺46 milliseconds [⫺66, ⫺26], control group: 11 milliseconds [⫺12, 34]; F1,25⫽15.12, P⫽.0007).
Training Effects: Secondary Outcome Measures Results for secondary stepping and grasping measures are summarized in Table 3. Both AP and ML perturbation trials showed evidence of changes in step pattern, but only for surface translations. For AP perturbations (trial block 1), perturbationbased training led to a complete elimination of responses involving 3 or more AP steps (from 27% of the trials to 0%), whereas the change was much smaller in the control group (from 21% of the trials to 15%) (see video, available at ptjournal.apta.org). The perturbationbased training group showed a greater increase than the control group in lateral displacement of the first step, but only for cable pulls, and in AP step displacement for “backward fall” trials for both perturbation types. The perturbation-based training group showed a greater reduction than the control group in time to foot-off, but
Volume 90
Number 4
Physical Therapy f
485
Perturbation-Based Balance Training Program for Older Adults
Figure 3. Effect of training on primary measures: (A) frequency of trials with multi-step reactions, (B) frequency of trials with extra lateral steps, (C) frequency of trials with foot collisions, and (D) handrail contact time, as determined before and after intervention for participants in the perturbation-based training group (PERT) and the control group (CON). Panel A shows data based on both anteroposterior perturbation trials (trial block 1) and mediolateral perturbation trials (trial block 2). Panel B shows data derived from anteroposterior perturbation trials (trial block 1). Panel C shows data derived from mediolateral perturbation trials (trial block 2). Panel D shows data derived from all backward fall trials in trial block 3, where participants were instructed to grasp the handrail as quickly as possible following the perturbation. Values shown are means with standard deviation error bars. The perturbation-based training program had a significantly larger training effect (preintervention-postintervention difference) than the control training program in the frequency of multi-step reactions (P⫽.034) and frequency of foot collisions (P⫽.0046) in surface translation trials and in handrail contact time in cable pull trials (P⫽.004).
only for cable pull trials. No other secondary step variable showed significant training effects. For grasping reactions (trial block 3), the control group showed a greater reduction in grasping errors than the perturbationbased training group (surface translation trials). Exploratory analysis of 486
f
Physical Therapy
Volume 90
physical functioning and psychological measurements failed to reveal any statistically significant training effects, other than a small improvement in “maximum balance range” in the perturbation-based training group; this is the maximum anteroposterior range of motion of the COM when the par-
Number 4
ticipant leans forward and backward at the ankles (eTab. 2, available at ptjournal.apta.org).
Discussion Perturbation-based balance training led to improvements in 3 of the 4 targeted balance reaction features. April 2010
Perturbation-Based Balance Training Program for Older Adults Specifically, the perturbation-based training group showed greater reductions than the control group in multi-step reactions and foot collisions in the surface translation trials and more rapid handrail contact in the cable pull trials. The degree to which each of the changes in the perturbation-based training group exceeded the change in the control group was comparable to, or larger than, the mean difference between fallers and nonfallers reported previously.17 These findings support the potential for perturbation-based balance training to reduce fall risk by improving the ability to respond to balance perturbations. The finding that handrail contact time was reduced in the cable pull trials provides clear evidence that some benefits of training with BOS perturbations generalized to COM perturbations. The generalizability of improvements in reactive balance control to a different perturbation type also is supported by a study showing that individuals who completed training on a moving platform were subsequently better able to recover from slips induced during overground walking.46 Surprisingly, our initial analyses failed to show an improvement in handrail contact time in the surface translation trials; however, such an effect did emerge when the confounding effect of grasp errors was removed. The perturbation-based training group was more likely than the control group to make such errors following training, probably because they were trained to grasp the rail quickly, without an explicit focus on accuracy. Future training programs should target both speed and accuracy. Although perturbation-based training led to substantive reduction in handrail contact time (eg, 120 milliseconds for cable pulls), the net (perturbation-based training group minus control group) reduction in April 2010
biceps muscle onset latency was much smaller (eg, 12 milliseconds for cable pulls) and was not statistically significant. Comparable effects of training on leg muscle latencies have been found in previous studies (17–18 milliseconds29,47). The present results indicate that the primary benefit of perturbation-based training was a reduction in movement time, rather than time required to detect instability and initiate the response. Although the ANOVA failed to provide statistical evidence to show a mean training benefit in reducing multi-step reactions in cable pull trials, it did appear that there was a substantial improvement for many participants. Ten of the 16 participants in the perturbation-based training group improved by 50% or more in the surface translation trials, and 5 of these 10 participants also improved by 50% or more in the cable pull trials. In contrast, no control group participants improved by 50% or more in both the surface translation and cable pull trials. The failure of multi-step and foot collision perturbation-based training effects to show greater evidence of transfer to COM perturbations may have been due, at least in part, to methodological limitations. Subsequent to completion of the present study, we discovered that cable pulls resulted in a slower and less destabilizing rise in perturbatory ankle torque compared with surface translations.36 This result led to improved baseline performance (ie, fewer multi-step reactions and limb collisions); thus, there was less potential for improvement during training. This floor effect in the cable pull trials was particularly pronounced for collisions, which occurred in ⬍6% of the trials at baseline versus 56% of the surface translation trials. The one training target that did not exhibit the hypothesized training effect was frequency of AP perturba-
tion trials in which additional lateral steps occurred following the initial AP-directed step. Such steps are thought to reflect an impaired ability, during the landing phase, to arrest lateral falling motion that arises after the swing foot is lifted17,38 (see part B of the video, available at ptjournal.apta.org). The cable pull trials offered little potential for improvement because such lateral steps occurred in ⬍5% of baseline trials, but it is not clear why improvements were not seen in the surface translation trials. Although perturbation-based training did not specifically target increases in step length, we found a modest increase for backward-fall perturbations (38% increase for surface translations, 23% increase for cable pulls). A previous perturbationbased training study22 demonstrated a more pronounced increase (⬃90%) in backward step length; however, that study involved participants with Parkinson disease, who tend to take shorter compensatory steps compared with age-matched controls.48,49 Although some effects of perturbationbased training on reactive balance control generalized to both COM and BOS perturbations, it appears that there were no benefits for other aspects of balance control. We found little evidence of training-related changes in measures related to control of unperturbed stance and gait (eg, postural sway, Timed “Up & Go” Test; eTab. 2, available at ptjournal.apta.org). Although this lack of improvement could be due to ceiling effects, it may reflect the specificity-of-training principle and the need for therapists to tailor balance training programs to target specific aspects of balance control. Further work is needed to determine the extent to which the training benefits that were observed reduce fall risk in daily life. The failure of previous efforts to demonstrate such
Volume 90
Number 4
Physical Therapy f
487
Perturbation-Based Balance Training Program for Older Adults an effect23,24 suggests that such studies will require a large sample size. Indirect evidence is provided by previous retrospective14 –16 and prospective17 studies demonstrating associations between the variables that were improved by training (multistep reactions, limb collisions, speed of arm reactions) and fall risk. Additional evidence from video surveillance studies50,51 shows examples of real-life falls that were preceded by step reactions involving limb collisions or multiple steps. Biomechanical considerations also provide reason to believe that the training benefits may help prevent falls. Limb collisions can jeopardize stability by preventing or delaying successful reestablishment of a stable base of support,37 and reduction in handrail contact time will help to ensure that individuals are able to grasp a stable object in sufficient time to prevent a fall. Although it is possible that multiple stepping is a preplanned strategy,39 participants were instructed to respond by taking as few steps as possible, so observed extra steps were likely a consequence of a poorly planned or executed initial step that was insufficient to halt the falling COM and recover equilibrium.52,53 Although multi-step reactions might be considered innocuous, provided a fall is prevented, each additional step is, in itself, destabilizing (particularly during the phase of single-limb support) and presents additional opportunity for a slip, trip, or limb collision to occur; thus, each new step may exacerbate fall risk.37 There are a number of possible mechanisms for the observed training effects. Previous work suggests that the main factor contributing to multi-stepping is an impaired ability to arrest COM motion during the landing phase of the initial step,52,53 rather than impaired initiation and execution of the stepping movement.38 Training may have improved the ability to recover equilibrium 488
f
Physical Therapy
Volume 90
during the landing phase by enhancing feedback control based on sensory information regarding foot contact, loading of the swing limb, and COM motion in relation to the newly established BOS.54 For participants with age-related declines in sensation, training may have resulted in a sensory “re-weighting” so as to rely more on information from less-impaired senses.55 Although age-related strength deficits also may contribute to impaired control of the step landing phase,56 isometric test results (eTab. 2, available at ptjournal.apta.org) suggest that the perturbation-based training was of insufficient intensity to increase strength. With regard to cognitive contributions to balance control,57,58 previous work has suggested that a rapid reallocation of cognitive resources occurs subsequent to perturbation onset to plan step trajectory.59 Thus, a more rapid and effective reallocation of cognitive resources, due to training, may have contributed to improved ability to avoid limb collisions. It also is likely that learning a different stepping strategy (lateral steps rather than crossovers) helped to reduce the likelihood of collisions.60 Evidence that perturbationbased training can enhance neural processing comes from a study showing improved voluntary reaction time21; however, we observed only small improvements in onset time for the perturbation-evoked reactions, likely because these reactions are triggered at a more automatic level.18,19 The observed training-related improvements in movement time during grasping reactions could have been due to strengthening of the specific neuromotor pathways responsible for planning and executing the reaching movement.61 Apparent training-related improvements in the control group for some balance measures were most likely either a placebo effect or an artifact
Number 4
related to the fact that the participants were more familiar with the balance testing protocol during the posttraining assessment. Within-session adaptation effects, which often are observed in studies of perturbation-evoked reactions,62,63 may have been retained in the posttraining test session. This possibility is supported by a previous study showing an improved ability to respond to a slip perturbation occurred during a session of repeated exposure to perturbations and was retained 4 months later.64 It is unlikely that the control group’s training program yielded physiological changes, as this stretching and relaxation program was of extremely low intensity (note the lack of improvement in sit-andreach flexibility; eTab. 2, available at ptjournal.apta.org). One potential limitation of this study pertains to the fact that multiple statistical analyses were performed, increasing the likelihood of a type I statistical error. Alpha can be adjusted to correct for the likelihood of a type I error (eg, Bonferroni correction); however, this adjustment is somewhat controversial, and may, in fact, have deleterious effects (eg, increased likelihood of a type II error).65 Within the 8 primary analyses, the training-related reduction in foot collisions and handrail contact time remains significant even when the Bonferonni correction is applied (reducing alpha from .05 to .006); however, the multi-step finding is no longer significant and should be conservatively viewed only as evidence of a potential trend. Further research is needed to determine the extent to which the present results apply to higher-risk populations. Although our cohort exhibited impaired control of change-in-support reactions compared with young adults36 and, therefore, had potential for improvement with training, they showed relatively high levels of physical function and mobility (eg, mean April 2010
Perturbation-Based Balance Training Program for Older Adults Timed “Up & Go” Test scores of ⬍7 seconds) as a consequence of our efforts to avoid confounding effects by excluding individuals with medical conditions affecting balance. Paradoxically, the few recruits who had lower levels of physical functioning withdrew from the study, most commonly because they disliked the large perturbations used during the first test session (4/7 withdrawals). The initial perturbation-based training sessions were much less daunting than the test sessions; thus, it is possible that such individuals would have been willing to complete perturbation-based training had they not first experienced the preintervention perturbation tests. In support of this possibility, our initial pilot tests, which did not include the balance testing protocol, showed that at-risk individuals who were referred to a falls clinic were willing and able to complete the perturbation-based training.8 Further work also is needed to determine how best to implement perturbation-based training programs in a clinical setting. The development of an inexpensive and compact commercial perturbation-delivery system would promote widespread clinical application. There also is potential to use existing equipment, such as programmable treadmills, to deliver the perturbations.23,24 Manually delivered perturbations are another possibility.22,29,66,67 Although it may be difficult to deliver highly unpredictable perturbations or to accurately control progression of perturbation magnitude when using manual methods,66 improvements in balance reactions have been observed with these methods.22,29 In summary, the present results indicate that perturbation-based balance training is an effective intervention to counter age-related impairments in balance recovery reactions. There have been few studies in this area to date, and the present randomized April 2010
controlled trial is the first to demonstrate improvements in stepping and grasping reactions in a nonclinical older adult population. Strengths of this study include: (1) a training program using principles of optimal motor learning to target specific features of reactions known to be associated with increased fall risk, (2) use of assessment methods that differed from those of the training program in order to assess generalizable training benefits, and (3) screening and randomization procedures that resulted in well-matched control and experimental groups that were free of confounding medical conditions or medication use. Further work is needed to examine the effects in clinical populations, to determine whether a maintenance program is required to retain the training benefits, to explore less expensive methods for delivering the training perturbations, and to assess whether training benefits reduce fall risk in daily life. All authors provided concept/idea/research design and writing. Dr Mansfield and Ms Peters provided data collection and analysis. Dr Liu and Dr Maki provided fund procurement and facilities/equipment. Dr Liu provided project management and consultation (including review of manuscript before submission). The authors thank Aaron Marquis, Tracy Lee, and Areeba Adnan for their assistance with data collection and processing. They also thank Stephen Lord and the Prince of Wales Medical Research Institute, Sydney, Australia, for providing the FallScreen equipment and falls-risk assessment software. Ethics approval was obtained from the Research Ethics Board of Sunnybrook Health Sciences Centre. Preliminary data from this study were presented at the Festival of International Conferences on Caregiving, Disability, Aging and Technology; June 2007; Toronto, Ontario, Canada, and the International Society for Postural and Gait Research; July 2007; Burlington, Vermont. This study was supported by the Canadian Institutes of Health Research (grant NET54025), the Ontario Neurotrauma Foundation (summer internship grant ONF2007-
PREV-INT-452), and the Canadian Foundation for Innovation. The views contained in this publication are those of the grantees and do not necessarily reflect those of the funding agencies. This trial was registered with ClinicalTrials. gov prior to recruiting the first participant (identifier: NCT00187317). This article was received February 26, 2009, and was accepted October 30, 2009. DOI: 10.2522/ptj.20090070
References 1 Kenny RA, Rubenstein LZ, Martin FC, Tinetti ME; for the American Geriatrics Society, British Geriatrics Society, and American Academy of Orthopaedic Surgeons Panel on Falls Prevention. Guideline for the prevention of falls in older persons. J Am Geriatr Soc. 2001;49:664 – 672. 2 Gillespie LD, Gillespie WJ, Robertson MC, et al. Interventions for preventing falls in elderly people. Cochrane Database Syst Rev. 2001;3:CD000340. 3 Province MA, Hadley EC, Hornbrook MC, et al. The effects of exercise on falls in elderly patients: a preplanned metaanalysis of the FICSIT trials. JAMA. 1995; 273:1341–1347. 4 Rubenstein LZ, Josephson KR, Trueblood PR, et al. Effects of a group exercise program on strength, mobility, and falls among fall-prone elderly men. J Gerontol A Biol Sci Med Sci. 2000;55:M317–M321. 5 Wolf SL, Sattin RW, Kutner M, et al. Intense Tai Chi exercise training and fall occurrences in older, transitionally frail adults: a randomized, controlled trial. J Am Geriatr Soc. 2003;51:1693–1701. 6 Lord SR, Castell S, Corcoran J, et al. The effect of group exercise on physical functioning and falls in frail older people living in retirement villages: a randomized, controlled trial. J Am Geriatr Soc. 2003;51: 1685–1692. 7 Mansfield A, Peters AL, Liu BA, Maki BE. A perturbation-based balance training program for older adults: study protocol for a randomised controlled trial. BMC Geriatr. 2007;7:12. 8 Maki BE, Cheng KC-C, Mansfield A, et al. Preventing falls in older adults: new interventions to promote more effective change-in-support balance reactions. J Electromyogr Kinesiol. 2008;18:243–254. 9 Maki BE, McIlroy WE. Postural control in the older adult. Clin Geriatr Med. 1996; 12:635– 658. 10 Maki BE, McIlroy WE. Control of rapid limb movements for balance recovery: age-related changes and implications for fall prevention. Age Ageing. 2006;35 (suppl 2):ii12–ii18. 11 Shumway-Cook A, Woollacott MH. Motor Control: Theory and Practical Applications. Baltimore, MD: Williams & Wilkins; 1995.
Volume 90
Number 4
Physical Therapy f
489
Perturbation-Based Balance Training Program for Older Adults 12 McIlroy WE, Maki BE. Task constraints on foot movement and the incidence of compensatory stepping following perturbation of upright stance. Brain Res. 1993;616: 30 –38. 13 Jensen JJ, Brown LA, Woollacott MH. Compensatory stepping: the biomechanics of a preferred response among older adults. Exp Aging Res. 2001;27:361–376. 14 Wolfson LI, Whipple R, Amerman P, Kleinberg A. Stressing the postural response: a quantitative method for testing balance. J Am Geriatr Soc. 1986;34:845– 850. 15 Chandler JM, Duncan PW, Studenski SA. Balance performance on the postural stress test: comparison of young adults, healthy elderly, and fallers. Phys Ther. 1990;70: 410 – 415. 16 Rogers MW, Hedman LD, Johnson ME, et al. Lateral stability during forwardinduced stepping for dynamic balance recovery in young and older adults. J Gerontol A Biol Sci Med Sci. 2001;56:M589 – M594. 17 Maki BE, Edmondstone MA, Perry SD, et al. Control of rapid limb movements for balance recovery: do age-related changes predict falling risk? In: Duysens J, SmitsEngelsman BCM, Kingma H, eds. Control of Posture and Gait. Maastricht, the Netherlands: International Society for Posture and Gait Research; 2001:126 –129. 18 Maki BE, McIlroy WE. The role of limb movements in maintaining upright stance: the “change-in-support” strategy. Phys Ther. 1997;77:488 –507. 19 Luchies CW, Wallace D, Pazdur R, et al. Effects of age on balance assessment using voluntary and involuntary step tasks. J Gerontol A Biol Sci Med Sci. 1999;54: M140 –M144. 20 Maki BE, McIlroy WE. Change-in-support balance reactions in older persons: an emerging research area of clinical importance. Neurol Clin. 2005;23:751–783. 21 Rogers MW, Johnson ME, Martinez KM, et al. Step training improves the speed of voluntary step initiation in aging. J Gerontol A Biol Sci Med Sci. 2003;58:46 –51. 22 Jo ¨ bges M, Heuschkel G, Pretzel C, et al. Repetitive training of compensatory steps: a therapeutic approach for postural instability in Parkinson’s disease. J Neurol Neurosurg Psychiatry. 2004;75:1682–1687. 23 Shimada H, Obuchi S, Furuna T, Suzuki T. New intervention program for preventing falls among frail elderly people: the effects of perturbed walking exercise using a bilateral separated treadmill. Am J Phys Med Rehabil. 2004;83:493– 499. 24 Protas EJ, Mitchell K, Williams A, et al. Gait and step training to reduce falls in Parkinson’s disease. Neurorehabilitation. 2005; 20:183–190. 25 Thomas S, Reading J, Shephard RI. Revision of the Physical Activity Readiness Questionnaire (PAR-Q). Can J Sport Sci. 1992;174:338 –345. 26 Kernan WN, Viscoli CM, Makuch RW, et al. Stratified randomization for clinical trials. J Clin Epidemiol. 1999;52:19 –26.
490
f
Physical Therapy
Volume 90
27 Weerdesteyn V, Rijken H, Guerts ACH, et al. A five-week exercise program can reduce falls and improve obstacle avoidance in the elderly. Gerontology. 2006;52: 131–141. 28 Sihvonen S, Sipila¨ S, Taskinen S, Pertti E. Fall incidence in frail older women after individualized visual feedback-based balance training. Gerontology. 2004;50: 411– 416. 29 Marigold DS, Eng JJ, Dawson AS, et al. Exercise leads to faster postural reflexes, improved balance and mobility, and fewer falls in older persons with chronic stroke. J Am Geriatr Soc. 2005;53:416 – 423. 30 Rose DJ, Clark S. Can the control of bodily orientation be significantly improved in a group of older adults with a history of falls? J Am Geriatr Soc. 2000;48:275–282. 31 Vickers JN, Livingston LF, Umeris-Bohnert S, Holden D. Decision training: the effects of complex instruction, variable practice and reduced delayed feedback on the acquisition and transfer of a motor skill. J Sports Sci. 1999;17:357–367. 32 Vickers JN, Reeves M-A, Chambers KL, Martell S. Decision training: cognitive strategies for enhancing motor performance. In: Williams AM, Hodges NJ, eds. Skill Acquisition in Sport: Research, Theory and Practice. London, United Kingdom: Routledge, Taylor & Francis Group; 2004:103–120. 33 Payne RA. Relaxation Techniques: A Practical Handbook for the Health Care Professional. 2nd ed. Edinburgh, Scotland: Churchill Livingstone; 2000. 34 Maki BE, McIlroy WE, Perry SD. Influence of lateral destabilization on compensatory stepping responses. J Biomech. 1996;29: 343–353. 35 Maki BE, McIlroy WE, Fernie GR. Changein-support reactions for balance recovery: control mechanisms, age-related changes and implications for fall prevention. IEEE Eng Med Biol Mag. 2003;22:20 –26. 36 Mansfield A, Maki BE. Are age-related impairments in change-in-support balance reactions dependent on the nature of the balance perturbation? J Biomech. 2009;42: 1023–1031. 37 Maki BE, Edmondstone MA, McIlroy WE. Age-related differences in laterally directed compensatory stepping behavior. J Gerontol A Biol Sci Med Sci. 2000;55: M270 –M277. 38 McIlroy WE, Maki BE. Age-related changes in compensatory stepping in response to unpredictable perturbations. J Gerontol A Biol Sci Med Sci. 1996;51:M289 –M296. 39 Luchies CW, Alexander NB, Schultz AB, Ashton-Miller J. Stepping responses of young and old adults to postural disturbances: kinematics. J Am Geriatr Soc. 1994;42:506 –512. 40 Maki BE. A Posture Control Model and Balance Test for the Prediction of Relative Postural Stability With Special Consideration to the Problem of Falling in the Elderly [PhD dissertation]. Glasgow, Scotland: Bioengineering Unit, University of Strathclyde; 1987.
Number 4
41 Winter DA. Biomechanics and Motor Control of Human Movement. 2nd ed. Toronto, Ontario, Canada: John Wiley & Sons Canada Ltd; 1990. 42 McIlroy WE, Maki BE. Changes in early “automatic” postural responses associated with the prior-planning and execution of a compensatory step. Brain Res. 1993;631: 203–211. 43 Chow S-C, Shao J, Wang H. Sample Size Calculations in Clinical Research. New York, NY: Marcel Dekker Inc; 2003. 44 Conover WJ, Iman RL. Rank transformations as a bridge between parametric and nonparametric statistics. Am Stat. 1981; 35:124 –129. 45 Lord SR, Clark RD. Simple physiological and clinical tests for the accurate prediction of falling in older people. Gerontology. 1996;42:199 –203. 46 Bhatt T, Pai Y-C. Generalization of gait adaptation for fall prevention: from moveable platform to slippery floor. J Neurophysiol. 2009;101:948 –957. 47 Gatts SK, Woollacott MH. Neural mechanisms underlying balance improvement with short term Tai Chi training. Aging Clin Exp Res. 2006;18:7–19. 48 Burleigh-Jacobs A, Horak FB, Nutt JG, Obeso JA. Step initiation in Parkinson’s disease: influence of levodopa and external sensory triggers. Mov Disord. 1997;12: 206 –215. 49 Damiano NK, McIlroy WE, Maki BE, Verrier MC. Compensatory stepping in Parkinson’s disease: do PD patients use external cues to decrease postural instability? Soc Neurosci Abstr. 2000;26:164. 50 Holliday PJ, Fernie GR, Gryfe CI, Griggs GT. Video recording of spontaneous falls of the elderly. In: Gray BE, ed. Slips, Stumbles, and Falls: Pedestrian Footwear and Surfaces. Philadelphia, PA: American Society for Testing and Materials; 1990:7–16. [Video CD in .mpg format available at: http://sunnybrook.ca/research/?page⫽sri_ proj_csia_focus_afep.] 51 Robinovitch S, Feldman F, Wan D, et al. Video recording of real-life falls in long term care provides new insight on the cause and circumstances of falls in older adults. Paper presented at: International Society for Posture and Gait Research; 21–25 June, 2009; Bologna, Italy. 52 Maki BE, McIlroy WE. The control of foot placement during compensatory stepping reactions: does speed of response take precedence over stability? IEEE Trans Rehabil Eng. 1999;7:80 –90. 53 Pai Y-C, Rogers MW, Patton J, et al. Static versus dynamic predictions of protective stepping following waist-pull perturbations in young and older adults. J Biomech. 1998;31:1111–1118. 54 Maki BE, Perry SD, Norrie RG, McIlroy WE. Effect of facilitation of sensation from plantar foot-surface boundaries on postural stabilization in young and older adults. J Gerontol A Biol Sci Med Sci. 1999; 54:M281–M287.
April 2010
Perturbation-Based Balance Training Program for Older Adults 55 Woollacott MH, Shumway-Cook A, Nashner LM. Aging and posture control: changes in sensory organization and muscular coordination. Int J Aging Hum Dev. 1986;23:81–97. 56 Maki BE, McIlroy WE. Control of compensatory stepping reactions: age-related impairment and the potential for remedial intervention. Physiother Theory Pract. 1999;15:69 –90. 57 Maki BE, McIlroy WE. Cognitive demands and corticol control of human balancerecovery reactions. J Neural Transm. 2007; 114:1279 –1296. 58 Woollacott MH, Shumway-Cook A. Attention and the control of posture and gait: a review of an emerging area of research. Gait Posture. 2002;16:1–14. 59 Zettel JL, Holbeche A, McIlroy WE, Maki BE. Redirection of gaze and switching of attention during rapid stepping reactions evoked by unpredictable postural perturbation. Exp Brain Res. 2005;165:392– 401. 60 Mille M-L, Johnson ME, Martinez KM, Rogers MW. Age-dependent differences in lateral balance recovery through protective stepping. Clin Biomech. 2005;20:607– 616. 61 Sveistrup H, Woollacott MH. Practice modifies the developing automatic postural response. Exp Brain Res. 1997;114:33– 43. 62 McIlroy WE, Maki BE. Adaptive changes to compensatory stepping responses. Gait Posture. 1995;3:43–50.
Invited Commentary The field of motor learning has affected the practice of physical therapy in many ways, some of which are reflected in the study by Mansfield et al.1 This study used an intervention based on the principles of optimal motor learning and tested the effects of training on automatic balance reactions. Specifically, the authors targeted well-defined aspects of balance reactions and tested whether perturbation training would reduce: (1) multi-step reactions and frequency of lateral steps during forward perturbations, (2) frequency of foot collisions during lateral perturbations, and (3) time required to grasp a handrail during a backward perturbation. They chose these outcome measures because previous studies demonstrated impairments in older people with a history of falls in these specific characteristics of bal-
April 2010
63 Horak FB. Adaptation of automatic postural responses. In: Bloedel J, Ebner TJ, Wise SP, eds. The Acquisition of Motor Behavior in Vertebrates. Cambridge, MA: MIT Press; 1996. 64 Bhatt T, Pai Y-C. Prevention of slip-related backward balance loss: the effect of session intensity and frequency on long-term retention. Arch Phys Med Rehabil. 2009; 90:34 – 42. 65 Perneger T. What’s wrong with Bonferroni adjustments. Br Med J. 1998;316: 1236 –1238. 66 Rose DJ. Fallproof!: A Comprehensive Balance and Mobility Training Program. Champaign, IL: Human Kinetics; 2003. 67 Horak FB, Wrisley DM, Frank J. The Balance Evaluation Systems Test (BESTest) to differentiate balance deficits. Phys Ther. 2009;89:484 – 498. 68 Podsiadlo D, Richardson S. The Timed “Up & Go”: A test of basic functional mobility for frail elderly persons. J Am Geriatr Soc. 1991;39:142–148. 69 Powell LE, Myers AM. The Activitiesspecific Balance Confidence (ABC) Scale. J Gerontol A Biol Sci Med Sci. 1995;50: M28 –M34. 70 Lord SR, Menz HB, Tiedemann A. A physiological profile approach to falls risk assessment and prevention. Phys Ther. 2003; 83:237–252.
71 McIlroy WE, Maki BE. The “deceleration response” to transient perturbation of upright stance. Neurosci Lett. 1994;175:13–16. 72 McIlroy WE, Maki BE. Preferred placement of the feet during quiet stance: development of a standardized foot placement for balance testing. Clin Biomech. 1997;12: 66 –70. 73 Hoeger WWK, Hopkins DR. A comparison of the sit and reach and the modified sit and reach in the measurement of flexibility in women. Res Q Exerc Sport. 1992;63: 191–195. 74 Holbein-Jenny MA, McDermott K, et al. Validity of functional stability limits as a measure of balance in adults aged 23–73 years. Ergonomics. 2007;50:631– 646. 75 Brauer SG, Burns TR, Galley P. A prospective study of laboratory and clinical measures of postural stability to predict community-dwelling fallers. J Gerontol A Biol Sci Med Sci. 2000;55:M469 –M476. 76 Row BS. Weight-Bearing Speed of Movement in Older Adults: The Effects of HighVelocity Resistance Training [PhD dissertation]. University Park, PA: Pennsylvania State University; 2003. 77 Endler NS, Edwards JM, Vitelli R, Parker JDA. Assessment of state and trait anxiety: Endler Multidimensional Anxiety Scales. Anxiety Research. 1989;2:1–14.
Fay B. Horak, Laurie A. King
ance reactions. The authors asked a very practical question, and the results have direct implications for balance training in physical therapy. One important, practical finding is that balance reactions can be trained and improved. Not long ago, balance responses were considered hardwired reflexes that could not be influenced by practice, training, and feedback. Considering stepping responses for postural recovery as complex motor skills allows therapists the opportunity to develop exercise interventions targeted to each patient’s specific impairments of his or her balance responses. What is still unknown is how that improvement carries over to real-life situations. The authors did not find significant carryover when partici-
pants were tested on a different type of perturbation. This finding is important because real-life encounters will be in varying directions, magnitudes, and types of destabilizing forces. It may be that carryover did exist but was not detected by the chosen outcome measures or that a ceiling effect masked any potential changes. The authors point out in the discussion that the clinical, pulley perturbation (which was used only in the pretesting and posttesting periods) was much less challenging than the platform perturbations used during training. Therefore, the participants may not have shown significant deficits in the pretest session that could be improved by the posttest session using the pulley. Another clinically relevant finding was that the benefits of perturbation
Volume 90
Number 4
Physical Therapy f
491
Perturbation-Based Balance Training Program for Older Adults 55 Woollacott MH, Shumway-Cook A, Nashner LM. Aging and posture control: changes in sensory organization and muscular coordination. Int J Aging Hum Dev. 1986;23:81–97. 56 Maki BE, McIlroy WE. Control of compensatory stepping reactions: age-related impairment and the potential for remedial intervention. Physiother Theory Pract. 1999;15:69 –90. 57 Maki BE, McIlroy WE. Cognitive demands and corticol control of human balancerecovery reactions. J Neural Transm. 2007; 114:1279 –1296. 58 Woollacott MH, Shumway-Cook A. Attention and the control of posture and gait: a review of an emerging area of research. Gait Posture. 2002;16:1–14. 59 Zettel JL, Holbeche A, McIlroy WE, Maki BE. Redirection of gaze and switching of attention during rapid stepping reactions evoked by unpredictable postural perturbation. Exp Brain Res. 2005;165:392– 401. 60 Mille M-L, Johnson ME, Martinez KM, Rogers MW. Age-dependent differences in lateral balance recovery through protective stepping. Clin Biomech. 2005;20:607– 616. 61 Sveistrup H, Woollacott MH. Practice modifies the developing automatic postural response. Exp Brain Res. 1997;114:33– 43. 62 McIlroy WE, Maki BE. Adaptive changes to compensatory stepping responses. Gait Posture. 1995;3:43–50.
Invited Commentary The field of motor learning has affected the practice of physical therapy in many ways, some of which are reflected in the study by Mansfield et al.1 This study used an intervention based on the principles of optimal motor learning and tested the effects of training on automatic balance reactions. Specifically, the authors targeted well-defined aspects of balance reactions and tested whether perturbation training would reduce: (1) multi-step reactions and frequency of lateral steps during forward perturbations, (2) frequency of foot collisions during lateral perturbations, and (3) time required to grasp a handrail during a backward perturbation. They chose these outcome measures because previous studies demonstrated impairments in older people with a history of falls in these specific characteristics of bal-
April 2010
63 Horak FB. Adaptation of automatic postural responses. In: Bloedel J, Ebner TJ, Wise SP, eds. The Acquisition of Motor Behavior in Vertebrates. Cambridge, MA: MIT Press; 1996. 64 Bhatt T, Pai Y-C. Prevention of slip-related backward balance loss: the effect of session intensity and frequency on long-term retention. Arch Phys Med Rehabil. 2009; 90:34 – 42. 65 Perneger T. What’s wrong with Bonferroni adjustments. Br Med J. 1998;316: 1236 –1238. 66 Rose DJ. Fallproof!: A Comprehensive Balance and Mobility Training Program. Champaign, IL: Human Kinetics; 2003. 67 Horak FB, Wrisley DM, Frank J. The Balance Evaluation Systems Test (BESTest) to differentiate balance deficits. Phys Ther. 2009;89:484 – 498. 68 Podsiadlo D, Richardson S. The Timed “Up & Go”: A test of basic functional mobility for frail elderly persons. J Am Geriatr Soc. 1991;39:142–148. 69 Powell LE, Myers AM. The Activitiesspecific Balance Confidence (ABC) Scale. J Gerontol A Biol Sci Med Sci. 1995;50: M28 –M34. 70 Lord SR, Menz HB, Tiedemann A. A physiological profile approach to falls risk assessment and prevention. Phys Ther. 2003; 83:237–252.
71 McIlroy WE, Maki BE. The “deceleration response” to transient perturbation of upright stance. Neurosci Lett. 1994;175:13–16. 72 McIlroy WE, Maki BE. Preferred placement of the feet during quiet stance: development of a standardized foot placement for balance testing. Clin Biomech. 1997;12: 66 –70. 73 Hoeger WWK, Hopkins DR. A comparison of the sit and reach and the modified sit and reach in the measurement of flexibility in women. Res Q Exerc Sport. 1992;63: 191–195. 74 Holbein-Jenny MA, McDermott K, et al. Validity of functional stability limits as a measure of balance in adults aged 23–73 years. Ergonomics. 2007;50:631– 646. 75 Brauer SG, Burns TR, Galley P. A prospective study of laboratory and clinical measures of postural stability to predict community-dwelling fallers. J Gerontol A Biol Sci Med Sci. 2000;55:M469 –M476. 76 Row BS. Weight-Bearing Speed of Movement in Older Adults: The Effects of HighVelocity Resistance Training [PhD dissertation]. University Park, PA: Pennsylvania State University; 2003. 77 Endler NS, Edwards JM, Vitelli R, Parker JDA. Assessment of state and trait anxiety: Endler Multidimensional Anxiety Scales. Anxiety Research. 1989;2:1–14.
Fay B. Horak, Laurie A. King
ance reactions. The authors asked a very practical question, and the results have direct implications for balance training in physical therapy. One important, practical finding is that balance reactions can be trained and improved. Not long ago, balance responses were considered hardwired reflexes that could not be influenced by practice, training, and feedback. Considering stepping responses for postural recovery as complex motor skills allows therapists the opportunity to develop exercise interventions targeted to each patient’s specific impairments of his or her balance responses. What is still unknown is how that improvement carries over to real-life situations. The authors did not find significant carryover when partici-
pants were tested on a different type of perturbation. This finding is important because real-life encounters will be in varying directions, magnitudes, and types of destabilizing forces. It may be that carryover did exist but was not detected by the chosen outcome measures or that a ceiling effect masked any potential changes. The authors point out in the discussion that the clinical, pulley perturbation (which was used only in the pretesting and posttesting periods) was much less challenging than the platform perturbations used during training. Therefore, the participants may not have shown significant deficits in the pretest session that could be improved by the posttest session using the pulley. Another clinically relevant finding was that the benefits of perturbation
Volume 90
Number 4
Physical Therapy f
491
Perturbation-Based Balance Training Program for Older Adults training did not generalize to other aspects of balance control such as postural sway and the Timed “Up & Go” Test. However, the pretraining Activities-specific Balance Confidence Scale and Timed “Up & Go” Test scores also were very high, indicating insensitive clinical tests to the type of balance impairment they were targeting. Balance control consists of many different underlying systems, so it is not expected for improvement in one system (stepping responses) to benefit other systems (sway in quiet stance or anticipatory postural adjustments associated with the sit-to-stand task, turning, and walking in the Timed “Up & Go” Test).
Author Response
f
Physical Therapy
Volume 90
F.B. Horak, PT, PhD, is Research Professor of Neurology and Adjunct Professor of Physiology and Biomedical Engineering, Department of Neurology, Oregon Health and Sciences University, West Campus, Building 1, 505 NW 185th Ave, Beaverton, OR 970063499 (USA). Address all correspondence to Dr Horak at:
[email protected]. L.A. King, PT, PhD, is Post-doctoral Fellow, Department of Neurology, Oregon Health and Sciences University. DOI: 10.2522/ptj.20090070.ic
Reference 1 Mansfield A, Peters AL, Liu BA, Maki BE. Effect of a perturbation-based balance training program on compensatory stepping and grasping reactions in older adults: a randomized controlled trial. Phys Ther. 2010; 90:476 – 491.
Avril Mansfield, Amy L. Peters, Barbara A. Liu, Brian E. Maki
We thank Horak and King for their comments1 on our article.2 We are in complete agreement, but would like to clarify one detail. Our results did, in fact, show evidence of one carryover effect: an improvement in the reactions to center-of-mass perturbations (cable pulls) that resulted from training with base-of-support perturbations (surface translations). This carryover effect was observed in the measure that was the target of the
492
In summary, this is a very well executed study with important clinical implications. Stepping and reaching reactions to recover equilibrium are modifiable and can improve with training. However, physical therapists need to identify specific impairments of balance control before designing specific treatments for balance deficits rather than identifying a general “balance problem” and using general exercises for “balance.” More research is needed to identify the best training protocol and intensity of exercise to allow for maximal benefit and carryover of training to daily life.
grasp training (ie, reduction in the time required to contact the handrail).
2 Mansfield A, Peters AL, Liu BA, Maki BE. Effect of a perturbation-based balance training program on compensatory stepping and grasping reactions in older adults: a randomized controlled trial. Phys Ther. 2010; 90:476 – 491.
DOI: 10.2522/ptj.20090070.ar
References 1 Horak FB, King LA. Invited commentary on “Effect of a perturbation-based balance training program on compensatory stepping and grasping reactions in older adults: a randomized controlled trial.” Phys Ther. 2010;90:491– 492.
Number 4
April 2010
Perturbation-Based Balance Training Program for Older Adults training did not generalize to other aspects of balance control such as postural sway and the Timed “Up & Go” Test. However, the pretraining Activities-specific Balance Confidence Scale and Timed “Up & Go” Test scores also were very high, indicating insensitive clinical tests to the type of balance impairment they were targeting. Balance control consists of many different underlying systems, so it is not expected for improvement in one system (stepping responses) to benefit other systems (sway in quiet stance or anticipatory postural adjustments associated with the sit-to-stand task, turning, and walking in the Timed “Up & Go” Test).
Author Response
f
Physical Therapy
Volume 90
F.B. Horak, PT, PhD, is Research Professor of Neurology and Adjunct Professor of Physiology and Biomedical Engineering, Department of Neurology, Oregon Health and Sciences University, West Campus, Building 1, 505 NW 185th Ave, Beaverton, OR 970063499 (USA). Address all correspondence to Dr Horak at:
[email protected]. L.A. King, PT, PhD, is Post-doctoral Fellow, Department of Neurology, Oregon Health and Sciences University. DOI: 10.2522/ptj.20090070.ic
Reference 1 Mansfield A, Peters AL, Liu BA, Maki BE. Effect of a perturbation-based balance training program on compensatory stepping and grasping reactions in older adults: a randomized controlled trial. Phys Ther. 2010; 90:476 – 491.
Avril Mansfield, Amy L. Peters, Barbara A. Liu, Brian E. Maki
We thank Horak and King for their comments1 on our article.2 We are in complete agreement, but would like to clarify one detail. Our results did, in fact, show evidence of one carryover effect: an improvement in the reactions to center-of-mass perturbations (cable pulls) that resulted from training with base-of-support perturbations (surface translations). This carryover effect was observed in the measure that was the target of the
492
In summary, this is a very well executed study with important clinical implications. Stepping and reaching reactions to recover equilibrium are modifiable and can improve with training. However, physical therapists need to identify specific impairments of balance control before designing specific treatments for balance deficits rather than identifying a general “balance problem” and using general exercises for “balance.” More research is needed to identify the best training protocol and intensity of exercise to allow for maximal benefit and carryover of training to daily life.
grasp training (ie, reduction in the time required to contact the handrail).
2 Mansfield A, Peters AL, Liu BA, Maki BE. Effect of a perturbation-based balance training program on compensatory stepping and grasping reactions in older adults: a randomized controlled trial. Phys Ther. 2010; 90:476 – 491.
DOI: 10.2522/ptj.20090070.ar
References 1 Horak FB, King LA. Invited commentary on “Effect of a perturbation-based balance training program on compensatory stepping and grasping reactions in older adults: a randomized controlled trial.” Phys Ther. 2010;90:491– 492.
Number 4
April 2010
Research Report Quality-of-Life Change Associated With Robotic-Assisted Therapy to Improve Hand Motor Function in Patients With Subacute Stroke: A Randomized Clinical Trial Nancy G. Kutner, Rebecca Zhang, Andrew J. Butler, Steven L. Wolf, Jay L. Alberts
Background. At 6 months poststroke, most patients cannot incorporate their affected hand into daily activities, which in turn is likely to reduce their perceived quality of life.
Objective. This preliminary study explored change in patient-reported, healthrelated quality of life associated with robotic-assisted therapy combined with reduced therapist-supervised training.
Design and Setting. A single-blind, multi-site, randomized clinical trial was conducted.
Participants. Seventeen individuals who were 3 to 9 months poststroke participated. Intervention. Sixty hours of therapist-supervised repetitive task practice (RTP) was compared with 30 hours of RTP combined with 30 hours of robotic-assisted therapy.
Measurements. Participants completed the Stroke Impact Scale (SIS) at baseline, immediately postintervention, and 2 months postintervention. Change in SIS score domains was assessed in a mixed model analysis.
Results. The combined therapy group had a greater increase in rating of mood from preintervention to postintervention, and the RTP-only group had a greater increase in rating of social participation from preintervention to follow-up. Both groups had statistically significant improvement in activities of daily living and instrumental activities of daily living scores from preintervention to postintervention. Both groups reported significant improvement in hand function postintervention and at follow-up, and the magnitude of these changes suggested clinical significance. The combined therapy group had significant improvements in stroke recovery rating postintervention and at follow-up, which appeared clinically significant; this also was true for stroke recovery rating from preintervention to follow-up in the RTP-only group.
N.G. Kutner, PhD, is Professor of Rehabilitation Medicine, Emory University School of Medicine, 1441 Clifton Rd NE, Atlanta, GA 30322 (USA). Address all correspondence to Dr Kutner at:
[email protected]. R. Zhang, MS, is Senior Associate, Biostatistics and Bioinformatics, Rollins School of Public Health, Atlanta, Georgia. A.J. Butler, PT, PhD, FAHA, is Associate Professor of Rehabilitation Medicine, Emory University School of Medicine. S.L. Wolf, PT, PhD, FAPTA, FAHA, is Professor of Rehabilitation Medicine and Cell Biology, Emory University School of Medicine. J.L. Alberts, PhD, is Associate Professor of Biomedical Engineering, Cleveland Clinic Foundation, and Cleveland FES Center, L. Stokes Cleveland VA Medical Center, Cleveland, Ohio. [Kutner NG, Zhang R, Butler AJ, et al. Quality-of-life change associated with robotic-assisted therapy to improve hand motor function in patients with subacute stroke: a randomized clinical trial. Phys Ther. 2010;90:493–504.] © 2010 American Physical Therapy Association
Limitations. Outcomes of 30 hours of RTP in the absence of robotic-assisted therapy remain unknown.
Conclusion. Robotic-assisted therapy may be an effective alternative or adjunct to the delivery of intensive task practice interventions to enhance hand function recovery in patients with stroke. Post a Rapid Response or find The Bottom Line: www.ptjournal.org April 2010
Volume 90
Number 4
Physical Therapy f
493
Health-Related Quality of Life and Robotic-Assisted Therapy for Hand Function
S
troke is the most prevalent cause of adult-onset disability in the United States. An estimated 5.8 million people who have had a stroke have residual neurological deficits.1,2 At 6 months poststroke, about 65% of patients cannot incorporate their affected hand into their usual activities,1 a limitation of distal motor function that understandably is associated with reduced perception of quality of life.3 There has been limited research on quality of life and rehabilitation therapies among people with conditions such as stroke that limit mobility. Repetitive task practice (RTP) strategies, developed for use in constraintinduced movement therapy (CIMT), can improve hand function and individuals’ perception of their healthrelated quality of life (HRQOL).4 Support for the use of RTP as a fundamental basis for retraining upperextremity function has been addressed in 2 excellent reviews.5,6 Briefly, RTP consists of breaking a task down into specific segments. These segments then are practiced individually and require successful completion before the entire task is “put together.” The activities the patient performs and the feedback provided are consistent with models of a massed RTP practice schedule.7 Repetitive task practice– based interventions, however, are labor-intensive therapies. Robotic devices continue to improve in design, control, and usability and may offer a solution to decrease the therapist time demands necessary for the delivery of RTP-
Available With This Article at ptjournal.apta.org • Audio Abstracts Podcast This article was published ahead of print on February 25, 2010, at ptjournal.apta.org.
494
f
Physical Therapy
Volume 90
type interventions. The ability of several device-oriented approaches to facilitate improvement in movement capability among patients following stroke has been explored.8 –11 In this regard, sophisticated upperextremity robotic systems (eg, MITManus*) have been developed.8,12,13 The vast majority of upper-extremity robotic systems (eg, MIT-Manus/ InMotion 2.0,* GENTILE/s,† MIME,‡ and others) are focused on improving gross reaching movements of the upper extremity.14 –18 These devices have shown effectiveness in improving proximal upper-extremity function and motor control10 and have yielded useful data for understanding recovery mechanisms.8,9 A systematic review of the rehabilitation literature indicates strong evidence that intensity and task specificity are primary indicators of an effective treatment program following stroke.19,20 In addition, training should be repetitive, functional, meaningful, and challenging to the patient; these are characteristics of task practice interventions that have been shown to improve upperextremity function in patients with stroke.4,21,22 Despite their effectiveness, widespread implementation of RTP interventions in their current forms faces substantial obstacles (primarily cost of delivering the service and the duration of each session) that limit the potential for clinical acceptance. The use of a robotic device as a therapeutic adjunct to task practice is appealing because this approach may enhance the recovery process and potentially de-
* Interactive Motion Technologies Inc, 37 Spinelli Place, Cambridge, MA 02318. † This system is not commercially available. It has been and is currently being used in research protocols at the University of Reading in the United Kingdom. ‡ This system is not commercially available; developed at the VA-Rehabilitation Research and Development Center at Palo Alto, California.
Number 4
crease the time spent in therapistdirected task practice. The robotic device used in this project focuses on improving distal motor function by enhancing active range of motion (AROM) of the wrist and fingers and reducing spasticity (exaggerated reflex response to slow or fast stretch of wrist or finger flexor muscles) about the wrist. The device is based on principles of motor learning, as it engages and provides feedback to the individual during the performance of repetitive activities that are intended to transfer to functional distal motor activities. The robotic device provides consistent and precise therapy for long durations without fatigue. Previous studies have shown that the quantity and quality of afferent information are greater during consistent and larger-amplitude movements.23–25 The resultant increase in the quantity and quality of afferent information provided to the patient during robotic device use may facilitate motor learning or relearning. Augmenting the motor learning or relearning processes with a robotic device may allow for a reduction in the duration of therapist-directed RTP without compromising improvements in motor function or HRQOL in patients with stroke. A meta-analysis of robot-assisted therapy effects on upper-limb function in patients with stroke concluded that robotic therapy typically improves proximal limb control.18,26 However, there is limited clinical acceptance of robot-assisted therapy.9 Although the motor effects of robotic-assisted therapy are beginning to be documented, there is little evidence to date about the effects on patientassessed HRQOL when roboticassisted therapy is used in conjunction with an RTP intervention. This article reports HRQOL outcomes that were gathered as part of April 2010
Health-Related Quality of Life and Robotic-Assisted Therapy for Hand Function a preliminary randomized clinical trial. The primary aim of the clinical trial was to determine the clinical effectiveness of using a distally focused robotic device in conjunction with an abbreviated duration of RTP to improve upper-extremity motor function in 2 groups of patients with stroke: (1) an RTP-only group whose participants received 60 hours of therapist-supervised RTP and (2) a combined therapy group whose participants received 30 hours of robotic-assisted therapy and 30 hours of therapist-supervised RTP. The well-validated Stroke Impact Scale (SIS)27 was used as the primary outcome measure to assess HRQOL. This preliminary study was conducted to estimate the magnitude of change and standard error between the 2 treatment groups in order to determine the sample size needed for a larger randomized clinical trial. Based on preliminary data that suggested HRQOL benefits associated with the combined therapy, we wanted to explore the relative HRQOL benefits of receiving therapist-supervised RTP only compared with receiving combined therapy among patients with reduced hand function secondary to stroke.
Method Design Overview A prospective, parallel group, randomized clinical trial with blinded assessments was conducted. All participants gave written informed consent prior to inclusion in the study. Setting and Participants A total of 109 patients with subacute stroke were assessed for eligibility. Participant eligibility criteria included: first clinical stroke diagnosis; status 3 to 9 months poststroke; MiniMental Status Examination28 score of ⬎24; ability to stand independently for 2 minutes; passive range of motion (PROM) of ⱖ45 degrees for abduction, flexion, or external rotation of the shoulder or pronation of the April 2010
forearm; active extension of the wrist of ⱖ10 degrees; active extension of the metacarpophalangeal and interphalangeal joints of the thumb; and ⱖ10 degrees of extension in at least 2 additional digits. Of the 109 patients assessed, 72 did not meet the inclusion criteria (in most cases due to overly high or low functional levels based on the specified criteria), 4 refused to participate, and 12 were excluded for other reasons (eg, lack of transportation, anticipated move to another area). Twenty-one patients with stroke were recruited at Emory University and at the Cleveland Clinic Foundation. Three patients withdrew after randomization due to transportation difficulties, and 1 patient did not complete the 2-month follow-up evaluation. Twelve patients completed training and evaluation at Emory University, and 5 patients were trained and evaluated at the Cleveland Clinic. Randomization and Interventions After screening by the study coordinator, participants were randomly assigned by the sealed envelope method to receive either 60 hours of RTP or 30 hours of RTP ⫹ 30 hours of robotic-assisted (ie, Hand Mentor§ [HM]) therapy over the course of 3 weeks. Therapists and evaluators had occupational or physical therapy backgrounds. Staff at the 2 research sites were trained to deliver standardized treatment and patient evaluation procedures. Research staff blinded to treatment assignment conducted interview-based outcome assessments. The participants were not blinded to their assignment but were not informed of the study hypotheses or primary outcome measures. The RTP tasks were selected by each participant, in collaboration with the §
Kinetic Muscles Inc, 2103 E Cedar St, #3, Tempe, AZ 85281 (http://www.kineticmuscles. com).
trainer, on the basis of personal preference, relevance, and interest. Most importantly, the tasks were selected so that the participant was challenged, rather than simply completing activities because of their apparent ease. Impairment training was not part of the RTP protocol for either group. Some typical activities were: ironing, potting a plant, holding cards, and so on. In the present context, the variables manipulated were related to the temporal and spatial domains for task completion. Upon selection, the tasks were broken into segments that required successful completion before the entire task was “put together.” For example, for the activity of potting a plant, the first component was reaching to grasp and release a hand shovel, then raising the shovel to the top of a pail and placing it back on the table. These individual actions were reinforced through RTP. Ultimately, the entire complement of movements included shoveling dirt into the pot and placing seedlings or plant stems in the pot. As can be deduced, this sequence placed progressively greater demands on multijoint control and sequencing. Improvements in spatial control were accomplished by having the therapist progressively move the object farther from the participant, thus imposing greater joint ranges of motion. Temporal domain elements were engaged by requiring the participant to repeat the task components or total task activity as frequently as possible during a defined time interval. The activities that the participant performed and the feedback provided were consistent with models of a massed RTP practice schedule.29 An RTP approach in which summary feedback is provided avoids potential complications related to feedback type and schedule for patients with stroke, especially as the optimal feedback schema for this population is unknown.
Volume 90
Number 4
Physical Therapy f
495
Health-Related Quality of Life and Robotic-Assisted Therapy for Hand Function were: motor control, recruitment, and spasticity reduction.30,31
Figure 1. Hand Mentor robotic system and its components.
Participants in the combined therapy group divided their time equally between the RTP activities with a therapist and supervised HM use. The selection of RTP tasks for this group was identical to task selection for the RTP-only group, and all RTP training was delivered in the same manner (eg, increasing temporal and spatial requirements of the activity over the course of practice based on the participants’ performance). Participants did not wear or use the HM during the performance of any RTP activities. The HM and its components are illustrated in Figure 1, and the system has been described in detail previously.30 Briefly, an air muscle (pneumatic actuator) provides the force necessary to extend the wrist and resist wrist flexion. Activation of the air muscle rotates a bar about a pivot point positioned in line with the axis of rotation about the wrist. This action extends the wrist and fingers. Wrist extension position is measured by a potentiometer, which is centered in-line with the pivot point and the axis of wrist flexion. Resistance to wrist flexion is measured by force496
f
Physical Therapy
Volume 90
sensitive resistors (FSRs). The FSR output is a measure of the resistance offered by multijoint stiffness attributable to changes in viscoelastic properties and muscle length stretch sensitivity in both the finger and wrist flexor muscle groups. Excessive force on the hand is prevented in several ways. A microcompressor has a maximum output pressure of 28 psi, thus limiting the maximum force produced by the air muscle. To prevent excessive extension of the wrist, a physical stop also is provided that limits motion of the activation bar at 60 degrees of wrist extension. The HM provided biofeedback to the participant during various activities requiring varying levels of wrist control. The therapist monitored HM usage and provided assistance in the donning and doffing of the device. The therapist also adjusted difficulty levels based on participant performance and comfort level. The primary goal of the HM is to improve AROM about the wrist and fingers (flexion-extension), wrist control, and initiation of distal movement. The 3 HM protocols used
Number 4
Briefly, the goal of the motor control protocol was to increase AROM at the wrist. The participants started with the wrist in a neutral position. Real-time wrist position was represented by a horizontal row of light-emitting diodes (LEDs) on the control box. Participants were instructed, via the ARM control box, to extend their wrist to the target LED. Once that target was achieved, the participants actively performed a wrist flexion, past neutral, to obtain the wrist flexion target. There were 2 motor control modes: a wrist flexion-extension mode and an extension-flexion mode. If a participant was unable to achieve the wrist extension goal, the air muscle would be activated and slowly extend the wrist to the target position. The difficulty of the task (eg, range of motion necessary to reach the LED targets) was systematically increased as a participant had multiple consecutive successful trials. By changing the sensitivity, success across the difficulty levels encapsulated a wrist AROM between 11 and 76 degrees. The aim of the recruitment protocol was to increase active wrist extensor muscle activity via feedback and assistive motion of the fingers and wrist. The participants were asked to relax their wrist and hand; this point was considered the initial starting position. Participants had to actively extend their hand and wrist to their maximum and hold their hand and wrist at that position for approximately 10 seconds. Over the next 20 seconds, the air muscle was inflated; during this time the fingers and wrist were extended to 60 degrees (or less if significant flexor tone [velocitydependent resistance to stretch] was present). The resulting position was held for 10 seconds, after which air pressure was released and the hand and fingers returned to the initial poApril 2010
Health-Related Quality of Life and Robotic-Assisted Therapy for Hand Function sition. This process was repeated for 10 to 20 cycles. The electromyographic (EMG) activity of the wrist extensors was presented as a line in the LED display. After each cycle, participants were encouraged to increase the height of the EMG and position lines on subsequent cycles using visual biofeedback. The aim of the spasticity protocol was to decrease flexor tone of the fingers and wrist via feedback and assistive motion of the fingers and wrist. The initial position of the hand was at approximately 30 degrees of flexion. The air muscle was inflated to bring the wrist to approximately half of the participants’ available PROM. The amount of force necessary to achieve this position was measured by FSRs within the ARM device while a potentiometer provided wrist position. A flexor stiffness ratio (ratio of force to position) was calculated and displayed on the LCD graph, and the participants’ goal was to decrease the amount of wrist flexor activity during the next 60 seconds. If flexor activity was reduced, the height (number) of illuminated LEDs, which was scaled to each participant, also decreased while wrist extension increased. After 60 seconds, the air pressure was released from the air muscle, and the wrist returned to its initial flexed position. Participants were encouraged to decrease the height of the line through relaxation on subsequent trials. They started in the easiest mode (ie, least amount of active wrist extension and flexion necessary to achieve the goal) and progressed to more difficult levels (ie, more active control of wrist extension and flexion) after consistently achieving an 85% success rate on a specific protocol. Participants in both groups adhered to all procedures and protocols and completed all of the training
April 2010
sessions. No adverse events were reported. Outcome Measures Data for the outcome measures were collected before the intervention, immediately postintervention, and 2 months postintervention. Primary outcome measure. Healthrelated quality of life was assessed by the 8 subscales and overall stroke recovery rating of the Stroke Impact Scale (SIS) version 3.0.32 The SIS version 3.0 comprises 59 items and assesses the domains of strength, memory/thinking, mood, communication, activities of daily living/instrumental activities of daily living (ADL/IADL), mobility, hand function, and social participation. An overall rating of stroke recovery also is included. Each domain contains a general description of the type of questions that follow and a statement with a reference to a specific time period (1, 2, or 4 weeks). Respondents score their performance on a 5-point scale (eg, “no strength” to “a lot of strength”; “none of the time” to “all of the time”). Duncan et al27 have shown the SIS to be valid, reliable, and sensitive to change, and other investigators also have concluded that the SIS has good psychometric properties.33 A 10- to 15-point change in a domain score may represent a clinically significant change.34 Covariates. Symptoms of depression are prevalent among patients who have had a stroke.35 Depressed mood was assessed by the 20-item Center for Epidemiologic Studies– Depression (CES-D) Scale.36 Additional covariates included in the analyses were age, sex, and race. Depressed mood, age, sex, and race influence reported HRQOL.3,33 Although the randomization process may ensure that intervention groups are not significantly different on these variables at baseline, with a small sample size, it is difficult to
ensure that intervention groups are balanced with respect to these characteristics. Follow-up. Follow-up was at 2 months postintervention for both groups. Sample Size Data to determine sample size were derived from an analysis of grasping kinetics among patients who participated in a previous CIMT study,4 in which an effect size of 1.0 was present from pretreatment to posttreatment. Using 1.0 as the estimated effect size, with a sample of 8 participants in the combined therapy group, pretreatment to posttreatment change could be detected with power of 0.68 using a 2-sided significance level of .05. In addition, assuming the same effect size, the test for time ⫻ treatment interaction effects (with treatment being either combined therapy or RTP alone) had power of 0.67 using a 2-sided significance level of .05 with 8 participants per group. Data Analysis Internal consistency reliability of each of the 8 SIS domains was evaluated by Cronbach alpha. All domains had adequate reliability (alpha exceeded 0.7). Baseline characteristics of the 2 intervention groups were compared by t test (continuous variables) and chi-square analysis (categorical variables). The HRQOL outcomes for the 2 groups, as measured by average change in SIS scores, were examined in a mixedeffects model, with random effects for patient and the estimate of interest being the time ⫻ treatment interaction. The 2 factors—treatment (RTP only, combined therapy) and time (preintervention, postintervention, and follow-up)—were included in the model, and the average response for each combination was modeled. The SIS scores were adjusted for participants’ age, sex, race, and baseline
Volume 90
Number 4
Physical Therapy f
497
Health-Related Quality of Life and Robotic-Assisted Therapy for Hand Function CES-D Scale score. Analyses were performed using SAS proc MIXED in SAS version 9.0.㛳 Role of the Funding Source This study was funded by a grant (R21 HD057020) from the National Institute of Neurological Disorders and Stroke, National Institutes of Health (NIH), to Dr Alberts. The study sponsor had no role in the study’s conduct or reporting. Recommendations received from NIH reviewers in the application phase were taken into consideration in the study design.
Results Figure 2 details the recruitment and passage of participants through the trial. At 2 months postintervention, 7 participants in the RTP-only group and 10 participants in the combined therapy group had successfully completed the study. Randomization resulted in a distribution of patient demographic and clinical characteristics that was not statistically different at baseline for the 2 intervention groups (Tab. 1). The average age of the participants who were randomly assigned to the RTPonly group was 51 years, and the average age of the participants who
Figure 2. Flow of participants through the trial. RTP⫽repetitive task practice.
㛳
SAS Institute Inc, PO Box 8000, Cary, NC 27513.
Table 1. Characteristics of Study Participants as a Whole and by Intervention Groupa Total (nⴝ17)
RTP-Only Group (nⴝ7)
Combined Therapy Group (nⴝ10)
P
57.4 (13.4)
51.0 (11.3)
61.9 (13.4)
.10
Variable Mean age (SD) Men/women
10/7
5/2
5/5
.38
9/8
3/4
6/4
.49
234.4 (121.8)
184.1 (126.5)
269.6 (111.1)
.16
12/5
5/2
7/3
.95
28.5 (1.4)
29.0 (1.3)
28.2 (1.5)
.27
9.6 (6.2)
7.3 (3.8)
11.3 (7.2)
.20
White/African American Days after stroke, mean (SD) Ischemic stroke/hemorrhagic stroke MMSE score, mean (SD) CES-D Scale score, mean (SD) a
RTP⫽repetitive task practice, MMSE⫽Mini-Mental State Examination, CES-D⫽Center for Epidemiologic Studies–Depression.
498
f
Physical Therapy
Volume 90
Number 4
April 2010
Health-Related Quality of Life and Robotic-Assisted Therapy for Hand Function were randomly assigned to the combined therapy group was approximately 62 years (P⫽.10). Participants in the RTP-only group were, on average, 184.1 days poststroke, and participants in the combined therapy group were, on average, 269.6 days poststroke (P⫽.16). Average CES-D Scale baseline scores for participants in the 2 groups did not differ significantly (P⫽.20) and were below the suggested cutoff (16) for possible clinical depression.36 As measured by the Fugl-Meyer Motor Assessment (maximum score⫽66),37 participants were similar in severity of upper-limb motor impairment at baseline, with a mean (SD) score of 39.4 (6.6) in the RTP-only group and 33.6 (9.7) in the combined therapy group.
Table 2. Mean (SD) Stroke Impact Scale (SIS) Domain Scores and Mean (SD) Stroke Recovery Rating, by Intervention Group and Study Visita RTP-Only Group (nⴝ7) SIS Domain
Follow-up
Strength
58.0 (11.2)
56.3 (12.0)
60.7 (17.2)
Memory
86.2 (19.4)
85.7 (14.9)
86.7 (16.7)
Mood
89.3 (5.9)
90.1 (5.8)
89.3 (7.1)
Communication
90.4 (14.4)
87.8 (16.2)
90.3 (14.8)
ADL/IADL
67.9 (15.8)
80.0 (16.3)
75.7 (21.3)
Mobility
75.4 (17.7)
80.2 (17.2)
74.2 (26.0)
Hand function
40.0 (31.4)
57.9 (26.9)
57.9 (27.5)
Social participation
51.8 (15.5)
61.2 (22.2)
67.9 (20.3)
Stroke recovery
42.9 (24.8)
55.7 (21.3)
63.6 (23.0)
SIS Domain
a
April 2010
Postintervention
Combined Therapy Group (nⴝ10)
Mean (SD) SIS scores for participants in the 2 groups at baseline, postintervention, and 2-month follow-up are shown in Table 2. Baseline SIS scores were not significantly different for the participants in the 2 treatment groups. When average SIS score changes observed in the 2 groups were compared, group ⫻ time interaction effects were evident for 2 SIS domains (Tab. 3). First, compared with average change in mood ratings in the RTP-only group, the average change in mood ratings in the combined therapy group was greater from preintervention to postintervention. The estimated effect was ⫺8.0 (P⫽.03). Second, compared with average change in social participation ratings in the combined therapy group, the average change in social participation ratings in the RTP-only group was greater from preintervention to follow-up. The estimated effect was 18.4 (P⫽.008). The data reported in Table 2 provide information about the nature of these changes in the 2 groups.
Baseline
Postintervention
Follow-up
Strength
47.5 (18.4)
Baseline
54.4 (9.3)
55.7 (18.5)
Memory
73.6 (21.8)
76.1 (19.4)
76.1 (16.1)
Mood
74.7 (13.2)
81.9 (13.0)
70.2 (11.5)
Communication
79.3 (17.7)
84.3 (14.9)
82.1 (14.9)
ADL/IADL
58.8 (18.8)
67.5 (17.7)
62.6 (17.3)
Mobility
58.1 (17.5)
63.3 (16.1)
62.2 (20.8)
Hand function
24.0 (16.0)
52.0 (17.5)
47.0 (22.3)
Social participation
49.1 (22.0)
52.8 (14.0)
48.7 (21.4)
Stroke recovery
48.0 (18.6)
67.0 (13.8)
66.0 (19.1)
RTP⫽repetitive task practice, ADL/IADL⫽activities of daily living/instrumental activities of daily living.
As indicated in Table 4, both RTP only and the combined therapy were associated with a statistically significant improvement in rating of the ADL/IADL domain from preintervention to postintervention, but the estimated changes did not meet the criterion for potential clinical significance. Both RTP only and the combined therapy were associated with a statistically significant improvement in rating of hand function at postintervention and at follow-up, and the magnitude of these changes suggested clinical significance for both intervention groups. Combined therapy was associated with a statistically significant improvement in stroke recovery rating postintervention and at follow-up, and both of these changes suggested clinical sig-
nificance. The RTP-only intervention was associated with a statistically significant improvement in stroke recovery rating at follow-up, and this change suggested clinical significance.
Discussion Patients who have had a stroke can experience accelerated gains in quality of life after participating in targeted interventions that focus on improving upper-extremity strength.38 Our study indicated improvement in ratings of hand function from baseline to follow-up among patients whose therapy time was divided between the HM and RTP, as well as among patients who received twice as much RTP. Based on ratings obtained with the SIS, assessment of hand function improved over time,
Volume 90
Number 4
Physical Therapy f
499
Health-Related Quality of Life and Robotic-Assisted Therapy for Hand Function Table 3. Estimated Intervention Group ⫻ Time Interaction Effects on Stroke Impact Scale (SIS) Domain Scores and Stroke Recovery Ratinga From Preintervention to Postintervention SIS Domain
Estimate
95% CI
P
From Preintervention to Follow-up Estimate
95% CI
P
Strength
⫺9.2
⫺20.7 to 2.3
.11
⫺5.0
⫺23.0 to 13.0
.56
Memory
⫺4.2
⫺10.1 to 1.7
.15
⫺0.9
⫺8.5 to 6.6
.79
Mood
⫺8.0
⫺15.3 to ⫺0.7
.03b
Communication
⫺8.3
⫺22.9 to 6.2
.24
1.6
⫺8.8 to 12.0
.74
5.6
⫺1.9
⫺10.9 to 7.1
.65
⫺4.1
⫺16.5 to 8.3
ADL/IADL Mobility
⫺4.7 to 16.5
.25
⫺16.8 to 12.6
.76
⫺7.2 to 18.5
⫺29.9 to 6.7
.19
⫺3.7
⫺25.6 to 18.1
Social participation
3.4
⫺12.1 to 19.0
.64
18.4
5.7 to 31.1
Stroke recovery
4.3
⫺25.1 to 9.3
.34
4.3
⫺16.7 to 25.3
Hand function
⫺11.6
5.9 ⫺2.1
.36 .48 .71 .008b .66
a
All analyses were adjusted for the effects of age, sex, race, and baseline Center for Epidemiologic Studies–Depression Scale score. 95% CI⫽95% confidence interval, ADL/IADL⫽activities of daily living/instrumental activities of daily living. b Average change over time in repetitive task practice– only group and average change over time in combined therapy group significantly different.
with no significant difference between the 2 intervention groups. In addition, the combined therapy group reported improvement in perceived overall stroke recovery postintervention and at follow-up, and the RTPonly group reported improvement in perceived overall stroke recovery at follow-up. The clinical significance of hand function improvement may generalize to a more positive perception of individuals’ poststroke level of functioning. Among the participants in this study, hand function was rated lower at enrollment than any of the other domains measured by the SIS. This observation also was true of the 222 patients who were enrolled into the EXCITE trial.4 Involuntary activation of flexor muscles coupled with extensor paresis affects hand function, impeding the ability to perform dexterous activities that require skilled hand use (ie, control and coordination of grasping forces and appropriate opening and closing of the hand). A qualitative study of patients who were ⱖ7 months poststroke showed that residual impairments in ability to handle newspapers, use paper clips, put things in envelopes, and 500
f
Physical Therapy
Volume 90
hang clothes in the closet were a source of great distress and reduced perception of life quality.3 Although the neurophysiologic bases for motor recovery after stroke are not completely understood, neuronal cortical connections and cortical representation areas appear modifiable by sensory input, experience, and learning.39 – 41 Rehabilitation therapies that incorporate principles of motor learning, such as RTP, can facilitate cortical reorganization.42
may result in an improved perception of hand function. The consistent, repetitive, and progressive nature of the HM may increase the quantity and quality of sensorimotor information provided to the patient, which has been shown to modulate motor cortex function and excitability43,44 and to promote motor learning.45 An increase in quantity and quality of afferent information provided to the patient during robotic device use may facilitate motor learning or relearning.
Robotic devices, such as the HM, that use motor learning principles (ie, engaging the user, providing meaningful feedback during the performance of repetitive activities that are intended to transfer to functional distal motor activities, and utilizing a massed practice training schedule) may be a valuable component of or adjunct to task practice interventions. Although the exact mechanism responsible for improved rating of hand function following an abbreviated RTP intervention while using the HM is unknown, it is likely that an increase in the AROM of the wrist and fingers and a possible reduction in spasticity of the hemiparetic limb
It is fully acknowledged that sole use of the HM, in the absence of therapist-directed RTP, would not be expected to result in the same magnitude of improvement in hand function rating that was seen in the combined therapy group in the current study. It is not envisioned that the introduction of a robotic system will replace therapist-directed rehabilitation approaches. Rather, the appeal of robotic systems such as the HM is that they can provide consistent and precise therapy for long durations without fatigue while generally requiring relatively minimal supervision. Therefore, we envision the HM as providing a “helping hand” to
Number 4
April 2010
Health-Related Quality of Life and Robotic-Assisted Therapy for Hand Function Table 4. Estimated Differences in Mean Stroke Impact Scale (SIS) Domain Scores and Mean Stroke Recovery Rating, by Time and Intervention Groupa Preintervention to Postintervention SIS Domain and Stroke Recovery
Preintervention to Follow-up
RTP-Only Group
Combined Therapy Group
RTP-Only Group
Combined Therapy Group
Estimate
⫺2.86
6.32
2.58
7.57
95% CI
⫺11.84 to 6.12
⫺1.08 to 13.74
⫺11.22 to 16.38
⫺4.02 to 19.16
.50
.09
.69
.18
Estimate
⫺2.93
1.27
0.29
1.22
95% CI
⫺7.54 to 1.67
⫺2.53 to 5.06
⫺5.49 to 6.07
⫺3.65 to 6.09
.19
.48
.91
.60
Estimate
⫺2.38
5.61
⫺0.29
⫺6.19
95% CI
⫺8.09 to 3.33
0.89 to 10.21
⫺8.43 to 7.85
⫺13.03 to 0.65
.38
.02
.94
.07
Estimate
⫺4.17
4.18
⫺0.15
1.95
95% CI
⫺15.48 to 7.14
⫺5.21 to 13.56
⫺11.44 to 11.14
⫺7.54 to 11.45
.44
.35
.98
.66
Estimate
8.49
6.89
7.53
1.88
95% CI
0.39 to 16.60
0.18 to 13.61
⫺2.35 to 17.40
⫺6.42 to 10.17
.04
.04
.12
.63
Estimate
1.91
3.82
⫺1.45
2.64
95% CI
⫺5.16 to 8.98
⫺2.00 to 9.65
⫺10.92 to 8.02
⫺5.32 to 10.61
.57
.18
.74
.48
Estimate
14.85
26.47
17.58
21.37
95% CI
0.64 to 29.06
14.69 to 38.26
0.84 to 34.22
7.31 to 35.44
.04
⬍.001
.04
.006
Estimate
4.93
1.48
15.67
⫺2.74
95% CI
⫺7.07 to 16.94
⫺8.49 to 11.46
5.95 to 25.39
⫺10.93 to 5.45
.39
.75
.004
.48
9.31
17.20
20.39
16.08
⫺4.08 to 22.7
6.12 to 28.27
4.32 to 36.46
2.56 to 29.60
.16
.005
.02
.02
Strength
P Memory
P Mood
P Communication
P ADL/IADL
P Mobility
P Hand function
P Social participation
P Stroke recovery Estimate 95% P a
All analyses were adjusted for the effects of age, sex, race, and baseline Center for Epidemiologic Studies–Depression Scale score. RTP⫽repetitive task practice, 95% CI⫽95% confidence interval, ADL/IADL⫽activities of daily living/instrumental activities of daily living.
April 2010
Volume 90
Number 4
Physical Therapy f
501
Health-Related Quality of Life and Robotic-Assisted Therapy for Hand Function the therapist, thereby decreasing the time necessary for RTP interventions, which— despite their effectiveness— have not been adopted clinically on a widespread basis. At the same time, the degree to which hand function and stroke recovery rating outcomes may reflect the particular skills or personality of the therapists who worked with the patients remains unknown. In addition, the relatively small sample size of this study demands that additional studies be conducted to specify the amount of RTP and robotic-assisted intervention time that is both effective and acceptable to patients. Change over time in reported HRQOL of patients who have had a stroke may reflect response shift rather than true change, as Ahmed et al46 have discussed. Based on their study of a large number of patients with stroke and controls, however, Ahmed et al concluded that improvement in HRQOL over time is likely to be real rather than the result of individuals’ reconceptualization or recalibration over the first 6 months poststroke.46 Stroke is increasingly recognized as a significant and expensive medical and societal problem.47 Even patients who are highly recovered following stroke may report significant residual disability in hand function,48 indicating that rehabilitation interventions that target this dimension of upper-extremity functioning should be made widely available. Roboticassisted therapy has an important role in technologic intervention in the modern neurorehabilitation setting.49 Patients in the combined therapy group in the current study used the HM device under the supervision of a therapist. Based on feedback from the therapists and patients who participated in this study, we believe that the HM device could be used in a home environment or in a group setting in which the therapist oversees a number of patients with stroke 502
f
Physical Therapy
Volume 90
who use HM systems simultaneously. A home and group clinical mode of HM use has the potential to dramatically reduce the amount of therapistdirected RTP without compromising perceived gains in hand function associated with a more traditional timeand labor-intensive RTP intervention. Reduced therapist-directed rehabilitation time without loss in quality of life ratings suggests that incorporation of robotic-assisted therapy may be an effective approach to enhancing motor recovery in patients who have had a stroke, while decreasing labor, the most costly aspect of delivering physical therapy interventions. This reduction in cost may promote greater acceptance of RTP approaches in clinical environments and the acceptance of robotic systems that function as adjuncts to therapist-directed interventions. Limitations The combined therapy group received 30 hours of RTP compared with 60 hours for the RTP-only group. A limitation of the study was the lack of understanding about the necessary dose of RTP to elicit the current level of improvements in motor function.50 Limiting training to 30 hours of RTP might be sufficient for patients to reach a plateau in terms of improved motor function. In this case, the robotic device may not be contributing to change in the perception of hand function. We are currently engaged in studies examining the dose response of RTP and RTP⫹HM to dissociate the contributions of task practice and robotic therapy to enhanced perception of hand function. Although a potential benefit of utilizing a robotic device is the reduction of the amount of therapist-directed RTP, thereby potentially providing a cost savings to the delivery of RTP interventions, the results from the current study do not directly assess the potential cost savings. In the cur-
Number 4
rent study, the HM was used in a clinical environment and under the supervision of a therapist. It is unclear whether the same level of intensity and adherence would occur if a patient used the device in a home environment. Future studies in which the HM is used in a home environment to complete outpatient RTP will be conducted. These subsequent studies will provide important data regarding any cost savings associated with the HM, feasibility of home use, and patient adherence to intended use of the system. It also will be important to continue to assess the potential effect of decreased therapist time on patients’ rating of the social participation dimension of their perceived quality of life. At the same time, the current data are the first to show that HRQOL measures specific to distal hand function can be improved following a combined robotic and RTP therapeutic approach.
Conclusion Receipt of 30 hours of therapistsupervised repetitive task practice combined with 30 hours of roboticassisted therapy during the subacute phase of stroke recovery resulted in patient-rated improvement in hand function similar to that observed with receipt of 60 hours of therapistsupervised repetitive task practice. Patients assigned to the 2 interventions also reported a higher level of overall stroke recovery at followup compared with baseline, and patients in the combined therapy group reported higher overall stroke recovery postintervention as well. Incorporating robotic-assisted therapy may be an effective approach to enhancing motor recovery in patients who have had a stroke while decreasing labor, which is the most costly aspect of delivering physical therapy interventions. A potential decrease in therapist labor may enApril 2010
Health-Related Quality of Life and Robotic-Assisted Therapy for Hand Function hance the clinical acceptance and potential reimbursement of task practice approaches such as CIMT and modified CIMT. Dr Kutner, Dr Butler, Dr Wolf, and Dr Alberts provided writing. Ms Zhang provided data analysis. Dr Kutner, Dr Butler, Dr Wolf, and Dr Alberts provided concept/idea/research design and project management. The authors thank Veronica Rowe, OTR, Kelly Crabtree, PT, Libby Rosenstein, OTR, and Vanessa Franco, OT/L, for their invaluable assistance in the training and evaluation of participants and Anil Thota for assistance in data processing. Ethical approval was obtained from the Emory University and Cleveland Clinic Foundation institutional review boards. Dr Wolf is a paid consultant for Kinetic Muscles Inc. He is not a shareholder or investor in the company. This study was funded by a grant (R21 HD057020) from the National Institute of Neurological Disorders and Stroke, National Institutes of Health, to Dr Alberts. Trial NCT00729625 registered at: https:// register.clinicaltrials.gov/prs/app/action/Filter OrSelectProtocol/selectaction/View/ts/2/uid/ U0000O71. This article was received May 15, 2009, and was accepted December 2, 2009. DOI: 10.2522/ptj.20090160
References 1 Dobkin BH. Rehabilitation after stroke. N Engl J Med. 2005;352:1677–1684. 2 Heart disease and stroke statistics—2008 update: a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation. 2008;117:e25– e146. 3 Clarke P, Black SE. Quality of life following stroke: negotiating disability, identity, and resources. J Appl Gerontol. 2005;24:319 – 336. 4 Wolf SL, Winstein CJ, Miller JP, et al. Effect of constraint-induced movement therapy on upper extremity function 3 to 9 months after stroke: the EXCITE randomized clinical trial. JAMA. 2006;296: 2095–2104. 5 Woldag H, Hummelsheim H. Evidencebased physiotherapeutic concepts for improving arm and hand function in stroke patients: a review. J Neurol. 2002;249: 518 –528. 6 Mulder T, Zijlstra W, Geurts A. Assessment of motor recovery and decline. Gait Posture. 2002;16:198 –210.
April 2010
7 Wolf SL. Revisiting constraint-induced movement therapy: are we too smitten with the mitten? Is all nonuse “learned”? And other quandaries. Phys Ther. 2007;87: 1212–1223. 8 Krebs HI, Volpe BT, Aisen ML, Hogan N. Increasing productivity and quality of care: robot-aided neuro-rehabilitation. J Rehabil Res Dev. 2000;37:630 – 652. 9 Krebs HI, Hogan N, Aisen ML, Volpe BT. Robot-aided neurorehabilitation. IEEE Trans Rehabil Eng. 1998;6:75– 87. 10 Daly JJ, Hogan N, Perepezko EM, et al. Response to upper-limb robotics and functional neuromuscular stimulation following stroke. J Rehabil Res Dev. 2005;42: 723–736. 11 Reinkensmeyer DJ, Kahn LE, Averbuch M, et al. Understanding and treating arm movement impairment after chronic brain injury: progress with the arm guide. J Rehabil Res Dev. 2000;37:653– 662.
23 Brooks VB, Stoney SD Jr. Motor mechanisms: the role of the pyramidal system in motor control. Ann Rev Physiol. 1971;33: 337–392. 24 Humphrey DR, Schmidt EM, Thompson WD. Predicting measures of motor performance from multiple cortical spike trains. Science. 1970;170:758 –762. 25 Waldvogel D, vanGelderen P, Ishii K, Hallett M. The effect of movement amplitude on activation in functional magnetic resonance imaging studies. J Cereb Bood Flow Metab. 1999;19:1209 –1212. 26 Brewer BR, McDowell SK, WorthenChaudhari LC. Poststroke upper extremity rehabilitation: a review of robotic systems and clinical results. Top Stroke Rehabil. 2007;14:22– 44. 27 Duncan PW, Wallace D, Lai SM, et al. The Stroke Impact Scale version 2.0: evaluation of reliability, validity and sensitivity to change. Stroke. 1999;30:2131–2140.
12 Volpe BT, Ferraro M, Lynch D, et al. Robotics and other devices in the treatment of patients recovering from stroke. Curr Neurol Neurosci Rep. 2005;5:465– 470. 13 Krebs HI, Ferraro M, Buerger SP, et al. Rehabilitation robotics: pilot trial of a spatial extension for mit-manus. J Neuroeng Rehabil. 2004;1:5. 14 Ferraro M, Palazzolo JJ, Krol J, et al. Robotaided sensorimotor arm training improves outcome in patients with chronic stroke. Neurology. 2003;61:1604 –1607. 15 Volpe BT, Krebs HI, Hogan N, et al. A novel approach to stroke rehabilitation: robot-aided sensorimotor stimulation. Neurology. 2000;54:1938 –1944. 16 Lum PS, Burgar CG, Shor PC, et al. Robotassisted movement training compared with conventional therapy techniques for the rehabilitation of upper-limb motor function after stroke. Arch Phys Med Rehabil. 2002;83:952–959. 17 Lum PS, Burgar CG, Van der Loos M, et al. Mime robotic device for upper-limb neurorehabilitation in subacute stroke subjects: a follow-up study. J Rehabil Res Dev. 2006;43:631– 642. 18 Kwakkel G, Kollen BJ, Krebs HI. Effects of robot-assisted therapy on upper limb recovery after stroke: a systematic review. Neurorehabil Neural Repair. 2008;22: 111–121. 19 Kwakkel G, Kollen B, Lindeman E. Understanding the pattern of functional recovery after stroke: facts and theories. Restor Neurol Neurosci. 2004;22:281–299. 20 Van Peppen RPS, Kwakkel G, WoodDauphine´e S, et al. The impact of physical therapy on functional outcomes after stroke: what’s the evidence? Clin Rehabil. 2004;18:833– 862. 21 Blanton S, Wolf SL. An application of upper-extremity constraint-induced movement therapy in a patient with subacute stroke. Phys Ther. 1999;79:847– 853. 22 Taub E, Wolf SL. Constraint induction techniques to facilitate upper extremity use in stroke patients. Top Stroke Rehabil. 1997;3:38 –59.
28 Folstein MF, Folstein SE, McHugh PR. “Mini-Mental State”: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189 –198. 29 Lee TD, Magill RA, Weeks DJ. Influence of practice schedule on testing schema theory predictions in adults. J Mot Behav. 1985;17:283–299. 30 Rosenstein L, Ridgel AL, Thota A, et al. Effects of combined robotic therapy and repetitive-task practice on upper extremity function in a patient with chronic stroke. Am J Occup Ther. 2008;62:28 –33. 31 Frick EM, Alberts JL. Combined use of repetitive task practice and an assistive robotic device in a patient with subacute stroke. Phys Ther. 2006;86:1378 –1386. 32 Stroke Impact Scale version 3.0. Available from the Rehabilitation Outcomes Research Center for Veterans (RORC) at: https:// www.maa.nsw.gov.au/getfile.aspx?Type⫽ document&ID⫽36614&ObjectType⫽3& ObjectID⫽3322. 33 Edwards B, O’Connell B. Internal consistency and validity of the Stroke Impact Scale 2.0 (SIS 2.0) and SIS-16 in an Australian sample. Qual Life Res. 2003;12:1127– 1135. 34 Cook C, McCluskey A, Bowman J. Increasing the Use of Outcome Measures by Occupational Therapists [final research report]. Penrith South, New South Wales, Australia: University of Western Sydney; 2006. 35 Whyte EM, Mulsant BH, Vanderbilt J, et al. Depression after stroke: a prospective epidemiological study. J Am Geriatr Soc. 2004;52:774 –778. 36 Radloff LS. The CES-D scale: a new selfreport depression scale for research in the general population. Appl Psychol Meas. 1977;1:385– 401. 37 Fugl-Meyer AR, Jaasko L, Leyman I, et al. The post-stroke hemiplegic patient: a method of evaluation of physical performance. Scand J Rehabil Med. 1975;7: 13–31.
Volume 90
Number 4
Physical Therapy f
503
Health-Related Quality of Life and Robotic-Assisted Therapy for Hand Function 38 Studenski S, Duncan PW, Perera S, et al. Daily functioning and quality of life in a randomized controlled trial of therapeutic exercise for subacute stroke survivors. Stroke. 2005;36:1764 –1770. 39 Johansson BB. Brain plasticity and stroke rehabilitation: The Willis Lecture. Presented at: 24th American Heart Association International Conference on Stroke and Cerebral Circulation; February 4, 1999; Nashville, Tennessee. 40 Merzenich MM. Variability in hand surface representations in areas 3b and 1 in adult owl and squirrel monkeys. J Comp Neurol. 1987;258:281–296. 41 Jenkins WM. Reorganization of neocortical representations after brain injury: a neurophysiological model of the bases of recovery from stroke. Prog Brain Res. 1987;71:249 –266.
Invited Commentary
f
Physical Therapy
Volume 90
47 Dion JE. Management of ischemic stroke in the next decade: stroke centers of excellence. J Vasc Interv Radiol. 2004;15: S133–S141. 48 Lai S-M, Studenski S, Duncan PW, Perera S. Persisting consequences of stroke measured by the Stroke Impact Scale. Stroke. 2002;33:1840 –1844. 49 Rocksmith ER, Reding MJ. New developments in stroke rehabilitation. Curr Atheroscler Rep. 2002;4:277–284. 50 Dobkin BH. Confounders in rehabilitation trials of task-oriented training: lessons from the designs of the excite and scilt multicenter trials. Neurorehabil Neural Repair. 2007;21:3–13.
Alma S. Merians
There is an increasing interest in the use of robotic-assisted therapy to facilitate and augment upperextremity movement for people with hemiparesis. This focus is an attempt either to improve patients’ rehabilitation outcomes beyond our current capabilities or, as Kutner et al1 suggest, to decrease the therapist time demands necessary for the delivery of repetitive task practice. The robotic assistive devices have been used in accordance with current neuroscience literature in animals and motor control literature in humans. They take advantage of recent improvements in robotic design, the development of haptic interfaces, and the use of virtual reality simulations, interfaced with the robots. Most of the robots have been designed to train the shoulder and elbow, supporting the limb as needed, with the limb moved either passively2,3 or requiring active movement.4 –7 Far fewer robots have been designed to train the arm and wrist7 or arm and hand together.8 –12 Some of the robots emphasize impairment-based practice,3,4 whereas other robots focus on more task-based practice,7,9,10 with the complexity of sen-
504
42 Kaplon RT, Prettyman MG, Kushi CL, Winstein CJ. Six hours in the laboratory: a quantification of practice time during constraint-induced therapy (CIT). Clin Rehabil. 2007;21:950 –958. 43 Ridding MC, Brouwer B, Nordstrom MA. Reduced interhemispheric inhibition in musicians. Exp Brain Res. 2000;133:249 –253. 44 Kaelin-Lang A, Cohen LG. Enhancing the quality of studies using transcranial magnetic and electrical stimulation with a new computer-controlled system. J Neurosci Methods. 2000;102:81– 89. 45 Rossini PM, Dal Forno G. Integrated technology for evaluation of brain function and neural plasticity. Phys Med Rehabil Clin N Am. 2004;15:263–306. 46 Ahmed S, Mayo NE, Corbiere M, et al. Change in quality of life of people with stroke over time: true change or response shift? Qual Life Res. 2005;14:611– 627.
sory feedback ranging from simple4 to complex, 2- and 3-dimensional, interactive simulations.9 –14 Two reviews found that robotassisted therapy showed potential to improve upper-extremity function15,16 and improve strength.16 However, neither review could confirm evidence for improvement in activities of daily living, which may reflect an inability of activities of daily living scales to accurately demonstrate changes in paretic limb function.15 In this study, Kutner et al showed that 30 hours of therapist-supervised repetitive task practice combined with 30 hours of robotic-assisted therapy resulted in similar reports of improvement in hand function, as measured by the patients’ responses to 5 questions on the Stroke Impact Scale (ie, carrying heavy objects, turning a knob, opening a can/jar, tying a shoelace, and picking up a dime), as well as improvement in the measure of overall stroke recovery. The robotic device used in this experiment was the Hand Mentor.
Number 4
The Hand Mentor has preprogrammed activities that provide continuous stretch of preset durations to the finger flexors (termed “antispasticity program”), a wrist flexion or extension strengthening protocol, and a muscle recruitment protocol, where the patient receives feedback regarding the intensity of the electromyographic signal related to wrist and finger movement. As this is an impairment-based system, it appears that, in addition to comparing the use of the robot as a therapeutic adjunct to task practice, the authors also were comparing particular therapeutic principles, specifically taskbased training with a combination of task-based training and impairmentlevel interventions. The authors’ findings of similar improvements in reported hand function and stroke recovery raise several interesting questions. What is the role of impairment-level interventions versus functional training? Is the combination important, and if it is important, what is the ideal dosing for each component? Is the total intensity or amount of practice more important than the difference in approach?
April 2010
Health-Related Quality of Life and Robotic-Assisted Therapy for Hand Function 8 Brewer BR, McDowell SK, WorthenChaudhari LC. Poststroke upper extremity rehabilitation: a review of robotic systems and clinical results, Topics Stroke Rehabil. 1993;14:22– 44. 9 Merians AS, Tunik E, Fluet GB, et al. Innovative approaches to rehabilitation of upper extremity hemiparesis using virtual environments. Eur J Phys Rehabil Med. 2009;45:123–133. 10 Adamovich SV, Fluet GG, Mathai A, et al. Design of a complex virtual reality simulation to train finger motion for persons with hemiparesis: a feasibility study. J Neuroeng Rehabil. 2009;17:28. 11 Qiu Q, Fluet GG, Lafond I, et al. Coordination changes demonstrated by subjects with hemiparesis performing hand-arm training using the NJIT-RAVR robotically assisted virtual rehabilitation system. Conf Proc IEEE Eng Med Biol Soc. 2009;1: 1143–1146.
Author Response
f
Physical Therapy
Volume 90
17 Lin KC, Fu T, Wu CY, et al. Minimal detectable change and clinically important difference of the Stroke Impact Scale in stroke patients. Neurorehabil Neural Repair. 2010 Jan 6 [Epub ahead of print]. 18 Schmidt RA. Motor learning principles for physical therapy. In: Lister MJ, ed. Contemporary Management lf Motor Control Problems: Proceedings of the II STEP Conference. Alexandria, VA: Foundation for Physical Therapy. 1991:49 – 63. 19 Beebe JA, Lang CE. Relationships and responsiveness of six upper extremity function tests during the first six months of recovery after stroke. J Neurol Phys Ther. 2009:33:96 –103. 20 Jebsen RH, Taylor N, Trieschmann RB, et al. An objective and standardized test of hand function. Arch Phys Med Rehabil. 1969;50:311–319. 21 Wolf SL, Thompson PA, Morris DM, et al. The EXCITE trial: attributes of the Wolf Motor Function Test in patients with subacute stroke. Neurorehabil Neural Repair. 2005;19:194 –205.
Jay L. Alberts, Steven L. Wolf, Nancy G. Kutner
We appreciate the interest in our research1 shown by Merians,2 and we are pleased to have the opportunity to comment on some of the important points that she has raised. Merians observes that “it appears that, in addition to comparing the use of the robot as a therapeutic adjunct to task practice, the authors also were comparing particular therapeutic principles, specifically task-based training with a combination of task-based training and impairment-level interventions.” In essence, this project did present an opportunity to assess a robotic device that could be used as a therapeutic adjunct and allow for further study of a task-based training approach. It is important to note that the Hand Mentor is preloaded with patient modules that are designed to reduce spasticity, enhance motor unit recruitment, and encourage controlled active motor function about the wrist. The spasticity module is considered one that addresses a specific impairment. The time spent training in the spasticity module varied across patients and depended primarily on their level of 506
12 Connelly L, Stoykov ME, Jia Y, et al. Use of a pneumatic glove for hand rehabilitation following stroke. Presented at: 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society; September 2– 6, 2009; Minneapolis, Minnesota. 13 Nef T, Mihelj M, Riener R. ARMin: a robot for patient-cooperative arm therapy. Med Bio Eng Comput. 2007;45:887–900. 14 Takahashi CD, Der-Yeghiaian L, Le V, et al. Robot-based hand motor therapy after stroke. Brain. 2008;131(pt 2):425– 437. 15 Kwakkel G, Kollen BJ, Krebs HI. Effects of robot-assisted therapy on upper limb recovery after stroke: a systematic review. Neurorehabil Neural Repair. 2008;22: 111–121. 16 Mehrholz J, Platz T, Kugler J, Pohl M. Electromechanical and robot-assisted arm training for improving arm function and activities of daily living after stroke. Cochrane Database System Rev. 2008;4: CD006876
spasticity. The aim of the spasticity module was to reduce resistance to movement during wrist extension motion while the other modules were more focused on enhancing the participant’s capacity to recruit the wrist extensors and engage in active control of extensors and flexors about the wrist. During the taskbased training portion of each session, participants were encouraged to practice and engage in activities that were important to them. Although such encouragement was not robotically driven, it certainly falls within the domain of task-specific training. In our study, we reported that hand function and stroke recovery rating improved for participants in both intervention groups, leading Merians to raise several questions and points that we would like to address: What is the role of impairment-level interventions versus functional training? This question is not easily addressed and forms the cornerstone of much of the applied research cur-
Number 4
rently under way in neurorehabilitation. The question is posed in an “either/or” mode, when, in fact, given the rapidly constricting time for both inpatient and outpatient rehabilitation services, coupled with the paucity of substantive evidence for successful outcomes using impairment-based approaches in the treatment of the upper extremities of patients with either acute or chronic stroke, a functional approach is evolving as more practical and defensible. However, the basis for progressing functional training must entail consideration of impairment levels and how such levels are responsive to either repetition or shaping for retraining to capture a defined function.3 Is the combination important, and if it is important, what is the ideal dosing for each component? We agree that determining ideal dosing is an important issue. This point is of concern for all rehabilitation studies.4 Our collective inability to determine the relationship of optimal intensity or dosing to achieve best April 2010
Health-Related Quality of Life and Robotic-Assisted Therapy for Hand Function outcomes has plagued many rehabilitation studies. With respect to an intervention such as constraintinduced movement therapy (CIMT), we did not see a clear relationship between meticulously documented amount of training, using repetitive task practice or shaping, and outcome among patients with subacute or chronic stroke.5 The precise dosing necessary to achieve optimal results is unknown at this time. An acknowledged limitation of our study and similar long-term robotic and task practice interventions is the lack of a measure to monitor functional recovery that can be completed on a regular basis during the intervention and that truly reflects a change in function and not simply motor learning to that specific task. We recently provided initial kinematic and kinetic data during the performance of a bimanual dexterity task that may provide appropriate measures to track the recovery process.6 Is the total intensity or amount of practice more important than the difference in approach? As we acknowledge in the article, the outcomes that might be seen if patients received 30 hours of repetitive task practice (RTP) alone (without an additional 30 hours of robotic-assisted therapy) remain unknown and are of interest in ongoing studies. Given the current strong interest in task-based therapy, the possible influence of impairment-level interventions is important to remember. The Stroke Impact Scale (SIS) is a widely used instrument with wellestablished reliability and validity. Lin et al7 recently proposed standards for determining minimal clinically important change on SIS subscales, including hand function. They concluded that the mean change score of a group of patients with stroke on the hand function subscale should reach 17.8 points in order to be considered clinically imApril 2010
portant. In our study, the estimated change in mean scores of the combined therapy group exceeded this threshold value, both preintervention to postintervention and preintervention to follow-up; the estimated change in mean scores of the RTP-only group did not exceed this threshold value, even though these changes were statistically significant. Merians suggests that this difference may indicate “an essential role for impairment-based interventions.” The improvement in hand function following the combined RTP-Hand Mentor intervention is encouraging and does suggest that an impairmentbased robotic system such as the Hand Mentor may have an important role, primarily as an adjunct to functional task practice. The possibility of utilizing technology (ie, robotic or virtual reality) is appealing because of the consistent and controllable nature of these systems. An important unknown is whether the use of these systems provides a cost-effective alternative or adjunct to the current delivery of care model. Given that study participants were trained within 3 to 9 months poststroke, a concern during this time frame is how to separate changes in function resulting from the intervention from changes due to spontaneous recovery. In patients with subacute stroke, differentiating change in function resulting from an intervention from changes due to spontaneous recovery is challenging, as we note with respect to the issue of change in quality-of-life ratings over time. As Merians suggests, double-baseline testing, separated by a period of time, may help to control for this issue. The value of implementing additional assessments always must be balanced against possible increased respondent burden and willingness to continue participation. Rather than using doublebaseline testing for clinical measures, we collected data for
kinematic and kinetic measures of hand function to complement data for clinical measures and provide insight into changes in motor control patterns that may be responsible for improvements in hand function. Nevertheless, we appreciate Merians’ point and recommend a combined approach of double-baseline testing and the utilization of quantitative measures of function that are not necessarily part of the training and are precise enough in nature to detect subtle changes in motor functioning. Clinicians would like to be able to discriminate between recovery based on compensation and recovery based on true improvement in motor abilities of the hemiparetic limb. Discriminating between recovery based on compensation and recovery based on true improvement in motor abilities of the hemiparetic limb is another research challenge. Quality-of-life measures provide patient perceptions of intervention effects and can be especially useful for judging the acceptability of a therapy. The National Institutes of Health Program Announcement under which our research was funded was titled Increasing Quality of Life in Mobility Disorders. However, we appreciate the difficulty of differentiating between compensatory mechanisms and true recovery. In our earlier work examining the effects of CIMT on the control of grasping forces and torques, we demonstrated that the motor improvements associated with CIMT appear to be related to actual improvements in patients’ ability to initiate and control grasping forces over time (ie, following CIMT, the rate of force production was similar to patterns others have reported in individuals who were healthy).8 More recently, for participants from this project, we have provided initial data suggesting that individuals in both groups exhibited a fundamental
Volume 90
Number 4
Physical Therapy f
507
Health-Related Quality of Life and Robotic-Assisted Therapy for Hand Function change in motor control strategy following each intervention from primarily a feedback control of grasping forces and torques to a greater reliance on feedforward control of grasping forces and torques during a functional bimanual dexterity task.6 Furthermore, participants were better able to focus their grasping forces (ie, produce smaller shear force) following these 2 interventions.6 These changes in specific kinematic and kinetic variables and motor control suggest that recovery was based on true improvement of the hemiparetic limb rather than a simple compensatory activity. To provide the reader with a context for changes in participants’ functional ability over the course of our study, mean scores on the FuglMeyer Motor Assessment (FMA) improved at postintervention by 7 to 8 points in both intervention groups, and this change was maintained at follow-up in both intervention groups. In a companion manuscript in process from this study in which the FMA is an outcome, our data are consistent with this sustained improvement in FMA at the 2-week follow-up evaluation. Development of robots to assist hand function is particularly challenging. Inconsistent use of the same outcome tests and measures and kinematic analyses hinders in-
508
f
Physical Therapy
Volume 90
terpretation of results and impedes progress of the development and incorporation of these technologies. We agree that development of robots to assist hand function is particularly challenging, which may explain why, to date, few studies have shown meaningful gain in dexterous function following a robotic intervention. The multiple degrees of freedom and extensive number of intrinsic and extrinsic muscles engaged, coupled with the vast amount of sensory information gathered, make the hand part of an exquisite control system. Developing a robotic system that takes all of these variables into account would be daunting and most likely cost-prohibitive. The importance of hand function in daily life is universally appreciated in the rehabilitation and bioengineering communities. The development of effective robotic systems for the hand requires a multidisciplinary approach between these 2 communities. We are encouraged to see more of these types of collaborations throughout North America and Europe. Such efforts will yield effective and practical solutions to improving hand function. As Merians suggests, it is incumbent on clinicians and bioengineers to utilize consistent outcomes across studies. The selection of outcomes should not be based on the “historical value” of a test per se. Outcomes, whether clinical or biomechanical, must be reliable, pre-
Number 4
cise, and quantitative in order for robotic or other technology-driven approaches to be translated into patient care. DOI: 10.2522/ptj.20090160.ar
References 1 Kutner NG, Zhang R, Butler AJ, et al. Quality-of-life change associated with robotic-assisted therapy to improve hand motor function in patients with subacute stroke: a randomized clinical trial. Phys Ther. 2010;90:493–504. 2 Merians A. Invited commentary on “Qualityof-life change associated with roboticassisted therapy to improve hand motor function in patients with subacute stroke: a randomized clinical trial.” Phys Ther. 2010; 90:504 –506. 3 Wolf SL. Revisiting constraint-induced movement therapy: are we too smitten with the mitten? Is all nonuse “learned”? And other quandaries. Phys Ther. 2007;87: 1212–1223. 4 Wolf SL, Winstein CJ, Miller JP, et al. Looking in the rear view mirror when conversing with back seat drivers: the EXCITE trial revisited. Neurorehabil Neural Repair. 2007;21:379 –387. 5 Wolf SL, Maddy D, Newton H, et al. The EXCITE trial: relationship of intensity of constraint induced movement therapy to improvement in the Wolf Motor Function Test. Restor Neurol Neurosci. 2007;25: 549 –562. 6 Alberts JL, Wolf SL. The use of kinetics as a marker for manual dexterity after stroke and stroke recovery. Top Stroke Rehabil. 2009;16:223–236. 7 Lin KC, Fu T, Wu CY, et al. Minimal detectable change and clinically important difference of the Stroke Impact Scale in stroke patients. Neurorehabil Neural Repair. 2010 January 6 [Epub ahead of print]. 8 Alberts JL, Butler AJ, Wolf SL. The effects of constraint-induced therapy on precision grip: a preliminary study. Neurorehabil Neural Repair. 2004;18:250 –258.
April 2010
Research Report Effects of a 6-Week, Individualized, Supervised Exercise Program for People With Bleeding Disorders and Hemophilic Arthritis
R. Mulvany, PT, DPT, is Associate Professor, Department of Physical Therapy, College of Allied Health Sciences, University of Tennessee Health Science Center, 930 Madison Ave, 6th Floor, Memphis, TN 38163 (USA). Address all correspondence to Dr Mulvany at:
[email protected].
Ruth Mulvany, Audrey R. Zucker-Levin, Michael Jeng, Catherine Joyce, Janet Tuller, Jonathan M. Rose, Marion Dugdale
Background. People with bleeding disorders may develop severe arthritis due to joint hemorrhages. Exercise is recommended for people with bleeding disorders, but guidelines are vague and few studies document efficacy. In this study, 65% of people with bleeding disorders surveyed reported participating in minimal exercise, and 50% indicated a fear of exercise-induced bleeding, pain, or physical impairment.
A.R. Zucker-Levin, PT, PhD, MBA, GCS, is Associate Professor, Department of Physical Therapy, College of Allied Health Sciences, University of Tennessee Health Science Center.
Objective. The purpose of this study was to examine the feasibility, safety, and
M. Jeng, MD, is Associate Professor of Pediatrics, Stanford University, Palo Alto, California.
efficacy of a professionally designed, individualized, supervised exercise program for people with bleeding disorders.
Design. A single-group, pretest-posttest clinical design was used. Methods. Thirty-three patients (3 female, 30 male; 7–57 years of age) with mild to severe bleeding disorders were enrolled in the study. Twelve patients had coexisting illnesses, including HIV/AIDS, hepatitis, diabetes, fibromyalgia, neurofibromatosis, osteopenia, osteogenesis imperfecta, or cancer. Pre- and post-program measures included upper- and lower-extremity strength (force-generating capacity), joint range of motion, joint and extremity circumference, and distance walked in 6 minutes. Each patient was prescribed a 6-week, twice-weekly, individualized, supervised exercise program. Twenty participants (61%) completed the program.
Results. Pre- and post-program data were analyzed by paired t tests for all participants who completed the program. No exercise-induced injuries, pain, edema, or bleeding episodes were reported. Significant improvements occurred in joint motion, strength, and distance walked in 6 minutes, with no change in joint circumference. The greatest gains were among the individuals with the most severe joint damage and coexisting illness. Limitations. Limitations included a small sample size with concomitant disease, which is common to the population, and a nonblinded examiner.
Conclusions. A professionally designed and supervised, individualized exercise program is feasible, safe, and beneficial for people with bleeding disorders, even in the presence of concomitant disease. A longitudinal study with a larger sample size, a blinded examiner, and a control group is needed to confirm the results.
C. Joyce, MSW, is affiliated with the Comprehensive Hemophilia Clinic, Department of Hematology, College of Medicine, University of Tennessee Health Science Center. J. Tuller, RN, MPH, is Clinical Nurse Coordinator, Comprehensive Hemophilia Clinic, Department of Hematology, College of Medicine, University of Tennessee Health Science Center. J.M. Rose, PT, MS, ATC, Assistant Professor, Department of Physical Therapy, Health Science Center, College of Allied Health Sciences, University of Tennessee. M. Dugdale, MD, is Medical Director, Comprehensive Hemophilia Clinic, Department of Hematology, College of Medicine, University of Tennessee Health Science Center. [Mulvany R, Zucker-Levin AR, Jeng M, et al. Effects of a 6-week, individualized, supervised exercise program for people with bleeding disorders and hemophilic arthritis. Phys Ther. 2010;90:509 –526.] © 2010 American Physical Therapy Association
Post a Rapid Response or find The Bottom Line: www.ptjournal.org April 2010
Volume 90
Number 4
Physical Therapy f
509
Supervised Exercise Program for People With Bleeding Disorders and Hemophilic Arthritis
B
leeding disorders and the musculoskeletal pathologies that accompany them offer numerous challenges to health care providers. Hemophilia A, hemophilia B, and von Willebrand disease are the most common inherited bleeding disorders. These disorders affect people of all ages, with no ethnic or racial predilection.1 Hemophilia A and B are caused by an x-linked inherited deficiency of clotting factors VIII and IX, respectively, and are found almost exclusively in males, whereas females are carriers of the trait.1–3 Von Willebrand disease is caused by an inherited defect in or deficiency
Available With This Article at ptjournal.apta.org • The Bottom Line clinical summary • The Bottom Line Podcast • Audio Abstracts Podcast This article was published ahead of print on March 4, 2010, at ptjournal.apta.org.
of von Willebrand factor and affects both sexes equally.4 Bleeding disorders can be classified as mild, moderate, or severe, depending on the amount or efficacy of circulating factor levels. In moderate or severe bleeding disorders, minor trauma can cause bleeding in joints (hemarthrosis) or in muscles (hematoma). This bleeding may initiate a cycle of musculoskeletal degeneration leading to disabling arthritis.5,6 Hemarthrosis is the most common and disabling manifestation of hemophilia. Approximately 80% of hemorrhages associated with hemophilia are hemarthroses.5 Hemarthroses usually begin around 12 to 24 months of age and persist throughout life.5 Joints most frequently affected are knees, followed by elbows and ankles; less frequently, hips and shoulders.5–7 Any joint that has 3 or more bleeding episodes over a period of 3 to 6 months is a target joint and is much more susceptible to subsequent bleeding and arthritic changes.8,9
Repeated bleeding into joints causes hemophilic arthritis (hemarthropathy), producing joint tissue destruction similar to that seen in rheumatoid arthritis. This destruction results in invasive hypervascular synovial hypertrophy, chronic synovitis, articular cartilage damage, bony hypertrophy, and subchondral cysts.5–10 Hemophilic arthritis is manifested by pain,11 joint instability, malalignment, muscular atrophy, impaired range of motion (ROM), and impaired function5–7,11–13 (Fig. 1). Arnold and Hilgartner14 described 5 radiographic stages in the progression of hemophilic arthropathy (Tab. 1). In stage I, joint integrity is maintained with no skeletal or cartilaginous changes. In stage V, the articular cartilage is fibrotic and deteriorated with a complete loss of joint space.14 Many joints are ankylosed at this stage, leaving joint replacement the only viable option for functional movement (Fig. 2). In addition to hemarthroses, bleeding may occur directly into a muscle. The most commonly affected muscles are the iliopsoas, quadriceps,
Figure 1. Severe chronic synovitis in hemophilia, with invasive hypervascular synovial hypertrophy. Joint arthropathy led to total knee replacement in this young man.
510
f
Physical Therapy
Volume 90
Number 4
April 2010
Supervised Exercise Program for People With Bleeding Disorders and Hemophilic Arthritis Table 1. Arnold and Hilgartner14 Radiographic Staging of Hemophilic Arthropathy and Clinical Presentation
Stage
Joint Integrity
Skeleton
Cartilage
Synovium
Clinical Presentation: Progression Along the Continuum
I
No change
No change
No change
Swelling of synovium and soft tissues
Acute hemarthrosis: as blood fills joint capsule, joint becomes tense, swollen, hard, hot, and tender; often held in flexion, with restricted range of motion and pain
II
No change
Osteoporosis, especially of epiphyses; epiphyseal overgrowth
No change
Swollen, thickened, boggy; early reaction similar to rheumatoid arthritis
Subacute hemarthrosis: after 2 or more hemarthroses; thickened and boggy synovium, moderate restriction of range of motion; pigmented villonodular synovitis, similar to rheumatoid arthritis
III
Disorganization of joint
Osteoporosis, subchondral cysts, progressive overgrowth of epiphyses, widening of intercondylar notch of knee and trochlear notch of ulna
No significant narrowing of joint space; squaring of patella
Opacified, with hemosiderin deposits; synovial hypertrophy and vascular hyperplasia
Chronic hemarthrosis: after subacute joint involvement has been present for ⱖ6 mo
IV
Advanced disorganization, irreversible joint changes
Progression of stages II and III
Severe cartilage destruction, narrowed joint space, osteochondral lesion, fibrillation and erosion, irreversible changes
Opacified and fibrous
Destructive progression to end stage
V
Marked fibrosis, substantial disorganization of structures, irreversible changes
Extensive enlargement of epiphyses, enburnated bone ends
Loss of joint space, absence of cartilage
Little or no recognizable synovial tissue
Chronic, fibrotic, contracted; totally destroyed joint
gastrocnemius-soleus, and muscles of the forearm. Bleeding into muscles can cause severe problems, including compartment syndrome, neurovascular compromise, fibrosis, adhesions, contractures, hematomas, and pseudotumors.15–18 (Fig. 3). Less common sites for bleeding are the gastrointestinal tract, vital organs, spine, and within the cranium. Although bleeding at these sites is less common, it poses a much greater risk and requires immediate medical intervention.19,20
April 2010
Overview of Management for Bleeding Disorders Currently, there are 3 major methods of managing hemophilia: medications to promote clotting and to manage pain and coexisting diseases, surgery, and rehabilitative exercise. Medications to Promote Clotting Depending on severity, medical management options for hemophilia include infusion of concentrated purified factor replacement5,21,22 or a nasal spray to stimulate the release of stored factor into the bloodstream for clotting control.21 Likewise, man-
agement of von Willebrand disease depends on the type and severity, with medical options including tablets, liquids, nasal sprays, and factor concentrate infusions.21,23 Historically, the management of hemophilia and its secondary musculoskeletal impairments was greatly complicated by factor replacements derived from contaminated blood products, leading to the spread of HIV, hepatitis, and other blood-borne pathogens.24 Currently, the risk of secondary infection is diminished by the availability of purified factor replacement. Some pa-
Volume 90
Number 4
Physical Therapy f
511
Supervised Exercise Program for People With Bleeding Disorders and Hemophilic Arthritis
Figure 2. Bilateral stage IV and V elbows and knees due to multiple hemarthroses.
tients will use factor replacement as needed (episodic), whereas others receive regularly scheduled replacement (prophylactic). Improved safety and prophylactic use have resulted in less joint destruction and the ability to lead a more active lifestyle.22,25 Surgery When indicated, orthopedic surgical interventions for hemarthropathy are similar to those for rheumatoid arthritis and osteoarthritis, with the added complications of bleeding and coexisting diseases acquired from blood products.26 –32 Some of the most common orthopedic procedures are synovectomy, arthroscopy, total joint arthroplasty, and arthrodesis.26,28 –30,32,33 To ensure an optimal outcome, the medical team should involve the primary care doctor, surgeon, anesthetist, hematologist, nurse, physical therapist, occupational therapist, and social worker. Because of the complexities of rehabilitation, physical therapy should be initiated before and continued well after surgery.27,28,31,33 512
f
Physical Therapy
Volume 90
Rehabilitation and Exercise A major component of rehabilitation for people with hemarthropathy and muscle bleeds is therapeutic exercise.27,33–36 Therapeutic exercise is the selective application of stress to cause beneficial physiologic adaptation and to restore function. Because exercise is a form of biomechanical stress, a delicate balance exists between too much stress, which may cause bleeding and trauma, and not enough stress, producing a subtherapeutic response. This fine line between beneficial and detrimental activity has led many people with bleeding disorders to refrain from exercise for fear of initiating a bleeding episode.34,36 Paradoxically, refraining from exercise leads people with bleeding disorders to experience decreased function due to weakness, decreased ROM, and diminished quality of life. When surveyed by the National Hemophilia Foundation’s National Prevention Program, 60% of the adolescents with hemophilia reported that they limit or refrain from physical activity.37 The survey supported the findings that children with hemophilia
Number 4
generally have decreased strength (force-generating capacity) and flexibility, are less active, and have lower aerobic working capacity than their unaffected peers.33,34,36 – 43 These inadequate levels of strength and flexibility leave people with bleeding disorders even more vulnerable because strong, flexible muscles support joints, help attenuate stresses, and diminish the risk of injury.37 Thus, people with bleeding disorders who exercise appropriately may improve their strength and flexibility, which, in turn, could diminish the chances of developing recurrent hemarthrosis, synovitis, and subsequent joint destruction.34 – 43 Furthermore, because obesity and heart disease are as prevalent in people with bleeding disorders as they are in the general population, cardiovascular fitness should be included in any exercise program.44,45 Literature that examines the role of exercise on function in people with bleeding disorders is limited. Harris and Boggio46 performed a descripApril 2010
Supervised Exercise Program for People With Bleeding Disorders and Hemophilic Arthritis
Figure 3. Massive pseudotumor and compartment syndrome due to repeated hemorrhages in the gastrocnemius muscle.
tive review of 46 people with bleeding disorders and found that those who reported regular exercise had significantly greater ROM in the large joints than those who did not exercise. Likewise, Wittmeier and Mulder43,47 described the benefits of exercise and hypothesized that people with bleeding disorders would benefit from sports, fitness, and physical activity. In addition to these reports, experienced therapists have provided guidelines for exercise safety in people with bleeding disorders in publications by the World Federation of Hemophilia47,48 and the National Hemophilia Foundation.49 However, experimental design was not used when establishing these guidelines. Few studies that used experimental designs have been reported. Hilberg et al50 performed an experimental April 2010
exercise protocol on 9 people with bleeding disorders who participated in a 6-month specialized training program including gentle strength training with low resistance. Significant improvements in isometric muscle strength and proprioceptive performance were found, and generic guidelines for strengthening in people with bleeding disorders were provided, which include the use of low-resistance, high-repetition exercises over a 6-month period. Likewise, Pelletier et al,51 in a quasiexperimental study, found strength gains after a 3-week isometric exercise program in a single 12-year-old participant. These studies provide limited guidance in decision making when prescribing exercises to people with bleeding disorders. The literature implies that people with bleeding disorders could bene-
fit from a structured, supervised exercise program. However, there is little scientific evidence on which to base specific exercise prescriptions for people with bleeding disorders, and continuous rehabilitation under a physical therapist may not always be economically feasible. Due to the dearth of evidence-based literature, therapists must rely on their best judgment for developing exercise prescriptions when treating people with bleeding disorders. Therefore, it was the goal of this study to determine whether people with bleeding disorders can safely and effectively exercise and achieve improved levels of function using an individualized exercise program that was designed by a physical therapist and supervised by a trained fitness instructor. In contrast to the studies by Hilberg et al50 and Pelletier et al,51 our study offers a comprehensive ex-
Volume 90
Number 4
Physical Therapy f
513
Supervised Exercise Program for People With Bleeding Disorders and Hemophilic Arthritis ercise program that challenges strength, ROM, and cardiovascular function in patients with severe comorbidity.
Materials and Method Study Design A single-group, pretest-posttest clinical design was used, with measurements taken prior to initiation of the supervised exercise program and immediately following the 6-week intervention. Participants Participant selection for this Institutional Review Board–approved study was based on a sample of convenience from the patient population of the Comprehensive Hemophilia Clinic of the University of Tennessee Health Science Center. Thirty-three volunteers, (30 male, 3 female), met the inclusion criteria. All participants signed informed consent and liability release forms. Additional consent forms were signed by legally authorized representatives for those volunteers under the age of 18 years. None of the participants had participated in an exercise program for a 1 year prior to initiating the program. Our population represents a typical cohort of people with bleeding disorders. The study group had a range of severity of hemophilic arthritis: some had previous orthopedic surgery, some had a coexisting illness such as HIV or hepatitis, and some were on a regimen of prophylactic factor replacement. Inclusion criteria were: (1) diagnosis of mild, moderate, or severe hemophilia or von Willebrand disease; (2) willingness to exercise twice a week for 6 weeks and to complete the preand post-program evaluations; (3) ability to arrange transportation to and from data collection and exercise sessions; (4) approval by their hematologist to participate in the exercise program; (5) aged 7 to 60 years; and (6) agreement of those 514
f
Physical Therapy
Volume 90
participants with severe hemophilia to receive a prophylactic intravenous dose of factor prior to exercise. Exclusion criteria were: (1) the inability to attend exercise sessions at least twice a week for 6 consecutive weeks; (2) nonadherence to instruction on proper exercise technique; (3) surgical procedures performed 6 weeks prior to or during the exercise program; (4) participation in any other form of exercise, including rehabilitation, during the study; (5) changes in medication during the study; and (6) a major bleeding episode that posed a risk or prevented exercise. Outcome Measures At the pre-program session, one physical therapist (R.M.) with 20 years of experience in treating people with bleeding disorders was designated to complete the evaluation and prescribe an exercise regimen for each participant based on individual needs. The same therapist performed the post-program evaluation, following the same procedures as in the pre-program evaluation, to ensure consistency in testing and exercise prescription. Pre- and post-program data included in the musculoskeletal evaluation were functional walking, ROM, muscle strength, and circumferential measurements. Functional walking. The SixMinute Walk Test (6MWT) is used to measure walking ability and baseline cardiovascular function for people with disease or low levels of fitness.45 Due to pre-existing comorbidities in many of our participants, we determined that the 6MWT was the most appropriate test to measure cardiovascular function. Participants walked an 800-ft,* unobstructed, rectangular pathway following the * 1 ft⫽0.3048 m.
Number 4
guidelines of the American Thoracic Society.52 To ensure safety and to measure the exact distance walked in 6 minutes, the physical therapist followed closely with a stopwatch and Rollatape meter.† The Rollatape meter measures distances in 1-ft increments. The 6MWT is a submaximal, quantitative evaluation of functional exercise capacity and is reflective of ability to perform daily physical activities.52 A significant correlation (r⫽ .73) between the 6MWT and peak oxygen uptake has been reported for patients with end-stage lung diseases. The short-term reproducibility of 6MWT is excellent.52 ROM. Joint passive ROM of the knees, hips, ankles, and elbows was measured with a universal goniometer according to the method described by Norkin and White.53 Norkin and White reported the reliability of goniometric measurement as good to excellent. In addition to having 20 years of experience with our population, the measuring therapist helped develop the protocol and train researchers for goniometric measurement for people with bleeding disorders for the Universal Data Collection Study of the Centers for Disease Control and Prevention.54 Thus, her measurements are considered highly consistent from patient to patient and test to test for people with bleeding disorders. Muscle strength. Isometric muscle strength of bilateral hip extension, flexion, and abduction; knee flexion and extension; and elbow flexion and extension were measured with a Nicholas handheld dynamometer‡ while the patient was placed in standard testing positions, as described by Hislop and Montgomery.55 The Nicholas dynamome† Rollatape, 255 W Fleming, Watseka, IL 60970. ‡ Lafayette Instrument Co, 3700 Sagamore Pkwy N, Lafayette, IN 47402.
April 2010
Supervised Exercise Program for People With Bleeding Disorders and Hemophilic Arthritis ter provides specific, numeric results that we believe are more valuable in quantifying changes in ability to resist force. Isometric muscle testing was used because movement with maximal resistance can be painful for many individuals who have limited pain-free range. Testing isotonic force production could have resulted in erroneous results, as pain could inhibit muscle function and individuals might avoid maximal contraction in the painful ROM. The Nicholas handheld dynamometer was used to enhance objectivity and consistency and to measure small increments of change that we predicted would be necessary for assessing the population. It has been proven valid and reliable for measuring isometric muscle force.56 –58 Circumferential measurements. Knees and elbows were measured at the joint line, as well as 6 in§ above and 4 in below the joint line, to determine whether joint or muscle swelling developed. Although we did not find a study using circumferential measurement on the lower extremity, Taylor et al59 reported that arm volume measured circumferentially was highly reliable. We felt circumferential measurement would be the most time-efficient method for our population when considering the number of measurements we were performing. To ensure consistency of measurement, the same tape measure and landmarks were used. Individualized Exercise Program The levels of intensity and guidelines for exercise progression used in this study were devised by the prescribing physical therapist for this unique setting and for these participants. The attempt was to provide a practical, flexible strategy for exercise that addressed the varying joint conditions from acute to chronic status §
1 in⫽2.54 cm.
April 2010
and from stage I to stage V arthropathy. Each participant’s individualized exercise program was prescribed by the evaluating physical therapist and included exercises for strength, flexibility, and cardiovascular function. Each program began conservatively, with progression of exercise intensity based on tolerance of the previously performed exercise. Participants were taught to selfmonitor heart rate and to report any adverse reactions such as dizziness, chest pain, muscle bleeds, joint bleeds, increased pain, swelling, or fatigue. All major muscle groups were exercised. However, some participants were not able to perform isotonic strengthening exercises on all joints due to either severe pain or joint ankylosis. Likewise, flexibility exercises were tailored to those joints that were not ankylosed, yet had limited motion. Strengthening. Three intensity levels were established based on the available ROM, strength, history of pain, hemarthroses, and perceived fragility of each major joint (Appendix 1). The structural integrity of each joint and the biomechanical stress of lifting weight also were considered. For example, maximal resistance with heavy weight was avoided to prevent excessive and possibly deleterious compressive forces. These factors plus the results of the muscle strength measure were considered by the therapist when establishing initial levels of exercise intensity. Although the dynamometry testing was isometric, exercise prescription was isotonic to promote active ROM, muscle endurance, joint nutrition, proprioception, and motor control that would not be as effectively gained with isometric exercise. Isotonic exercise at submaximal intensity was tolerable throughout the painful ranges. Level 1 intensity was for the most fragile joints, identified target joints,
or previously injured muscles.8,9 Strengthening exercise at level 1 was approximately 40% of the result determined by the isometric Nicholas dynamometry muscle test with 1 set of 10 repetitions performed in the pain-free range. Level 2 intensity was prescribed for joints and muscles that had a history of bleeding and demonstrated moderate hemarthropathy but had no history of bleeding in the previous 6 months. Strengthening exercise at level 2 was approximately 50% of the result determined by the isometric Nicholas dynamometry muscle test with 1 set of 10 repetitions in the pain-free range. Level 3 intensity was prescribed for joints and muscles that demonstrated minimal or no signs of impairment. Level 3 intensity was approximately 60% of the result determined by the isometric Nicholas dynamometry muscle test with 1 set of 10 to 20 repetitions in the pain-free range. If no increased pain or swelling occurred, the intensity could be increased by 5% to 10% per week up to a maximum of 75% of the isometric Nicholas dynamometry muscle test. A set of 10 to 20 repetitions was added in the second week, and a third set was added in the following week. By the end of the third week, a participant who had no adverse effects could be performing 3 sets of 20 repetitions with up to 75% intensity. Any increase in pain or swelling required ceasing exercise for that structure and consulting with the physical therapist. Options for strengthening exercise included free weights, stationary resistance equipment, Thera-Band exercise bands,㛳 and functional strengthening activities. Although the resistance of Thera-Band exercise bands cannot 㛳
The Hygenic Corporation, 1245 Home Ave, Akron, OH 44310.
Volume 90
Number 4
Physical Therapy f
515
Supervised Exercise Program for People With Bleeding Disorders and Hemophilic Arthritis be quantified, some participants were not able to safely exercise with free weights or stationary resistance equipment due to pain, limited ROM, or muscle imbalance. Participants who used Thera-Band exercise bands exercised in standard positions and were progressed through the color hierarchy (Appendix 1). Functional exercise included walking, sit-to-stand, and stair-climbing tasks, with intensity graded based on required use of upper-extremity support. Level 1 intensity corresponded to using 2 hands, level 2 intensity corresponded to using 1 hand, and level 3 intensity required no hands to perform the activity. If a participant progressed past level 3, exercise progression was advanced in 1 of 2 ways: either increasing the number of repetitions of the functional activity or performing the activity with weights. Plyometrics were avoided as the high impact, torsion, and loading of joints were deemed too intense and potentially damaging to joint structures. Flexibility and ROM. An individualized exercise program including soft tissue flexibility and joint ROM was prescribed based on limitations found during the evaluation. Due to the fibrous nature of joint and muscle limitations in people with bleeding disorders, stretching was performed after warming the tissue with active exercise. Gentle, lowload, prolonged stretch guidelines within the pain-free range were used.60 Each stretch was held for a minimum of 2 minutes to a maximum of 20 minutes. For efficiency with prolonged stretching longer than 2 minutes, a body part would be positioned and externally stabilized with weights or straps in a stretched position while other body parts were exercised. Cardiovascular exercise. Each participant’s cardiovascular program began conservatively and was ad516
f
Physical Therapy
Volume 90
justed according to any precautions identified by the physician or any difficulties reported by the participant. The participant’s resting and post-walking heart rate and respiratory rate from the 6MWT were considered in the development of the cardiovascular workout. Because all participants were unaccustomed to cardiovascular exercise and musculoskeletal impairments prevented the normal stress testing process, we used the American College of Sports Medicine’s formula for estimating maximal heart rate (estimated HRmax⫽220 ⫺ age).45 The sessions began at 50% of maximum heart rate for the first 2 sessions and progressed by 5% to 10% increments up to a maximum of 70% if no adverse effects were reported. A final consideration was the ability of the participant to use equipment. For example, none of the participants had enough ROM in the lower extremities to exercise on a stationary bicycle. Each participant chose one cardiovascular activity from the following list: hydrotrack (aquatic treadmill), therapeutic pool, landbased treadmill, upper-extremity ergometer, cross-country skier, and low-impact aerobic exercise. For participants with severely limited ROM, aquatic exercise was most effective. Twenty minutes of cardiovascular exercise was the goal. Following the evaluation and individualized exercise prescription, each participant was scheduled to begin a twice-weekly supervised exercise program with a minimum of 2 days of rest between each session. Participants were not provided with a home exercise program but were instructed to continue with their normal daily activities. Participants with severe hemophilia were instructed to infuse with factor no later than 2 hours prior to each exercise session. Participants with mild or moderate hemophilia or von Willebrand disease were instructed to bring their
Number 4
appropriate coagulation medication to all exercise sessions. In order to ensure safety, consistency, and proper technique, each participant’s exercise session was supervised by a fitness instructor who possessed a master’s degree in fitness and exercise. Every effort was made to educate the fitness instructor to ensure safety of all participants. She reviewed the publications on bleeding disorders from the National Hemophilia Foundation,61,62 received training from the physical therapist, and indicated she understood the tenuous nature of the pathology and the importance of adherence to the prescribed exercise program. She guided the participant through the program on a twice-weekly schedule and recorded the exercise performed at each session. In the event of an adverse reaction, she was asked to document it and contact the physical therapist. The physical therapist consulted with the fitness instructor weekly to progress the exercise program. Participants who missed more than 2 exercise sessions during the 6-week time period were terminated from the study. At the end of 6 weeks, each participant was re-evaluated using the same testing procedures as for the pre-program evaluation. The physical therapist performed post-program evaluations and recorded the post-program data on blank evaluation forms without reviewing the pre-program data. Data Analysis Analyses were carried out on a personal computer using a spreadsheet (Microsoft Excel 2003#). Outcomes were evaluated with paired t tests. Data from both the left and right extremities were combined for analyses. This was done because people with bleeding disorders have drastically different patterns of joint destruction based on bleeding history of each individual joint. For exam# Microsoft Corp, One Microsoft Way, Redmond, WA 98052-6399.
April 2010
Supervised Exercise Program for People With Bleeding Disorders and Hemophilic Arthritis ple, a person with bleeding disorders may have one knee rated at stage V on the Arnold and Hilgartner scale, and the contralateral knee may show minimal or no joint pathology. This situation is very different from that of people with other forms of arthritis, who tend to have almost symmetrical destruction due to a systemic disease process.
Table 2. Demographic Characteristics of Study Participants (N⫽33) Characteristic
Enrolled in Study
Race Asian
1 16
6
White
16
13
30
18
3
2
Adult
20
12
Child
13
8
Sex
Female Group
Age (y) Adult X
40
40.5
SD
12.1
12
Range
19–57
26–27
14
15
Child X
Role of Funding Source This work was supported by Baxter Healthcare. Baxter Healthcare had no involvement in the design, conduct, or reporting of the study.
SD
3.2
3.9
Range
7–18
7–18
Diagnoses Hemophilia Mild
Results
Moderate
Participants Thirty-three participants volunteered for the study (Tab. 2). Twenty (61%) of the participants completed the program, attending a mean of 11.5 of the 12 scheduled sessions, and returned for post-program data collection. Thirteen participants (39%) did not complete the program due to transportation problems, illness, or scheduling difficulty. None of the 33 participants reported any adverse reactions from the exercise program.
Severe
Anthropometry ROM. Significant improvement in ROM was found in all joints when comparing pre- and post-program data (Tab. 3). A negative number indicates a lack of range; for example, the baseline range of knee extension was ⫺38 to 14 degrees, indicating April 2010
1
African American
Male
Effect sizes were calculated to account for group variability. The effect size indexes were calculated for all outcome measures. The calculations were performed by dividing the difference between the pre- and postprogram means by the standard deviation of the pre-program scores.63 An effect size of 0.20 or less represents a small change, 0.21 to 0.80 was considered a medium change, above 0.80 was considered a large change.
Completed Study
Von Willebrand disease Mild Moderate HIV positive
one participant lacked 38 degrees of knee extension and another participant had 14 degrees of knee hyperextension. Strength. Muscle strength was tested in 17 of 20 participants. Three participants were not tested at baseline assessment due to pain. These 3 participants participated in the entire exercise protocol, including strength training as tolerated, but were not tested for strength at the final assessment. One additional participant was not tested for hip flex-
30
18
1
1
3
3
26
14
3
2
1
1
2
1
11
9
ion and extension at post-program due to low back pain. For the 17 participants analyzed, significant improvement was seen in all muscle groups tested when comparing preand post-program data (Tab. 4). Circumferential measures. Circumferential measurements were taken on all participants. Significant differences were seen in all joints and limb segments tested in the upper extremity when comparing preand post-program data. However, no
Volume 90
Number 4
Physical Therapy f
517
Supervised Exercise Program for People With Bleeding Disorders and Hemophilic Arthritis Table 3. Range of Motion (in Degrees) and Effect Sizea Baseline Measurement Variable Knee flexionb Knee extension
c
Ankle dorsiflexionb Ankle plantar flexionb Elbow flexionc
X
SD
112.0
32.0
95% Confidence Interval
Final Measurement
Range 28 to 149
⫺7.1
10.7
⫺38 to 14
⫺2.6
10.3
⫺20 to 20
39.7
10.4
15 to 60
132.3
17.3
80 to 153
X
SD
118.4
30.8
Range 40 to 155
⫺2.1
6.9
⫺18 to 12
1.2
10.4
⫺20 to 26
47.0
9.5
137.2
14.9
d
102.1 to 121.9
Effect Size 0.2
⫺10.4 to ⫺3.8
0.46
⫺5.8 to 0.6
0.36
30 to 70
36.5 to 42.9
0.7
80 to 155
126.9 to 137.7
0.28
d
b
⫺12.4
20.9
⫺87 to 20
⫺8.5
17.9
⫺75 to 14
⫺18.9 to ⫺5.9
0.19
Elbow pronationc
59.0
23.0
5 to 85
64.5
20.1
20 to 92
51.9 to 66.1
0.24
Elbow supinationc
60.3
29.5
⫺30 to 90
65.3
26.0
⫺20 to 90
51.2 to 69.4
0.17
Elbow extension
d
d
a
N⫽40 accounting for bilateral limbs of 20 participants. A negative number indicates a lack of range of motion; for example, the baseline range of knee extension was ⫺38 to 14 degrees, indicating one participant lacked 38 degrees of knee extension and another participant had 14 degrees of knee hyperextension. b Pⱕ.01. c Pⱕ.05. d Hyperextension.
differences were found in the lower extremities (Tab. 5). Functional walking. The 6MWT was performed by 19 participants for pre- and post-program evaluations. Baseline and final assessments were not performed on one participant because of severe pain when walking. A significant (P⬍.01) improvement with a large effect size (0.90) was seen when comparing baseline distance walked (X⫽1,145 ft, SD⫽ 318, range⫽376 –1,617) with final distance walked (X⫽1,431 ft, SD⫽ 471, range⫽471–2,297).
Discussion
lenging comorbidities, including HIV/ AIDS, hepatitis, hypertension, cancer, diabetes, and osteogenesis imperfecta. Of the 20 participants who completed the study, 14 had at least 1 joint with severe stage IV or V arthropathy, and 1 participant had 8 stage V joints.
People with bleeding disorders may have a wide range of orthopedic and psychological sequellae of the disease, including weakness, joint destruction, pain, and fear, that may limit participation in fitness programs. Safety in exercise is a foremost concern for this population, with scarce guidance found in the literature. Our participants represented a typical cross-section of patients followed in a hemophilia clinic serving a large metropolitan area. In addition to issues related to bleeding disorders, our population had a variety of chal-
Due to the severity of joint destruction, our findings of statistically significant gains in ROM in all joints tested were unexpected. The mean arc of ROM increased as follows: ankle dorsiflexion and plantar flexion⫽11 degrees, knee flexion and
Table 4. Muscle Strength (in Newtons) and Effect Sizea
X
SD
Range
X
SD
Range
95% Confidence Interval
15.6
4.4
0.0 to 26.6
22.9
6.1
4.3 to 35.2
14.1 to 17.1
1.66
18.1
6.3
6.3 to 32.6
22.4
6.6
9.5 to 36.3
15.9 to 20.3
0.68
16.5
7.9
6.1 to 29.1
21.2
9.4
11.3 to 38.0
13.9 to 19.2
0.59
12.8
4.8
0.8 to 27.2
16.3
5.7
4.6 to 29.0
11.2 to 14.4
0.73
15.0
7.8
1.0 to 29.7
20.5
9.9
3.8 to 40.6
12.4 to 17.6
0.71
14.4
7.4
0.0 to 30.1
15.0
7.8
0.8 to 31.6
11.9 to 16.7
0.08
11.7
8.3
0.0 to 24.8
14.2
8.3
0.4 to 28.8
8.9 to 14.5
0.3
Baseline Measurements Variable Hip extensionb,c Hip flexion
b,c
Hip abductionc Knee flexion
c
Knee extensionc Elbow flexion
c
Elbow extensionc a b c
Final Measurements
Effect Size
N⫽34 accounting for bilateral limbs of 17 participants; 3 participants were not tested before exercise due to pain. n⫽32; at the post-program evaluation, one participant was experiencing low back pain, thus hip extension and flexion were not tested. Pⱕ.01.
518
f
Physical Therapy
Volume 90
Number 4
April 2010
Supervised Exercise Program for People With Bleeding Disorders and Hemophilic Arthritis Table 5. Circumferential Measurements (in Centimeters) and Effect Sizea Baseline Measurements Variable
95% Confidence Interval
Final Measurements
X
SD
Range
X
SD
Range
45.2
7.1
32.0 to 66.5
47.2
8.0
33.0 to 69.0
43 to 47.4
0.28
Knee at joint line
37.8
3.3
31.0 to 46.0
38.1
3.0
32.5 to 44.0
36.8 to 38.8
0.09
Knee 4 in below joint line
35.1
4.2
28.0 to 45.0
35.5
4.2
28.0 to 45.0
33.8 to 36.4
0.10
Knee 6 in above joint line
Elbow 6 in above joint line
b
Effect Size
29.8
5.3
19.5 to 40.0
30.4
5.7
19.8 to 41.0
28.2 to 31.4
Elbow at joint lineb
28.0
3.3
21.5 to 33.0
27.6
3.2
22.0 to 32.0
27 to 29
⫺0.12
0.11
Elbow 4 in below joint linec
25.6
3.5
19.0 to 33.5
26.3
3.8
19.5 to 31.5
24.5 to 26.7
0.2
a
N⫽40 accounting for bilateral limbs of 20 participants. b Pⱖ.05. c Pⱕ.01.
extension⫽11.4 degrees, elbow flexion and extension⫽8.8 degrees, and pronation and supination⫽10.5 degrees. The overall change in mean arc of ROM may not appear large. However, when we realized that some individuals had no improvement due to ankylosis, the significant findings can be attributed to large gains in some joints. For example, in elbow flexion, 17 of 40 joints showed 3 degrees or less of improvement in ROM. Three degrees may be a function of instrument or examiner error. Therefore, the significant findings were due to large ROM gains in 23 of the 40 joints. Our findings are consistent with the report of Harris and Boggio,46 who described the ROM of large joints in adults with hemophilic arthritis. They determined that individuals who participated in thrice-weekly exercise had significantly (P⫽.003) better ROM than those who had not participated in an exercise program. Our findings indicate that initiation of an exercise program in a group of patients with hemophilic arthritis who had not previously participated in a regular exercise program can significantly improve joint ROM. Clinical relevance can be found in the increase of function that accompanies increased ROM. For example, increased ROM of the elbow can promote the ability of individuals to reach their head, April 2010
neck, and face, allowing them to wash their face and hair, shave, apply makeup, brush and floss teeth, feed themselves, button collars, and don and doff jewelry or ties. Strength significantly improved for all muscle groups. Again, this finding was unexpected for the following reasons. In addition to severe arthropathy, 9 (45%) of the participants in our study were HIV positive or had AIDS, predisposing them to sarcopenia.64 The prescribed exercise program was intentionally conservative to avoid biomechanical stress on joints, thus diminishing the expected strength gains. Strength gains were not expected in muscles surrounding ankylosed joints, which are prevalent in this population. For example, 13 of the 34 joints tested for elbow flexion showed gains of 2 N or less, indicating the significant change in strength was due to large gains in the remaining 21 joints. These findings are in concert with those of Hilberg et al,50 who tested proprioceptive performance and isometric muscle strength in 9 participants with hemophilia who took part in a 6-month specialized training program. The specialized training program included gentle strength training with low resistance performed for 20 to
25 repetitions. The 9 participants showed significant (P⬍.05) improvement in maximal isometric leg muscle strength, as measured by leg press. This low-intensity, highrepetition program was performed to apply minimal stress to the lowerextremity joints. Likewise, Pelletier et al51 tested the effect of a 3-week isometric exercise program on a single 12-year-old participant with severe factor VIII deficiency and chronic knee arthrosis. Their intervention produced increased strength in the right hamstring and quadriceps muscle without adverse effect. When comparing our study to those of Hilberg et al50 and Pelletier et al,51 we find similarities in strength gain with more intensive isotonic exercise. Our participants were prescribed an exercise program based on their joint integrity, pain, bleeding history, strength, and available ROM. We allowed participants to progressively increase intensity of exercise up to 75% of the preprogram strength measures. The clinical relevance of this finding is that it indicates a more intensive isotonic strengthening program can be both safe and effective. Circumferential measurements were taken at the joint line of the knee and elbow to determine whether exercise increased joint swelling. Addi-
Volume 90
Number 4
Physical Therapy f
519
Supervised Exercise Program for People With Bleeding Disorders and Hemophilic Arthritis tional girth measurements were taken at specific distances above and below the knee and elbow to serve as a baseline in the event of a muscle bleed. We found a small decrease in the circumferential measure at the elbow joint line and a small, but significant, increase in the upperextremity girth measurements. Lowerextremity girth measurements did not show significant change. This result, in combination with clinical examination and participant selfreport, implies that the exercise program did not induce major muscle or joint bleeding. To elaborate on the safety of our program, at the conclusion of the study, several participants with severe hemophilia reported that their adherence to factor infusion prior to exercise was not 100%; yet no bleeding episodes were reported resulting from exercise. These findings agree with those of Pelletier et al,51 who did not find adverse effects of exercise on circumferential measures. Due to the low level of intensity and short duration of our program, it is unlikely that the small increases were due to muscle hypertrophy or significant physiologic adaptation. These small increases may have been due to measurement error. We believe that the improved performance in strength measures was likely the result of neuromotor adaptation60 and that hypertrophy might have been more evident with a longer period and higher intensity of exercise. However, hypertrophy is of special interest due to the sarcopenia related to HIV status in several participants. As previously stated, each limb was examined as independent observations with up to 40 observations made on 20 participants (ROM). This was done because typically people with bleeding disorders have asymmetry in joint destruction. To address any concerns about the validity of pairing data collected from both the right and left extremities, single520
f
Physical Therapy
Volume 90
limb analyses also were performed. Similar results were found for ROM, strength, and circumferential measures when comparing single and paired data65,66 (Appendix 2). Finally, we found significant gains in distance walked on the 6MWT. This gain was important for our participants because it reflects an improvement in functional exercise level such as walking, activities of daily living, and self-care.52 Several factors may have contributed to this functional improvement: increased stride length from improved ROM; improved muscular endurance; improved cardiopulmonary efficiency; improved circulation; and improved biomechanical loading on the joints from gains in ROM and muscle strength, resulting in a more comfortable and efficient gait.52 The improvement in functional walking also could result from behavioral and psychological factors such as increased confidence, improved body image, and decreased fear of movement or injury.
scribed conservative strengthening regimen may have failed to challenge participants with higher functioning. One physical therapist performed all evaluations, developed individualized exercise programs, and modified the programs, which may have led to examiner bias. To gather as much information from the study as possible, many measurements were collected and many comparisons were made, which may have resulted in an inflation of the type I error rate. Finally, some participants had pain, which precluded full participation for evaluation and intervention. Additional minor limitations include confounding variables imposed upon the health of the participants due to concomitant disease. Long-term follow-up was not formally performed; thus maintenance of gains were not assessed. Due to these limitations and limited sample size of 20 without a control group, further study is needed to confirm these results in people with bleeding disorders.
Conclusions Limitations Our participants represented a typical population with bleeding disorders managed at a metropolitan hemophilia clinic with a wide range of ages (7–57 years) and functional abilities ranging from independent to severely limited. This variability allows us to generalize our results, but at the same time it creates a number of limitations in applying the results that should be recognized. We believe this trade-off was acceptable because had stricter inclusion and exclusion criteria been imposed, the cohort of participants would have been diminished. Limitations of our study included nonrandomization of participants who served as their own controls. Safety issues were the highest concern; therefore, the independent variable (exercise) was not one specific protocol that could be applied to all participants, and the pre-
Number 4
Hemophilia and bleeding disorders are rare and may result in lifelong, chronic musculoskeletal problems and functional limitations. Physical therapists are challenged with developing interventions to protect people with bleeding disorders from episodes of bleeding and joint destruction but also to find ways to promote physical function and independence. Although exercise is recommended for recovery from bleeding episodes, as well as for health, fitness, and enhanced quality of life, there are few reported evidencebased studies on the benefits and safety of exercise in people with bleeding disorders. Therefore, the purposes of this study were to investigate the efficacy, safety, and feasibility of an individually designed exercise program for people with bleeding disorders and to initiate the development of evidence-based April 2010
Supervised Exercise Program for People With Bleeding Disorders and Hemophilic Arthritis guidelines for exercise in people with bleeding disorders. It is our recommendation that people with bleeding disorders should: • have permission from their physician to exercise, • have a thorough physical therapist evaluation and individualized exercise prescription prior to initiating exercise, • have coagulation medications available for mild and moderate bleeding disorders, • use prophylactic factor replacement for severe bleeding disorders, • start with a very conservative baseline exercise program that accommodates weakened joints and muscles, • exercise within their pain-free range, • avoid equipment or activities that apply biomechanical stress in the form of impact or torsion, • expect a longer period of time to achieve goals, and • be supervised by a fitness instructor who has been educated by and works closely with a physical therapist.
Following these guidelines, our participants improved, with no adverse events. Ten of the 20 participants requested to repeat the 6-week exercise program. The results of this study suggest that the combined efforts of physical therapists and trained fitness instructors to provide individually prescribed and supervised exercise programs for people with bleeding disorders can succeed. Consistent with the goals of Healthy People 2010,67 physical therapists can play an important role in the promotion of health and the development of exercise programs for underserved populations and can reduce health care disparities for individuals with disabilities. Our program proved effective, safe, and feasible for our population of people with bleeding disorders receiving care at a metropolitan hemophilia April 2010
clinic to improve health, fitness, and well-being through exercise. Improvement of function through exercise in people with bleeding disorders is understudied. Further study is needed to determine the benefits of exercise on variables such as gait parameters and pain and to promote the development of evidence-based exercise guidelines. Dividing the participants into subgroups based on severity of hemophilia to determine which patients would gain the most from an exercise program also should be explored. Dr Mulvany, Dr Zucker-Levin, Dr Jeng, and Dr Dugdale provided concept/idea/research design and writing. Dr Mulvany, Ms Joyce, Ms Tuller, and Mr Rose provided data collection. Dr Mulvany, Dr Zucker-Levin, and Dr Jeng provided data analysis, project management, and clerical support. Dr Mulvany, Ms Tuller, and Dr Dugdale provided fund procurement. Dr Mulvany, Dr Jeng, Ms Joyce, Ms Tuller, and Dr Dugdale provided participants. Mr Rose provided facilities/ equipment and institutional liaisons. All authors provided consultation (including review of manuscript before submission). The authors acknowledge the study participants for their determination and persistence; Angela Redden, fitness instructor; Penny Head, PT; and all student physical therapy researchers for their significant contributions, especially: Kellee Berry McBride, David Grigsby, April Webb, Ed Moyer, and Kristi Lott-Davidson. This project was approved by the University of Tennessee, Health Science Center, Institutional Review Board. Poster presentations of this work were given at the World Federation of Hemophilia Meeting; May 20, 2002; Seville, Spain, and the Annual Conference of the American Physical Therapy Association; June 18 –22, 2003; Washington, DC. An oral presentation of this work was given at the Global Nursing Symposium; September 20 –22, 2005; Dublin, Ireland. This work was funded by a grant from Baxter Healthcare. This article was received June 26, 2008, and was accepted December 15, 2009. DOI: 10.2522/ptj.20080202
References 1 Prevention and control of haemophilia: memorandum from a joint WHO/WFH meeting. Bull World Health Organ. 1991; 69:17–26. 2 McDaniel P. Focus on factors. J Intraven Nurs. 2000;23:282–289. 3 Susman-Shaw A, Harrington C. Haemophilia: the facts. Nurs Stand. 1999;14:39 – 46. 4 Perry JJ, Alving BM. von Willebrand’s disease. Am Fam Physician 1990; 41:219 – 224. 5 Medical and Scientific Advisory Committee (MASAC) Recommendation #132: Standards and Criteria for the Care of Persons With Congenital Bleeding Disorders. New York, NY: National Hemophilia Foundation; approved March 24, 2002. 6 Rodriguez-Merchan EC. Effects of hemophilia on articulations of children and adults. Clin Orthop Relat Res. 1996;328: 7–13. 7 Molho P, Verrier P, Stieltjes N, et al. A retrospective study on chemical and radioactive synovectomy in severe haemophilia patients with recurrent haemarthrosis. Haemophilia. 1999;5:115–123. 8 Jansen NW, Roosendaal G, Lafeber FP. Understanding haemophilic arthropathy: an exploration of current open issues. Br J Haematol. 2008;143:632– 640. 9 Kern M, Blanchette V, Stain AM, et al. Clinical and cost implications of target joints in Canadian boys with severe hemophilia A. J Pediatr. 2004;145:628 – 634. 10 Roosendaal G, Jansen NW, Schutgens R, Lafeber FP. Haemophilic arthropathy: the importance of the earliest haemarthroses and consequences for treatment. Haemophilia. 2008;14(suppl 6):4 –10. 11 Choiniere M, Melzack R. Acute and chronic pain in hemophilia. Pain. 1987; 31:317–331. 12 Johnson RP, Babbitt DP. Five stages of joint disintegration compared with range of motion in hemophilia. Clin Orthop Relat Res. 1985;201:36 – 42. 13 Rodriguez-Merchan EC. Pathogenesis, early diagnosis, and prophylaxis for chronic hemophilic synovitis. Clin Orthop Relat Res. 1997;343:6 –11. 14 Arnold WD, Hilgartner MW. Hemophilic arthropathy: current concepts of pathogenesis and management. J Bone Joint Surg Am. 1977;59:287–305. 15 Ashrani AA, Osip J, Christie B, Key NS. Iliopsoas haemorrhage in patients with bleeding disorders: experience from one centre. Haemophilia. 2003;9:721–726. 16 Rodriguez-Merchan EC. Common orthopaedic problems in haemophilia. Haemophilia. 1999;5(suppl 1):53– 60. 17 Shaheen S, Alasha E. Hemophilic pseudotumor of the distal parts of the radius and ulna: a case report. J Bone Joint Surg Am. 2005;87:2546 –2549. 18 Valentino LA, Martinowitz U, Doolas A, Murali P. Surgical excision of a giant pelvic pseudotumour in a patient with haemophilia A. Haemophilia. 2006;12:541–544.
Volume 90
Number 4
Physical Therapy f
521
Supervised Exercise Program for People With Bleeding Disorders and Hemophilic Arthritis 19 Meena AK, Jayalakshmi S, Prasad VS, Murthy JM. Spinal epidural haematoma in a patient with haemophilia-B. Spinal Cord. 1998;36:658 – 660. 20 Medical and Scientific Advisory Committee (MASAC) Recommendation #175: Guidelines for Emergency Department Management of Individuals With Hemophilia. New York, NY: National Hemophilia Foundation; approved October 15, 2006. 21 Medical and Scientific Advisory Committee (MASAC) Recommendation #190: Recommendations Concerning the Treatment of Hemophilia and Other Bleeding Disorders. New York, NY: National Hemophilia Foundation; approved March 21, 2009. 22 Medical and Scientific Advisory Committee (MASAC) Recommendation #179: Recommendation Concerning Prophylaxis (Regular Administration of Clotting Factor Concentrate to Prevent Bleeding). New York, NY: National Hemophilia Foundation; approved November 4, 2007. 23 Medical and Scientific Advisory Committee (MASAC) Recommendation #185: Recommendations Regarding Women With Inherited Bleeding Disorders. New York, NY: National Hemophilia Foundation; approved November 15, 2008. 24 Arnold DM, Julian JA, Walker IR. Mortality rates and causes of death among all HIVpositive individuals with hemophilia in Canada over 21 years of follow-up. Blood. 2006;108:460 – 464. 25 Manco-Johnson MJ, Abshire TC, Shapiro AD, et al. Prophylaxis versus episodic treatment to prevent joint disease in boys with severe hemophilia. N Engl J Med. 2007;357:535–544. 26 Gilbert MS, Weidel JD. Treatment of Hemophilia: Current Orthopedic Management. New York, NY: National Hemophilia Foundation, 1995. 27 Gilbert MS, Radomisli TE. Therapeutic options in the management of hemophilic synovitis. Clin Orthop Relat Res. 1997;343:88 –92. 28 Koch B, Cohen S, Luban NC, Eng G. Hemophiliac knee: rehabilitation techniques. Arch Phys Med Rehabil. 1982;63:379 –382. 29 Limbird TJ, Dennis SC. Synovectomy and continuous passive motion (CPM) in hemophiliac patients. Arthroscopy. 1987;3:74–79. 30 Norian JM, Ries MD, Karp S, Hambleton J. Total knee arthroplasty in hemophilic arthropathy. J Bone Joint Surg Am. 2002; 84:1138 –1141. 31 Salomon O, Steinberg DM, Seligshon U. Variable bleeding manifestations characterize different types of surgery in patients with severe factor XI deficiency enabling parsimonious use of replacement therapy. Haemophilia. 2006;12:490 – 493. 32 Silva M, Luck JV Jr. Long-term results of primary total knee replacement in patients with hemophilia. J Bone Joint Surg Am. 2005;87:85–91. 33 Dietrich SL. Rehabilitation and nonsurgical management of musculoskeletal problems in the hemophilic patient. Ann N Y Acad Sci. 1975;240:328 –337.
522
f
Physical Therapy
Volume 90
34 Gilbert MS. Should Older Children and Adults With Hemophilia Participate in Sports and Recreational Activities? New York, NY: National Hemophilia Foundation; 1996. 35 Heijnen L, Mauser-Bunschoten EP, Roosendaal G. Participation in sports by Dutch persons with haemophilia. Haemophilia. 2000;6:537–546. 36 Lane H, Audet M, Herman-Hilker S, Houghton S. Physical Therapy in Bleeding Disorders. New York, NY: National Hemophilia Foundation; 2004. 37 Nazzaro AM, Owens S, Hoots WK, Larson KL. Knowledge, attitudes, and behaviors of youths in the US hemophilia population: results of a national survey. Am J Public Health. 2006;96:1618 –1622. 38 Greene WB. Musculoskeletal aspects of hemophilia. Mo Med. 1984;81:136 –140. 39 Buzzard BM. Sports and hemophilia: antagonist or protagonist. Clin Orthop Relat Res. 1996;328:25–30. 40 Engelbert RH, Plantinga M, Van der Net J, et al. Aerobic capacity in children with hemophilia. J Pediatr. 2008;152:833– 838, 838e1. 41 Falk B, Portal S, Tiktinsky R, et al. Anaerobic power and muscle strength in young hemophilia patients. Med Sci Sports Exerc. 2000;32:52–57. 42 Koch B, Galioto FM Jr, Kelleher J, Goldstein D. Physical fitness in children with hemophilia. Arch Phys Med Rehabil. 1984;65:324 –326. 43 Wittmeier K, Mulder K. Enhancing lifestyle for individuals with haemophilia through physical activity and exercise: the role of physiotherapy. Haemophilia. 2007;13(suppl 2):31–37. 44 Hofstede FG, Fijnvandraat K, Plug I, et al. Obesity: a new disaster for haemophilic patients? A nationwide survey. Haemophilia. 2008;14:1035–1038. 45 Armstrong LE, Whaley MH, Brubaker PH, Otto RM. ACSM’s Guidelines for Exercise Testing and Prescription. 7th ed. Philadelphia, PA: Lippincott, Williams & Wilkins; 2006. 46 Harris S, Boggio LN. Exercise may decrease further destruction in the adult haemophilic joint. Haemophilia. 2006;12: 237–240. 47 Mulder K. Exercise for People With Hemophilia. Montreal, Quebec, Canada: World Federation of Hemophilia; 2006:44. 48 Jones P, Buzzard BM, Heijnen L. Go for It: Guidance on Physical Activity and Sports for People With Haemophilia and Related Disorders. Montreal, Quebec, Canada: World Federation of Hemophilia; 1998. 49 Anderson A, Forsyth A. Playing It Safe, Bleeding Disorders, Sports and Exercise. New York, NY: National Hemophilia Foundation; 2005:44. 50 Hilberg T, Herbsleb M, Puta C, et al. Physical training increases isometric muscular strength and proprioceptive performance in haemophilic subjects. Haemophilia. 2003;9:86 –93.
Number 4
51 Pelletier JR, Findley TW, Gemma SA. Isometric exercise for an individual with hemophilic arthropathy. Phys Ther. 1987;67: 1359 –1364. 52 ATS statement: guidelines for the SixMinute Walk Test. Am J Respir Crit Care Med. 2002;166:111–117. 53 Norkin C, White D. Measurement of Joint Motion: A Guide to Goniometry. Philadelphia, PA: FA Davis Co; 1995. 54 Universal Data Collection Joint Range of Motion Reference Guide: Orientation Manual for Physical Therapists. New York, NY: National Hemophilia Foundation/Centers for Disease Control and Prevention; 1997. 55 Hislop H, Montgomery J. Daniels and Worthingham’s Muscle Testing: Techniques of Manual Examination. Philadelphia, PA: WB Saunders Co; 1995. 56 Byl NN, Richards S, Asturias J. Intrarater and interrater reliability of strength measurements of the biceps and deltoid using a hand held dynamometer. J Orthop Sports Phys Ther. 1988;9:395–398. 57 Surburg PR, Suomi R, Poppy WK. Validity and reliability of a hand-held dynamometer applied to adults with mental retardation. Arch Phys Med Rehabil. 1992;73: 535–539. 58 Wadsworth CT, Krishnan R, Sear M, et al. Intrarater reliability of manual muscle testing and hand-held dynametric muscle testing. Phys Ther. 1987;67:1342–1347. 59 Taylor R, Jayasinghe UW, Koelmeyer L, et al. Reliability and validity of arm volume measurements for assessment of lymphedema. Phys Ther. 2006;86:205–214. 60 Kisner C, Colby LA. Therapeutic Exercise Foundations and Techniques. Philadelphia, PA: FA Davis Co; 2007:928. 61 Anderson A, Forsyth A. Playing It Safe. Bleeding Disorders, Sports and Exercise. New York, NY: National Hemophilia Foundation; 2000:44. 62 Anderson A, Holtzman TS, Masley J. Physical Therapy in Bleeding Disorders. New York, NY: National Hemophilia Foundation; 2000:30. 63 Portney L, Watkins M. Foundations of Clinical Research: Applications to Practice. Upper Saddle River, NJ: Pearson Prentice Hall; 2009:892. 64 Capili B, Anastasi JK. Body mass index and nutritional intake in patients with HIV and chronic diarrhea: a secondary analysis. J Am Acad Nurse Pract. 2008;20: 463– 470. 65 Derr J. Valid paired data designs: make full use of the data without “double-dipping.” J Orthop Sports Phys Ther. 2006;36:42– 44. 66 Menz HB. Analysis of paired data in physical therapy research: time to stop doubledipping? J Orthop Sports Phys Ther. 2005; 35:477– 478. 67 Healthy People 2010. Office of Disease Prevention and Health Promotion, US Department of Health and Human Services. Available at: http://www.healthypeople.gov/. Accessed January 16, 2008.
April 2010
Supervised Exercise Program for People With Bleeding Disorders and Hemophilic Arthritis Appendix 1. Strength Training Protocol and Progression
1. Progression to next level only if no adverse reaction to previous week of exercise. 2. Prophylaxis: Factor infusion recommended for people with severe hemophilia; people with mild and moderate hemophilia to have medications available, if needed. 3. Intensity⫽percent of isometric Nicholas dynamometry muscle test (INDMT) to assess pounds of weight to use or color of Thera-Band exercise band. The Hygenic Corporationa reports correspondence of colors to weight resistance as the following: yellow⫽2.5 lb, red⫽4.5 lb, green⫽5.0 lb, blue⫽7.5 lb, black⫽9.0 lb, and silver⫽15 lb 4. Repetition⫽to be done only in pain-free range. 5. Rate⫽5–10 seconds concentric with exhale; 5–10 seconds with inhale. Level 1: Prescribed for the most fragile joints, target joints, previously injured muscle, and joints with painful active range of motion, passive range of motion, or weight bearing. No acute swelling or bleeding within past 2 weeks. Intensity
No. of Repetitions
No. of Sets
Week 1
40%
10
1
Week 2
45%–50%
10–20
2
Week 3
50%–60%
10–20
3
Week 4
55%–65%
10–20
3
Week 5
60%–70%
10–20
3
Week 6
65%–75%
10–20
3
Progression
Level 2: Prescribed for joints and muscles with history of bleeding and chronic, mild-to-moderate impairment. No bleeding in past 6 months. Progression
Intensity
No. of Repetitions
No. of Sets
Week 1
50%
10
1
Week 2
55%–60%
10–20
2
Week 3
60%–70%
10–20
3
Week 4
65%–75%
10–20
3
Week 5
70%–75%
10–20
3
Week 6
75%
10–20
3
Level 3: Prescribed for joints and muscles with minimal history of bleeding and no signs of impairment. Intensity
No. of Repetitions
No. of Sets
Week 1
60%
10–20
1
Week 2
65%–70%
10–20
2
Week 3
70%–75%
10–20
3
Week 4
75%
10–20
3
Week 5
75%
10–20
3
Week 6
75%
10–20
3
Progression
(Continued)
April 2010
Volume 90
Number 4
Physical Therapy f
523
Supervised Exercise Program for People With Bleeding Disorders and Hemophilic Arthritis Appendix 1. Continued
Example: Participant 4, a 23-year-old man with severe hemophilia. Right elbow⫽level 1: Target joints; had 6 episodes of bleeding over past 2 months. Active and passive range of motion painful at end range of flexion and extension. No acute swelling, no bleeding within past 2 weeks. Right biceps muscle isometric Nicholas dynamometry muscle test⫽5 lb. Left elbow⫽level 3: Only 2 episodes of bleeding in past. Last episode of bleeding was 2 years previously. Pain-free motion, no swelling or crepitus. Normal end-feel. Left biceps muscle isometric Nicholas dynamometry muscle test⫽ 30 lb. Week 1: Right elbow flexion: 40% ⫻ 5 lb⫽lift 2 lb or use yellow Thera-Band for 1 set of 10 repetitions in pain-free range. Left elbow flexion: 60% ⫻ 30 lb⫽lift 18 lb or double thickness of black Thera-Band for 1 set of 10 –20 repetitions in pain-free range. a
The Hygenic Corporation, 1245 Home Ave, Akron, OH 44310.
524
f
Physical Therapy
Volume 90
Number 4
April 2010
Supervised Exercise Program for People With Bleeding Disorders and Hemophilic Arthritis Appendix 2. Statistical Results When Separating Out Left Versus Right Joint Results for Range of Motion, Manual Muscle Test, and Circumferential Measures Baseline Measurements
Range of Motion (°) (nⴝ20) Left knee flexion
a
Right knee flexiona Left knee extension
X
SD
115.9
29.0
108.1
35.6
⫺8.2
a
Right knee extension
⫺6.0
c
Final Measurements
Range
X
SD
28 to 149
122.5
28.3
43 to 155
40 to 145
114.3
33.3
40 to 152
⫺2.0
6.8
⫺16 to 12b
⫺2.2
7.4
⫺18 to 10b
12.7
⫺38 to 14
b
8.9
⫺24 to 14
b
Range
Left ankle dorsiflexiona
⫺4.3
9.5
⫺20 to 10
1.3
10.2
⫺20 to 20
Right ankle dorsiflexiond
⫺0.9
11.3
⫺18 to 18
1.2
11.1
⫺20 to 26
38.8
11.1
20 to 50
48.7
9.3
36 to 70
40.6
10.2
15 to 60
45.4
10.0
30 to 65
131.3
20.3
80 to 153
136.4
17.9
80 to 154
133.4
14.8
105 to 150
138.0
12.2
110 to 155
⫺12.4
22.6
⫺87 to 10b
⫺9.5
20.0
⫺75 to 10b
⫺12.4
20.2
⫺52 to 20
⫺7.5
16.5
⫺40 to 14b
Left elbow pronationc
60.5
20.0
10 to 85
65.5
18.5
20 to 92
Right elbow pronationd
57.5
26.7
5 to 85
63.6
22.4
20 to 90
Left elbow supinationd
65.3
24.4
0 to 90
68.8
21.9
15 to 90
Right elbow supinationd
55.3
34.5
⫺30 to 90
61.7
30.3
⫺20 to 90
Left ankle plantar flexion
a
Right ankle plantar flexionc Left elbow flexion
c
Right elbow flexionc Left elbow extensiona Right elbow extension
c
b
a
Pⱕ.01. Hypertension. Pⱕ.05. d Not significant. b c
Muscle Strength (N) (nⴝ17) Left hip extension
a,b
Right hip extensiona,b Left hip flexion
a,b
Right hip flexiona,b Left hip abduction
b
X
SD
Range
15.9
6.8
4.3 to 26.6
22.9
8.5
7.1 to 39.0
15.4
6.9
0.0 to 24.1
22.8
8.6
4.3 to 33.5
18.6
5.6
8.1 to 32.6
23.4
6.3
13.0 to 36.3
17.7
6.5
6.3 to 30.2
21.3
7.0
9.5 to 30.7
5.6
7.2 to 29.1
21.6
7.8
11.4 to 38.0
16.7
6.4
6.1 to 25.9
20.7
8.4
10.9 to 36.5
b
14.4
7.2
1.0 to 23.8
19.9
9.2
4.3 to 28.5
15.6
8.5
2.1 to 29.7
21.0
10.8
3.8 to 40.6
c
12.1
5.4
0.8 to 21.6
15.6
6.7
4.6 to 28.6
Right knee flexionb
13.5
6.8
4.0 to 27.2
17.0
6.3
6.6 to 29.0
c
13.6
7.6
1.6 to 30.1
16.7
7.6
2.3 to 28.4
15.1
7.5
0.0 to 29.6
17.4
8.1
0.8 to 31.6
11.2
6.4
0.3 to 16.6
13.6
7.4
0.4 to 26.0
12.2
6.5
0.0 to 24.8
14.8
7.2
0.5 to 28.8
Left knee flexion
Left elbow flexion
Right elbow flexionc Left elbow extension
b
Right elbow extensionc
c
Range
16.2
Right knee extensionb
b
Final Measurements
SD
Right hip abductionc Left knee extension
a
Baseline Measurements X
n⫽16. Pⱕ.01. Pⱕ.05. (Continued)
April 2010
Volume 90
Number 4
Physical Therapy f
525
Supervised Exercise Program for People With Bleeding Disorders and Hemophilic Arthritis Appendix 2. Continued Baseline Measurement
Circumferential Measure (cm) (nⴝ20)
Final Measurement
X
SD
Range
X
SD
Range
Left knee 6 in above joint linea
45.0
6.8
33.5 to 60.0
47.0
8.3
33.0 to 69.0
Right knee 6 in above joint linea
45.0
7.7
32.0 to 66.5
47.0
8.1
33.0 to 68.5
Left knee at joint lineb
37.9
3.6
31.0 to 46.0
38.3
3.2
32.5 to 44.0
37.7
3.1
32.5 to 43.0
37.9
2.9
32.5 to 43.5
28.6
4.4
28.0 to 43.5
32.9
4.4
28.0 to 42.5
28.6
4.3
28.0 to 45.0
32.4
4.2
28.0 to 45.0
29.0
5.4
21.0 to 38.0
30.0
5.8
19.8 to 37.5
30.0
5.5
19.5 to 40.0
31.0
5.8
20.0 to 42.0
27.7
3.3
21.5 to 33.0
27.3
3.2
22.0 to 31.5
28.2
3.4
22.5 to 33.0
27.8
3.3
22.0 to 32.0
25.3
3.6
19.0 to 32.0
26.2
4.0
19.5 to 34.0
25.9
3.7
21.0 to 33.5
26.4
3.7
19.5 to 32.0
Right knee at joint line
b
Left knee 4 in below joint lineb Right knee 4 in below joint line
b
Left elbow 6 in above joint linec Right elbow 6 in above joint line
b
Left elbow at joint lineb Right elbow at joint line
b
Left elbow 4 in below joint linea Right elbow 4 in below joint line
b
a
Pⱕ.01. b Not significant. c Pⱕ.05.
526
f
Physical Therapy
Volume 90
Number 4
April 2010
Research Report Effect of Intensive Outpatient Physical Training on Gait Performance and Cardiovascular Health in People With Hemiparesis After Stroke Jørgen Roed Jørgensen, Daniel Thue Bech-Pedersen, Peter Zeeman, Janne Sørensen, Lars L. Andersen, Michael Scho¨nberger
Background. Stroke can result in severe motor deficits, and many people who have survived a stroke have poor cardiovascular fitness, with potentially disabling effects on daily life. Objective. The objective of this study was to evaluate the impact of intensive physical training on gait performance and cardiovascular health parameters in people with stroke in the chronic stage.
Design. This was a single-group, pretest-posttest experimental study. Methods. Fourteen people with hemiparesis after cerebrovascular injury (mean age⫽58.4 years, mean time since injury⫽25 months) participated in a 12-week training intervention, 5 times per week for 1.5 hours per session. The intervention consisted of high-intensity, body-weight–supported treadmill training; progressive resistance strength training; and aerobic exercise. The main outcome measures were gait performance (Six-Minute Walk Test, 10-Meter Walk Test, and aerobic capacity) and parameters of cardiovascular health (systolic and diastolic blood pressures, body mass index, and resting heart rate).
Results. Significant improvements in all main outcome parameters were observed in response to the intervention. Gait speed during the Six-Minute Walk Test increased 62%, and systolic and diastolic blood pressures decreased 10% and 11%, respectively. Weekly testing of walking speed showed that most of the increase in the walking speed occurred in the first 8 weeks of training. Correlation analyses showed that improvements were unrelated to age, chronicity, or level of functioning.
Conclusions. High-intensity physical training for people with stroke in the chronic stage increased walking speed regardless of chronicity, age, or level of functioning. Further studies should investigate the intervention duration needed to reach the full potential of gait recovery.
J.R. Jørgensen is Physiotherapist, Center for Rehabilitation of Brain Injury, University of Copenhagen, Amagerfaelledveg 56A, DK-2300 Copenhagen, Denmark. Address all correspondence to Mr Jørgensen at:
[email protected]. D.T. Bech-Pedersen is Physiotherapist, Center for Rehabilitation of Brain Injury, University of Copenhagen. P. Zeeman is Physiotherapist, Center for Rehabilitation of Brain Injury, University of Copenhagen. J. Sørensen is Physiotherapist, Center for Rehabilitation of Brain Injury, University of Copenhagen. L.L. Andersen, PhD, is Postdoctoral Fellow, National Research Centre for the Working Environment, Copenhagen, Denmark. M. Scho¨nberger, PhD, is Research Fellow, School of Psychology and Psychiatry, Monash University, Melbourne, Victoria, Australia, and Monash-Epworth Rehabilitation Research Centre, Epworth Hospital, Melbourne, Victoria, Australia. [Jørgensen JR, Bech-Pedersen DT, Zeeman P, et al. Effect of intensive outpatient physical training on gait performance and cardiovascular health in people with hemiparesis after stroke. Phys Ther. 2010;90:527–537.] © 2010 American Physical Therapy Association
Post a Rapid Response or find The Bottom Line: www.ptjournal.org April 2010
Volume 90
Number 4
Physical Therapy f
527
Intensive Physical Training After Stroke
S
troke is the most common form of acquired brain injury and is one of the leading causes of death and disability worldwide.1 In Denmark alone, 11,000 people in a total population of 5.4 million have a stroke annually. A slightly higher relative number of people have a stroke each year in the United States.2 It has been estimated that at any given time, 30,000 to 40,000 people who have survived a stroke reside in Denmark, incurring high expenses for primary health care, social services, and disability pensions.3 Hemiparesis is a common consequence of stroke and often leads to problems across multiple systems, including loss of strength (force-generating capacity) and dexterity and poor motor control. Upper-extremity (UE) and lower-extremity (LE) pareses, frequently combined with significantly reduced overground walking speed and walking distance, constitute severe impediments to the ability to perform activities of daily living, participate in normal social life, and manage gainful employment. In addition to persistent motor deficits, concomitant poor endurance and increased fatigability can constitute a psychological burden to people with stroke, as well as their significant others. At 6 months after the onset of stroke, people are in the chronic stage, and the physical therapy goal often shifts from rehabilitation to maintenance training.4 Although persistent hemiparesis and concomitant low levels
Available With This Article at ptjournal.apta.org • Intervention Video • Audio Abstracts Podcast This article was published ahead of print on March 4, 2010, at ptjournal.apta.org.
528
f
Physical Therapy
Volume 90
of physical activity are independent risk factors for new cerebrovascular and cardiovascular events,5 rehabilitation and maintenance training time allocated by private and public physical therapy clinics in Denmark to people with chronic stroke amounts to no more than 1 hour once or twice per week. The lack of available equipment constitutes a further hindrance to training efficacy. These factors may explain why the intensity of conventional physical therapy delivered to people with hemiparesis tends to be low to moderate. Despite rehabilitative interventions, motor sequelae remaining after discharge often are considerable.6 The comprehensive Copenhagen Stroke Study showed that functional and neural recovery reached a plateau within 6 to 20 weeks after the lesion.7 The absence of further recovery over time may be ascribed partly to the facts that the abovementioned parameters are assigned the lowest priority in neurorehabilitation, that the intensity of physical rehabilitation is inadequate, or both.8 Body-weight–supported treadmill training (BWSTT) currently is gaining recognition as an effective way to improve walking ability after stroke. Evidence for this type of intervention has not yet been fully established, and research results are not yet unanimously favorable. On the basis of a review of 15 trials including 622 participants, Moseley et al9 concluded that there were no statistically significant differences between treadmill training, with or without body-weight support, and other interventions for walking speed or independence. However, they found a non–statistically significant tendency for independent walkers training with BWSTT to show improvements in walking speed. In addition, they noted that the treatment effects might be highly dependent on the intensity of the
Number 4
protocol used. Other researchers have obtained more promising results in favor of BWSTT compared with conventional treatments. Pohl et al10 found significantly higher overground walking speed, cadence, stride length, and Functional Ambulation Category scores in a study comparing the effects of highintensity, speed-dependent treadmill training with the effects of limited progressive treadmill training and conventional gait training for people with hemiparesis. These data support the use of a high-intensity training protocol, an observation supported by other research groups.11,12 However, more research is warranted in this area. Weakness after stroke is a common phenomenon, and emerging evidence suggests that it may be responsible for compromised motor function5 and that it is related to functional activity performance.13 Notably, gait ability is closely related to muscle strength.14,15 Consequently, the restoration of muscle strength should be a cornerstone of rehabilitation. A systematic review of the effects of progressive resistance strength training (PRST) revealed preliminary evidence that PRST programs reduce musculoskeletal impairment, but whether they enhance the performance of functional activities and participation in societal roles remains unknown.16 The failure to demonstrate consistent benefits of PRST may be due to the heterogeneity of symptoms typically found in people with hemiparesis.17 Along with impaired muscle strength and gait in people with hemiparesis, cardiovascular fitness after a stroke has been found to be as low as 50% to 70% of that of sedentary age- and sex-matched individuals.18 The level of physical activity is related to the aerobic capacity, and a critical level of aerobic capacity must be met to function independently.13 The enApril 2010
Intensive Physical Training After Stroke ergy level required to perform routine ambulation is 1.5- to 2-fold higher in people who have had a stroke than in people who are healthy, and this level can represent 76% of physiological capacity.13 On the basis of a recent Cochrane review,19 including 12 studies, the Danish National Board of Health concluded that aerobic exercise improves ambulation after stroke. The board consequently recommended that aerobic exercise be included in rehabilitation after stroke.3 Because weakness after stroke and a low level of fitness appear to be related to functional activity performance, it seems advisable to emphasize the importance of allocating rehabilitation time to the enhancement of strength training and aerobic exercise (AE) through highintensity training. Numerous studies have provided evidence that PRST increases strength and that AE increases peak oxygen consumption ˙ O2),14,20 but there is limited evi(V dence for the generalizability of such increases to increased functional activities in people with stroke. The paucity of evidence may be ascribed to the heterogeneity of study participants and to the fact that the relatively low levels of intensity and allocation of many interventions cannot elicit the desired generalizations. The combined effects of PRST, AE, and BWSTT have scarcely been investigated, although promising results have been demonstrated in a single study.21 It is clear that more research is needed in this area. At present, conventional stroke rehabilitation in Denmark does not comprise intensive strength training or BWSTT. The aim of this study was to investigate the effects of high-intensity BWSTT, PRST, and AE on gait performance and cardiovascular health parameters in people with postApril 2010
stroke hemiparesis in the chronic stage. We hypothesized that the training intervention would lead to significant improvements in participants’ gait performance and cardiovascular health, regardless of age, time interval between the injury and the start of the intervention, and amount of physical training received before the intervention.
Method Design We used a single-group, pretreatmentposttreatment design, with measurement of participants’ aerobic capacity before and after the 12-week training intervention. Additionally, participants’ performance during physical training was monitored each week (ie, 12 measurements). Participants The participants were people with cerebrovascular injuries (3 lefthemisphere lesions and 11 righthemisphere lesions) in the chronic stage and with hemiparesis resulting in moderate to severe UE and LE motor impairments. Ten of the participants had no volitional function of the affected UE; the remaining 4 participants were able to use the affected UE for support but not for any activities requiring dexterity. All 14 participants were unable to stand on the affected LE without support. None of the participants was able to perform volitional dorsiflexion or eversion of the affected ankle or isolated flexion of the hip or knee on the affected side while standing. Inclusion criteria were a chronicity of more than 3 months; moderate to severe hemiparesis, with pronounced spasticity (hypertonicity) and synergistic movements of both the UE and the LE and with no volitional dorsiflexion and eversion of the affected ankle joint and no volitional flexion of the hip and knee joints on the affected side while standing; and an age of at least 50 years. Gait function
had to be moderately to severely impaired, so that the maximum walking speed would be less than 50% of the normal walking speed for age-, height-, weight-, and sex-matched people.22 Participants had to be able to perform the Six-Minute Walk Test (6MWT). All referrals to the program were made by physicians or general practitioners who, having received detailed information about the intervention, had not found the strenuous nature of the intervention to constitute a contraindication to participation. Although exclusion criteria did not comprise atrial fibrillation and previous strokes or heart attacks, all participants in the present study had experienced only a first-ever stroke, and none had a medical history of atrial fibrillation or heart attacks. All participants were medically stable, independent with regard to basic activities of daily living, and motivated for strenuous physical exercise. Alcohol or substance abuse, psychiatric disorders, and any progressive diseases were exclusion criteria. Medications used by the participants were statins (n⫽8), antihypertensives (n⫽8), blood thinners (n⫽7), anticoagulants (n⫽6), diuretics (n⫽3), antispasmodics (n⫽3), anticonvulsants (n⫽3), 1-receptor blockers (n⫽2), antidepressants (n⫽4), and nonprescription analgesics (n⫽3). Of those medications, only the 1-receptor blockers affect exercise tolerance by lowering the maximum heart rate. There were no changes in medications over the study period, except that one individual began taking nonprescription analgesics and antidepressants during the study. A total of 31 people were referred to the program. Seven individuals were excluded from participation for the following reasons: gait function was too severely impaired for 2 people to adhere to the initial test require-
Volume 90
Number 4
Physical Therapy f
529
Intensive Physical Training After Stroke Table 1. Demographic and Medical Characteristics of Participantsa Percentiles Characteristic
a
Minimum
Maximum
25%
50%
75%
X (SD)
Age at injury (y)
51.3
70.1
53.3
57.7
61.3
58.4 (6.1)
Age at program entry (y)
52.4
71.4
55.9
58.9
63.6
60.4 (5.7)
Time between injury and beginning of training (mo)
2.7
84.7
6.5
18.1
31.3
24.6 (23.1)
Amount of training before intervention (h/wk)
0.0
6.0
1.9
2.8
4.2
3.0 (1.7)
Thirteen participants (93%) were men, and 1 participant (7%) was a woman.
ments, 2 people had an impairment other than hemiparesis, 2 people had to interrupt their participation because of illness, and 1 individual did not participate on a regular basis. Twenty-four people participated in the intervention, but 10 of these participants were excluded from the present study—not because of significantly different outcomes but to obtain group homogeneity with regard to age and etiology. Among the participants excluded were people with traumatic brain injuries, spaceoccupying lesions, and tumors. Consequently, the present study included only 14 of the 24 participants who completed the full 12-week intervention. All but 1 participant required some type of walking aid on admission; 11 used a cane or crutch, and 12 used an ankle-foot orthosis. Finally, 2 of the participants used wheelchairs, 2 participants used an electric scooter, and 1 participant used a wheeled walker. Table 1 shows the demographic and medical characteristics of the 14 participants. Time since injury varied from 3 months to 7 years. The distribution of time since injury was heavily skewed; 50% of the participants began the intervention within the first 16 months after the injury, and 86% began within the first 40 months. Preintervention training time with a physical therapist varied considerably. Although only 1 participant had received no training at all 530
f
Physical Therapy
Volume 90
and 2 participants had received more than 5 hours of training, 79% of the participants had received between 1.5 and 5 hours of training per week. Although the majority of the participants were men (only 1 woman was included in the study), this skewed sex distribution was coincidental and did not reflect the general referral pattern. All participants provided written informed consent before participation in the study. Intervention The intervention took place at the gait rehabilitation facility of the Center for Rehabilitation of Brain Injury (see video, available at ptjournal. apta.org). The facility had been especially equipped for the project, which comprised 12 weeks of training, 5 times per week for 1.5 hours per session. The intervention consisted of 4 key elements: BWSTT, AE, PRST, and functional training. The chief objective was to improve gait function (ie, ambulatory safety), walking speed, and walking distance. Moreover, the intervention aimed at improving maximum muscle strength and cardiorespiratory fitness. Each intervention week comprised 3 days with the main emphasis on strength training activities and 2 days with an emphasis on cardiorespiratory endurance training and functional training. Training sessions on all training days invariably began with BWSTT. To ensure safety and
Number 4
maximum training intensity, a personal physical therapist was assigned to each participant. All output data were collected on individual preprogrammed USB memory sticks (TGS*) and stored on the central computer together with heart rate data from all AE sessions. The participants’ personal physical therapists evaluated and revised training output daily to ensure continuous improvement. Decisions about progression and daily reprogramming of the USB memory sticks were made jointly by the physical therapists but often required ad hoc adjustments because of overestimation or underestimation of participants’ energy, strength, and endurance, which fluctuated not only from day to day but also during the course of the training sessions. Maximum walking speed was assessed every Monday with the 6MWT, the 10-Meter Walk Test (10MWT), or both to ensure continuous walking speed progress and to boost participants’ motivation and adherence. To promote the transfer of training progress to daily life and to secure optimum restitution after the intensive strength training, we modified the training schedule for weeks 4 and 10. In these weeks, the tasks consisted of treadmill training and functional training but not PRST.
* Technogym SpA, Via Perticari, 20 Gambettola (FC), Italy.
April 2010
Intensive Physical Training After Stroke BWSTT. The BWSTT system used was a prototype developed by the Center for Rehabilitation of Brain Injury in collaboration with Ergolet A/S,† which specializes in track lift systems. Mounted on a 10-m ceiling track, the track lift either can be used in a locked position or can ride freely along the rail, thus making it possible to put the harness on before stepping onto the treadmill. The lift system is an open-loop, force-controlled mechanism, operating with a pneumatic cylinder. Attached at one end to the lift and at the other end to the weight relief spreader bar, the lift cable travels over a 55-cm lever with a floating range from top to bottom of 50 cm; therefore, a constant amount of body-weight support is delivered when the center of mass fluctuates along a sinusoidal wave, which often is exacerbated by a pronounced limp. The weight relief can be adjusted from 0 to 80 kg; the maximum lifting capacity is 200 kg. The harnesses were Guldmann Active Trainer harnesses,‡ and the treadmill was a standard Technogym Runrace treadmill* with TGS USB memory sticks.
fluctuating speed, and cardiorespiratory endurance. For all components of gait, the highest safely attainable speed was encouraged. However, given the heterogeneity of the participants, we made allowances for individual gait patterns, muscle strength, and endurance. Although it was impossible to adhere strictly to a subdivision into separate and distinguishable walking periods, physical therapists attempted to observe the following basic principles: • All participants were encouraged to maintain a treadmill speed that was significantly higher than the average speed of the most recent 6MWT. • Holding on to the bar of the treadmill was permitted only when absolutely necessary. • The cardiorespiratory system was challenged by increasing speed and by increasing the treadmill gradient up to 10% to approach, as closely as possible, 80% of the maximum heart rate. • To promote movement quality at higher walking speeds and higher levels of cardiovascular intensity, some participants were fitted with an Activister,§ an elastic band that was twisted around the LE to correct excess lateral (external) or medial (internal) rotation, depending on the direction of the band. • Some participants required manual guidance from therapists during various parts of the gait cycle to enhance gait quality and walking speed. • Body-weight support, which was determined as the amount of support that would enhance gait quality, varied from 10 to 25 kg. • Treadmill speeds, intervals, gradient, and dosage were evaluated daily and increased whenever possible.
The chief goal of each treadmill session was for participants to walk as far and as fast as possible without breaking up the gait pattern or risking safety, while increasing the maximum walking speed whenever possible and constantly focusing on movement quality. Training sessions on the treadmill lasted up to 25 minutes and normally consisted of 3 walking periods of 6 to 8 minutes each, interspersed with breaks. Some participants walked less and needed more breaks because of fatigue. Ideally, the 3 walking periods were designed to focus on 3 components of gait: symmetry and balance, † Ergolet A/S, Taarnborgvej 12C, DK-4220 Korsoer, Denmark. ‡ V Guldmann A/S, Graham Bells Vej 21–23A, DK-8200 Århus N, Denmark.
April 2010
§
Ortopaedingenioerene, Gl. Darupvej 5C, DK-4000 Roskilde, Denmark.
Aerobic exercise. Aerobic exercise comprised 5 separate training stations with TGS USB memory sticks: BWSTT (Technogym Runrace), stationary bipedal and unipedal paretic leg cycling (Technogym Bikerace HC600*), unipedal paretic arm cycling (Technogym XT PRO Top600*), and body-weight– supported stair climbing (Technogym Steprace HC300*). Training intensity was monitored with a heart rate monitor transmitting continuously to the TGS key of the training station to ensure an adequate aerobic challenge. At the beginning of the intervention, some participants could not reach the heart rate target zone; in 2 cases, the explanation was the heart rate– lowering effect of 1-receptor blockers. The goal for these individuals was to gradually maximize the heart rate attained in each activity. Except for weeks 4 and 10, Mondays and Wednesdays were AE days, with approximately 1 hour of aerobic exercise after the initial 25 to 30 minutes of BWSTT. The actual amount of cardiorespiratory training achieved during the 2 weekly AE sessions depended on each participant’s endurance. At the beginning of the 12-week program, most participants were so deconditioned and unaccustomed to cardiorespiratory challenges that their endurance permitted only short bursts of 5 to 6 minutes of any given activity at an intensity of 80% of the maximum heart rate before they required a break and a drink of water. As fitness levels improved, so did endurance and acceptance of more time spent on each activity. Toward the end of the training program, most participants were able to tolerate 12 to 15 minutes of each cardiorespiratory activity at a heart rate of at least 80% of their estimated maximum heart rate (220 bpm minus age [in years]). The principal goal of AE was to increase power output (ability to perform work over time), measured in watts, on the display of the machine; consequently, the goal of each train-
Volume 90
Number 4
Physical Therapy f
531
Intensive Physical Training After Stroke Table 2. Progression of Number of Repetitions in Progressive Resistance Strength Training Week
Repetition Maximum
Week
Repetition Maximum
1
12, 12, 12
7
12, 10, 10, 8
2
10, 10, 10
8
10, 8, 8, 8, 6
3
8, 8, 8, 8
4
Functional training 8, 8, 8, 8
11
10, 8, 8, 6
6
8, 8, 8, 8
12
8, 6, 6, 4
Progressive resistance strength training. Progressive resistance strength training comprised 4 activities with TGS USB memory sticks: semiseated leg press, leg curl, leg extension, and seated leg press with a Technogym Isotonic Line with Power Control§; the equipment offered visual feedback with regard to range of motion and power output in watts for each repetition and average power output for each set of repetitions. Except for weeks 4 and 10, Tuesdays, Thursdays, and Fridays were PRST days, with approximately 1 hour of resistance training after the initial 25 to 30 minutes of BWSTT. Although only 3 to 5 sets were performed unilaterally for 4 resistance activities, the total time spent on resistance training usually amounted to f
8, 6, 6, 6, 4 Functional training
5
ing session was to exceed the highest previous power output achieved. This goal was emphasized before each activity, and feedback about performance and goal attainment was given continuously. Power output was measured in watts by the computer of each machine and saved on TGS keys. In addition to increasing power output, there could be other goals, such as improving cadence or symmetry. Furthermore, during training for overground walking and stair climbing, participants wore a heart rate monitor to ensure an adequate aerobic challenge. Cardiorespiratory training was evaluated daily and, when possible, adjusted and increased.
532
9 10
Physical Therapy
Volume 90
1 hour per session because transferring from one training station to another as well as resting between sets was very time-consuming. Progressive resistance strength training was performed unilaterally with the paretic leg to ensure the highest possible training intensity for the affected extremity. Sets, repetitions, and a 90second resting pause between sets were identical for all participants. The weight load of each set was adjusted so that participants could only just perform the number of repetitions required in the set, that is, to volitional failure. All repetitions had to be performed with as much burst as possible to promote maximum movement speed and range of motion. The progression of the number of repetitions during the course of the PRST program is shown in Table 2. Functional training. In weeks 4 and 10, the training program was changed to ensure the effectiveness of the intensive BWSTT, PRST, and AE. In week 4, functional training replaced all PRST and AE; 0.5 hour of BWSTT per day was retained. In week 10, functional training also replaced all PRST and AE, except on the middle day of the week (Wednesday), when PRST was retained. On Fridays of weeks 4 and 10, the functional training concluded with a “walkathon,” a self-paced 30-minute walking test in which the participants covered as much distance as possible, thus proving the effects of
Number 4
their efforts. The components of functional training depended on the functional level of each participant, the goal being to ensure optimum carryover from functional improvements to activities of daily living. Functional training comprised training for specific details of the gait pattern, gait training in a nonclinical setting, stair climbing, and transfers. For participants with the potential ability to learn to ride a tricycle, outdoor cycling also was introduced in weeks 4 and 10. Measures The following measures were used at the beginning and end of the program. Systolic and diastolic blood pressures and resting heart rate were measured on the nonparetic arm with an Omron M4 device㛳 after the participant had been seated for 10 minutes; the average of 3 consecutive measurements was calculated. Body weight was measured with a Tanita BWB-600 scale# (we subtracted 2 kg for clothing and shoes), height was measured, and the body mass index was calculated. Walking speed was tested with the 6MWT23 on an indoor 50-m track without disturbances. No encouragement was offered during the test, except for information provided 㛳
Omron Corp, Shiokoji Horikawa, Shimagyoku, Kyoto 600-8530, Japan. # Tanita Europe BV, Kruisweg 813-A, 2132NG Hoofddorp, the Netherlands.
April 2010
Intensive Physical Training After Stroke Table 3. Test Results Obtained Before and After Program Before Program
Mean Difference (95% Confidence Interval)
Test
X (SD)
Systolic blood pressure (mm Hg)
12
142.3 (17.7)
144.5
127.6 (14.2)
126.0
14.7 (6.2–23.2)
.005
Diastolic blood pressure (mm Hg)
12
88.0 (10.2)
87.0
78.3 (10.3)
78.5
9.7 (3.2–16.1)
.017
Resting heart rate (bpm)
12
70.3 (10.6)
73.0
66.9 (11.0)
63.5
3.4 (⫺2.2–9.1)
.27
2
Median
X (SD)
Median
Pa
Body mass index (kg/m )
11
28.9 (4.3)
28.5
28.1 (4.3)
27.5
0.8 (0.2–1.4)
.005
Six-Minute Walk Test (km/h)
14
2.1 (1.1)
2.1
3.4 (1.3)
3.8
1.3 (1.0–1.6)
⬍.001
10-Meter Walk Test (s)
14
18.9 (12.2)
14.5
11.5 (9.1)
8.4
7.4 (4.4–10.4)
.001
5
22.4 (3.8)
21.7
26.5 (2.7)
26.0
4.1 (⫺0.7–8.9)
.08
11
1,570.5 (1,546.0)
1,601 (679–2,523)
.003
Estimated aerobic capacity (mL O2/min/kg) Self-rated maximum walking distance (m) a
After Program
No. of Participants
1,500.0
3,171.4 (2,041.4)
3,000.0
As determined with the Wilcoxon matched-pairs signed rank test (1-tailed).
once per minute about time elapsed. Participants used their customary assistive devices during the test; however, they were requested to walk without support from an elbow crutch or cane when possible. The 10MWT was used to assess participants’ maximum walking speed over a short distance.24 A 10-m portion of the 50-m track was used. The fastest of 3 attempts was used. They were requested to walk without support from an elbow crutch or cane when possible. Participants’ cardiorespiratory endurance was estimated by use of the submaximal stationary ergometer test from the Åstrand and Rhyming calculation of aerobic capacity from heart rate during submaximal work25,26 in combination with the Borg Rating of Perceived Exertion.27 Only 5 participants were able to complete the initial cardiorespiratory test. An additional 3 participants were able to complete the final test. The reason why 3 more participants were able to complete the postintervention test was that their aerobic capacity had improved sufficiently. During the initial interview, participants were requested to estimate April 2010
their maximum walking distance in meters when walking without any interruptions or breaks and when using their customary assistive devices. Three participants were unable to provide an estimate of their maximum walking distance. Participants also were asked about the amount and nature of training that they were receiving at the time of the interview. Participants’ performance was monitored once per week throughout the intervention by administering the 6MWT on Mondays and calculating their average walking speed during the fastest treadmill training interval on the Monday training sessions each week. Data Analysis For inferential statistics, we used nonparametric procedures (Spearman correlations, Friedman test, and Wilcoxon signed rank test with ␣ set at .05). For the examination of improvements in test performance, 1-tailed tests were used. For investigation of the relationship between improvements in test performance and participants’ age, time since injury, and amount of training before the intervention, 2-tailed tests were used. Statistical analyses were per-
formed with GPOWER.28
SPSS
13.0**
and
Role of the Funding Source Subjects’ participation in the intervention was funded by public health care. Equipment and research expenses were funded by the Center for Rehabilitation of Brain Injury, which is a nonprofit rehabilitation facility.
Results Gait Performance and Aerobic Capacity The group of participants as a whole improved on almost all measures of gait performance and aerobic capacity (Tab. 3). The improvements were consistent across participants. On the 6MWT and the 10MWT, all participants showed improved performance. On the submaximal stationary ergometer test, 4 of 5 participants showed improvements, and 3 participants who were unable to complete the initial test were able to complete the final test. These objective findings were reflected by the participants’ own experiences. All participants experienced improvements in maximum walking dis** SPSS Inc, 233 S Wacker Dr, Chicago, IL 60606.
Volume 90
Number 4
Physical Therapy f
533
Intensive Physical Training After Stroke tance. Their self-rated improvement of 102% was considerably higher than the objective improvements (62%, 39%, and 19% on the 6MWT, on the 10MWT, and in aerobic capacity, respectively). Finally, systolic and diastolic blood pressures decreased by 10% (P⫽.005) and 11% (P⫽.017), respectively.
Figure 1. Improvement in participants’ walking speed on the Six-Minute Walk Test during the training period. Data are means (circles) and standard deviations (error bars).
Figure 2. Improvement in participants’ walking speed during the fastest treadmill training interval. Data are means (circles) and standard deviations (error bars).
Relationships Among Main Parameters We examined whether the improvement in aerobic capacity was related to the participants’ age and the time interval between the brain injury and the initiation of the intervention. Age and time from injury to the initiation of the intervention were not related to aerobic capacity at the initiation of the intervention or to the improvement in aerobic capacity. No significant relationship between the amount of standard physical therapy training received before the intervention and the physical improvements experienced during the intervention was observed. Time-Wise Progression of Improvements We examined the time course of improvements in the participants’ walking speed on the 6MWT during the training period. The number of participants for whom walking speed data were available varied from week to week, from 8 to 14. As shown in Figure 1, participants showed approximately linear improvements in walking speed until week 8, when a plateau was reached (Friedman test, P⬍.001). The 6MWT results from week 1 and week 12 resembled the test results from before the intervention and after the intervention, respectively. We next examined the time course of improvements in the participants’ walking speed during the fastest treadmill training interval. In most weeks, 1 to 3 participants were not able to attend the test; in week 10, 6
534
f
Physical Therapy
Volume 90
Number 4
April 2010
Intensive Physical Training After Stroke participants could not attend for logistical reasons. The dramatic increase in performance occurring in the first 3 weeks of training was followed by more gradual, but still continuous, improvements in speed (Friedman test, P⬍.001) (Fig. 2). At all time points, the participants’ walking speed during the fastest treadmill training interval was above the speed on the 6MWT. This finding was statistically significant at all time points except weeks 1, 8, and 10 (Wilcoxon matched-pairs signed rank test, 2-tailed).
Discussion The principal findings of the present study are that people with stroke in the chronic stage can achieve clinically relevant improvements in gait performance and cardiovascular health parameters through highintensity physical training consisting of a combination of BWSTT, PRST, AE, and functional training. Gait Performance and Aerobic Capacity Common physical impairments after stroke are poor strength, a low level of fitness, and slow ambulation. These impairments, in combination with cognitive deficits, lead to decreased levels of activity and participation. Interventions for people in comparable age groups, for whom improvements in gait function had a priority similar to that in the present study, included just one element of physical training to determine its effect. Isolated BWSTT gait interventions resulted in improvements on the 6MWT, albeit of a magnitude more modest than that in the present study.29,30 Likewise, an examination of the effects of isolated PRST on gait, various functional parameters, and strength in people with poststroke hemiparesis and similar in age and chronicity to the participants in the present study revealed that although significant strength improvements could be elicited through April 2010
PRST, no significant difference between intervention and control groups could be found for any performance-based measure of function. However, a nonsignificant improvement of 10% on the 6MWT was noted.31 It is likely that the best effects of various elements can be achieved when they are combined. The literature offers little evidence for the generalizability of increased strength to increased functional abilities, but if strength training is combined with goal-oriented exercise, the outcome may be different. Our intervention included various elements of training at the highest possible level of intensity; this design made it more difficult to determine which element was responsible for which portion of the overall effect. The results of the present study, however, showed that the intervention had large effects on all parameters. The cardiovascular improvements achieved—10% and 11% decreases in systolic and diastolic blood pressures, respectively—were capable of significantly reducing the risk of recurrent stroke.32 The effects of exercise on blood pressures in the present study, therefore, were consistent with or superior to the findings of a similar study. Rimmer et al33 found that moderate-intensity, shorter-duration exercise had more favorable effects on systolic and diastolic blood pressures than did lower-intensity exercise but that neither moderate-intensity exercise nor lower-intensity exercise could in˙ O2 duce significant changes in peak V ˙ O2. It may be preor submaximal V sumed that the intensity of that intervention study was too low to elicit ˙ O2. A reduced risk of changes in V recurrence is commonly achieved medically, but if risk reduction is induced through an increase in activity, then it will have other beneficial effects as well.
Although the speed of gait increased markedly in the present study, we did not include any objective measures of movement quality during gait. However, our clinical observations indicated that the increase in walking speed was principally attributable to improvements in stance phase and stride length in the affected leg, as well as to an increase in steps per minute. The clinical relevance of increased walking speed is irrefutable. It has been demonstrated that the functional walking speed in a community environment should be between 4.1 and 5.4 k/h13; therefore, the participants in the present study attained functionality. Strengthening of the affected leg is a key element in a more stable stance phase. Therefore, PRST may have contributed to the increased speed on the 6MWT. During the initial weeks of the intervention, walking speed gradually decreased during the performance of the 6MWT because of fatigue, but by the end of the intervention, most of the participants were able to maintain a constant speed throughout the test. This observation can be ascribed not only to increased muscular endurance and a higher level of fitness but also to a more energy-efficient gait pattern. Overall, no signs of increased spasticity were observed by therapists or reported by participants. During strenuous exercise, spasticity did increase temporarily in some participants but tended to abate when the exercise was stopped. Some participants felt uncomfortable about this increase in spasticity during exercise, interpreting it as a loss of control. Relationships Among Main Parameters Regardless of chronicity, aerobic capacity at the start of the intervention was low and not related to time since
Volume 90
Number 4
Physical Therapy f
535
Intensive Physical Training After Stroke injury. The comprehensive Copenhagen Stroke Study reported that neurological recovery and functional recovery reach a plateau within 6 to 20 weeks after the occurrence of the lesion.7 This report might imply that once motor function has reached a plateau, no further changes can occur unless there is a massive change in the level of activity. This notion is consistent with the fact that all participants in the present study showed the same amounts of improvements in physical parameters irrespective of age or time since injury. Consequently, it seems likely that this kind of intervention will have an effect on most people who have had a stroke.
increased during the intervention. Therefore, speed was still increasing during the final weeks of the intervention, albeit to a lesser extent than during the initial weeks. These findings indicate that longer training periods of high-intensity BWSTT, PRST, and AE may lead to further improvements in physiological capacity. The other intervention elements also showed this tendency. Most of the initial progression appeared to be attributable to gradual adaptation to the exercises and determination of the appropriate level of intensity. Later, when the rate of progression decreased, further progression likely was attributable to real physiological changes.
Because all participants were in the chronic stage at the beginning of the intervention and because there was a substantial difference in the amounts of physical training that they received before the intervention, (ranging from 0 to 7 hours per week), the previous rehabilitative training did not relate to aerobic capacity at the beginning of the intervention or to improvements after the intervention. These findings indicate that an individual’s potential for improvement is not related to a prior intervention. Most of the participants reported that training received before the intervention did not involve sweating or being out of breath because it was less intense than the intervention. Training intensity seems to play a crucial role in an individual reaching full potential for functional recovery. Another important aspect is the magnitude of the total training volume over the course of a training week. In addition, the elements of the training protocol are factors that determine outcome.
The walking speed on the 6MWT increased approximately linearly throughout the intervention, indicating that if the duration of the intervention had been longer, then there might have been further improvement.
Time-Wise Progression of Improvements The time course for treadmill speed improvements revealed a gradual decrease in the rate at which speed 536
f
Physical Therapy
Volume 90
Because there is good evidence in the literature for speed-dependent BWSTT, 2 of the 3 sessions in the present study were speed dependent. In one session, the goal was to reach the highest speed possible; in the other session, the goal was quality of movement rather than speed, so the speed was to be slightly above the most recent 6MWT speed. During the entire intervention, participants maintained a treadmill speed that was substantially higher than the 6MWT speed; therefore, the 6MWT speed could serve as a parameter for setting the treadmill speed in BWSTT sessions. It was possible to combine the interventions and sustain a high level of intensity for a 12-week period without any negative side effects. All participants but 1 were motivated during the entire intervention and were able to endure the intensity.
Number 4
A limitation of the present study was the lack of a control group. However, as previously shown,7 functional and neurological performance reaches a steady state within the first 6 months after injury. Because the average time since injury was more than 2 years in the present study, it is highly unlikely that the observed gains in ambulation and cardiovascular health could be ascribed to a natural course of recovery. However, the results of the present study should be tested in a future randomized controlled trial.
Conclusion A high dose of intensive physical training for participants with stroke in the chronic stage (a combination of BWSTT, PRST, and AE 5 times per week for 1.5 hours per session for 12 weeks) increased walking speed on the 6MWT by 62%, regardless of chronicity, age, or level of functioning. Weekly testing of the walking speed revealed an almost linear progression during the entire intervention, indicating that an undetected and dormant plateau of recovery after stroke was reached for this group of participants. Further studies should investigate the duration of intervention needed to reach the full potential of recovery of gait. All authors provided concept/idea/research design. Mr Jørgensen, Mr Bech-Pedersen, Mr Zeeman, Dr Andersen, and Dr Scho¨nberger provided writing. Mr Jørgensen, Mr BechPedersen, Mr Zeeman, and Mrs Sørensen provided data collection. Dr Scho¨nberger provided data analysis. Mr Bech-Pedersen and Mrs Sørensen provided fund procurement and participants. Mr Zeeman and Mrs Sørensen provided facilities/equipment. Mrs Sørensen, Dr Andersen, and Dr Scho¨nberger provided consultation (including review of manuscript before submission). This study was approved by the local ethics committee (De Videnskabsetiske Komite´er for Københavns og Frederiksberg Kommuner, Copenhagen, Denmark; study approval no. KF-01-240/04).
April 2010
Intensive Physical Training After Stroke This article was received December 18, 2008, and was accepted November 16, 2009. DOI: 10.2522/ptj.20080404
References 1 Murray CJ, Lopez AD. Mortality by cause for eight regions of the world: Global Burden of Disease Study. Lancet. 1997;349: 1269 –1276. 2 Rosamond W, Flegal K, Furie K, et al. Heart disease and stroke statistics: 2008 update—a report from the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Circulation. 2008;117:e25– e146. 3 Danish National Board of Health. Reference Program for the Treatment of Stroke Patients [in Danish]. Copenhagen, Denmark: Danish National Board of Health; 2005. 4 Demain S, Wiles R, Roberts L, McPherson K. Recovery plateau following stroke: fact or fiction? Disabil Rehabil. 2006;28:815– 821. 5 Lee IM, Paffenbarger RS Jr. Physical activity and stroke incidence: the Harvard Alumni Health Study. Stroke. 1998; 29:2049 –2054. 6 Kwakkel G, van Peppen R, Wagenaar RC, et al. Effects of augmented exercise therapy time after stroke: a meta-analysis. Stroke. 2004;35:2529 –2539. 7 Jorgensen HS, Nakayama H, Raaschou HO, et al. Outcome and time course of recovery in stroke, part II: time course of recovery. The Copenhagen Stroke Study. Arch Phys Med Rehabil. 1995;76:406 – 412. 8 MacKay-Lyons J, Makrides L. Cardiovascular stress during a contemporary stroke rehabilitation program: is the intensity adequate to induce a training effect? Arch Phys Med Rehabil. 2002;83:1378 –1383. 9 Moseley AM, Stark A, Cameron ID, Pollock A. Treadmill training and body weight support for walking after stroke. Cochrane Database Syst Rev. 2005;4:CD002840. 10 Pohl M, Mehrholz J, Ritschel C, Ru ¨ ckriem S. Speed-dependent treadmill training in ambulatory hemiparetic stroke patients: a randomized controlled trial. Stroke. 2002; 33:553–558.
April 2010
11 Sullivan KJ, Knowlton BJ, Dobkin BH. Step training with body weight support: effect of treadmill speed and practice paradigms on poststroke locomotor recovery. Arch Phys Med Rehabil. 2002;83:683– 691. 12 Eich HJ, Mach H, Werner C, Hesse S. Aerobic treadmill plus Bobath walking training improves walking in subacute stroke: a randomized controlled trial. Clin Rehabil. 2004;18:640 – 651. 13 Bohannon RW. Muscle strength and muscle training after stroke. J Rehabil Med. 2007;39:14 –20. 14 Weiss A, Suzuki T, Bean J, Fielding RA. High intensity strength training improves strength and functional performance after stroke. Am J Phys Med Rehabil. 2000;79: 369 –376. 15 Bohannon RW, Walsh S. Nature, reliability, and predictive value of muscle performance measures in people with hemiparesis following stroke. Arch Phys Med Rehabil. 1992;73:721–725. 16 Morris SL, Dodd KJ, Morris ME. Outcomes of progressive resistance strength training following stroke: a systematic review. Clin Rehabil. 2004;18:27–39. 17 Patten C, Lexell J, Brown HE. Weakness and strength training in people with poststroke hemiplegia: rationale, method, and efficacy. J Rehabil Res Dev. 2004;41:293– 312. 18 Pang MY, Eng JJ, Dawson AS, Gylfadottir S. The use of aerobic exercise training in improving aerobic capacity in individuals with stroke: a meta-analysis. Clin Rehabil. 2006;20:97–111. 19 Saunders DH, Greig CA, Mead GE, Young A. Physical fitness training for stroke patients. Cochrane Database Syst Rev. 2009; (4):CD003316. 20 Teixeira-Salmela LF, Nadeau S, McBride I, Olney SJ. Effects of muscle strengthening and physical conditioning training on temporal, kinematic and kinetic variables during gait in people with chronic stroke. J Rehabil Med. 2001;33:53– 60. 21 Sullivan KJ, Brown DA, Klassen T, et al. Effects of task-specific locomotor and strength training in adults who were ambulatory after stroke: results of the STEPS randomized clinical trial. Phys Ther. 2007; 87:1580 –1602. 22 Enright PL, Sherrill DL. Reference equations for the six-minute walk in healthy adults. Am J Respir Crit Care Med. 1998; 158:1384 –1387.
23 Guyatt GH, Sullivan PJ, Thompson PJ, et al. The six-minute walk: a new measure of exercise capacity in patients with chronic heart failure. Can Med Assoc J. 1985;132:919 –923. 24 Wade DT. Measurement in Neurological Rehabilitation. New York, NY: Oxford University Press; 1992. 25 Åstrand PO, Rodahl K. Textbook of Work Physiology: Physiological Bases of Exercise. 3rd ed. New York, NY: McGraw-Hill Book Co; 1986. 26 Åstrand PO, Rhyming I. A nomogram for calculation of aerobic capacity (physical fitness) from pulse rate during submaximal work. J Appl Physiol. 1954; 7:218 –221. 27 Borg G. A Simple Rating Scale for Use in Physical Work Tests [in Swedish]. Lund, Sweden: University of Lund; 1962:7–15. 28 GPOWER: a priori, post-hoc and compromise power analysis for MS DOS [computer program]. Bonn, Germany: Bonn University; 1992. 29 Ada L, Dean CM, Hall JM, et al. A treadmill and overground walking program improves walking in persons residing in the community after stroke: a placebocontrolled, randomized trial. Arch Phys Med Rehabil. 2003;84:1486 –1491. 30 Peurala SH, Tarkka IM, Pitka¨nen K, Sivenius J. The effectiveness of body weight– supported gait training and floor walking in patients with chronic stroke. Arch Phys Med Rehabil. 2005;86:1557–1564. 31 Ouellette MM, LeBrasseur NK, Bean JF, et al. High-intensity resistance training improves muscle strength, self-reported function, and disability in long-term stroke survivors. Stroke. 2004;35:1404 –1409. 32 Lewington S, Clarke R, Qizilbash N, et al. Age-specific relevance of usual blood pressure to vascular mortality: a meta-analysis of individual data for one million adults in 61 prospective studies. Lancet. 2002;360: 1903–1913. 33 Rimmer JH, Rauworth AE, Wang EC, et al. A preliminary study to examine the effects of aerobic and therapeutic (nonaerobic) exercise on cardiorespiratory fitness and coronary risk reduction in stroke survivors. Arch Phys Med Rehabil. 2009;90: 407– 412.
Volume 90
Number 4
Physical Therapy f
537
Research Report Muscle Activation and Perceived Loading During Rehabilitation Exercises: Comparison of Dumbbells and Elastic Resistance Lars L. Andersen, Christoffer H. Andersen, Ole S. Mortensen, Otto M. Poulsen, Inger Birthe T. Bjørnlund, Mette K. Zebis L.L. Andersen, PhD, is Researcher, National Research Centre for the Working Environment, Lersø Parkalle 105, DK 2100 Copenhagen, Denmark. Address all correspondence to Dr Andersen at: LLA@ NRCWE.DK. C.H. Andersen, MSc, is a PhD student at National Research Centre for the Working Environment. O.S. Mortensen, MD, PhD, is Researcher, National Research Centre for the Working Environment, and Department of Occupational and Environmental Medicine, Bispedjerg University Hospital, Copenhagen, Denmark. O.M. Poulsen, DrMedSci, is Director of Research, National Research Centre for the Working Environment. I.B.T. Bjørnlund, BSc (Physiotherapy), is Teacher in Physiotherapy, Faculty of Physiotherapy, Metropolitan University College, Copenhagen, Denmark. M.K. Zebis, PhD, is Researcher, National Research Centre for the Working Environment.
Background. High-intensity resistance training plays an essential role in the prevention and rehabilitation of musculoskeletal injuries and disorders. Although resistance exercises with heavy weights yield high levels of muscle activation, the efficacy of more user-friendly forms of exercise needs to be examined. Objective. The aim of this study was to investigate muscle activation and perceived loading during upper-extremity resistance exercises with dumbbells compared with elastic tubing.
Design. A single-group, repeated-measures study design was used. Setting. Exercise evaluation was conducted in a laboratory setting. Participants. Sixteen female workers (aged 26 –55 years) without serious musculoskeletal diseases and with a mean neck and shoulder pain intensity of 7.8 on a 100-mm visual analog scale participated in the study. Measurements. Electromyographic (EMG) activity was measured in 5 selected muscles during the exercises of lateral raise, wrist extension, and shoulder external rotation during graded loadings with dumbbells (2–7.5 kg) and elastic tubing (TheraBand, red to silver resistance). The order of exercises and loadings was randomized for each individual. Electromyographic amplitude was normalized to the absolute maximum EMG amplitude obtained during maximal voluntary isometric contraction and exercise testing. Immediately after each set of exercise, the Borg CR10 scale was used to rate perceived loading during the exercise. Results. Resistance exercise with dumbbells as well as elastic tubing showed increas-
[Andersen LL, Andersen CH, Mortensen OS, et al. Muscle activation and perceived loading during rehabilitation exercises: comparison of dumbbells and elastic resistance. Phys Ther. 2010;90:538 –549.]
ing EMG amplitude and perceived loading with increasing resistance. At the individually maximal level of resistance for each exercise— defined as the 3 repetitions maximum— normalized EMG activity of the prime muscles was not significantly different between dumbbells (59%– 87%) and elastic tubing (64%– 86%). Perceived loading was moderately to very strongly related to normalized EMG activity (r⫽.59 –.92).
© 2010 American Physical Therapy Association
Limitations. The results of this study apply only for exercises performed in a controlled manner (ie, without sudden jerks or high acceleration).
Post a Rapid Response or find The Bottom Line: www.ptjournal.org 538
f
Physical Therapy
Conclusions. Comparably high levels of muscle activation were obtained during resistance exercises with dumbbells and elastic tubing, indicating that therapists can choose either type in clinical practice. The Borg CR10 can be a useful aid in estimating intensity of individual rehabilitation protocols.
Volume 90
Number 4
April 2010
Electromyographic Comparison of Exercises With Dumbbells and Elastic Tubing
M
ore than half a century ago DeLorme1 recommended progressive resistance training for rehabilitation of injured servicemen. Today, high-intensity resistance training has become an essential part of various rehabilitation protocols.2 For instance, resistance training is used effectively in rehabilitation of work-related neck and shoulder pain,3,4 rotator cuff injury,5,6 Achilles tendinopathy,7 poststroke hemiplegia,8 and postoperative weakness in elderly patients.9 A key ingredient of strengthening protocols is training intensity, defined as the percentage of maximal voluntary force exerted. Electromyography (EMG) is commonly used to measure the level of muscle activation and provides a rough estimate of exercise intensity for specific muscles involved in the movement.10 –15 Training intensities of 60% and higher generally are recommended to obtain the desired physiological adaptations.16 To yield high levels of muscle activation, resistance training usually is performed on machines or with free weights.10,17 In clinical practice and for home-based rehabilitation, conventional resistance training devices may not always be feasible. Thus, the effectiveness of alternative exercise methods should be investigated. Strengthening exercises with elastic resistance have been shown to be a feasible alternative to heavy weights in certain situations.18,19 The material properties of commercially available elastic tubing theoretically allows for efficient resistance exercise.20 However, although some studies have demonstrated high levels of muscle activation for specific muscles using elastic resistance,13,15 other studies have shown low to medium levels of activation for most of the involved muscles.15,21–23 Overall, these studies indicate practical difficulties in
April 2010
determining the appropriate exercise intensity using elastic resistance. Thus, there is a need to further investigate whether high levels of muscle activation similar to those achieved with traditional devices, such as dumbbells, can be obtained with elastic resistance devices. Self-selected loadings in resistance training generally are lower than recommended— below 60% of the maximal dynamic load.24 –27 Although the repetitions maximum test is recommended to determine training intensity,28 this test is performed to local muscle exhaustion, which may be inconvenient in patients with pain who are undergoing rehabilitation. Patient-report rating scales of exercise intensity may be more appropriate in clinical practice. The Borg CR10 scale has been widely used for rating the perceived intensity of various physiological experiences, such as physical exertion.29 Thus, it would be relevant to investigate whether perceived loading rated on the Borg CR10 scale is related to the level of muscle activation. The aim of the present study was to investigate the level of muscle activation (EMG) and perceived loading (Borg CR10 scale) during graded rehabilitation exercises using dumbbells in comparison with elastic resistance. We hypothesized that the levels of muscle activation and perceived loading are similar when comparing elastic resistance with dumbbells. Furthermore, we hypothesized that perceived loading is related to the level of muscle activation.
Method Participants The study was performed in Copenhagen, Denmark. A group of 16 women (41⫾9.6 years; 168⫾4.9 cm, 64.5⫾11.0 kg) with primarily sedentary jobs (office workers, laboratory technicians) were recruited on a
voluntary basis for the study. Exclusion criteria were clinically assessed subacromial impingement syndrome, anamnestic history of disk prolapse, rheumatoid arthritis, or other serious musculoskeletal disorders. None of the recruited participants met these exclusion criteria. Musculoskeletal pain (100-mm visual analog scale) during the last 3 months was 7.8⫾ 19 mm (neck/shoulder), 5.4⫾15 mm (forearm), and 11⫾17 mm (low back) (mean ⫾ SD). Complete testing was performed on all 16 participants with both elastic tubing and dumbbells during the exercises described below. All participants were informed about the purpose and content of the project and gave written informed consent to participate in the study, which conformed to the Declaration of Helsinki and was approved by the local ethical committee (HC-2008-103). Maximal Voluntary Isometric Contraction Prior to the dynamic exercises described below, maximal voluntary isometric contractions (MVICs) were performed according to standardized procedures during neck extension, shoulder abduction, shoulder external rotation, and wrist extension to induce a maximal EMG response of the respective muscles.30 Two MVICs were performed for each muscle, and the trial with the higher EMG response was used for normalization
Available With This Article at ptjournal.apta.org • The Bottom Line clinical summary • The Bottom Line Podcast • Audio Abstracts Podcast This article was published ahead of print on February 4, 2010, at ptjournal.apta.org.
Volume 90
Number 4
Physical Therapy f
539
Electromyographic Comparison of Exercises With Dumbbells and Elastic Tubing Table 1. Comparison of Force Levels in Kilograms Between the Thera-Band Elastic Tubing (Range of 125%–150% Elongation of Resting Length) and Dumbbells Used in the Present Studya Elastic Tubing Range
Dumbbells
Red
2.0–2.2
2.0
Green
2.6–3.0
3.0
Blue
3.7–4.1
4.0
Black
5.0–5.6
5.0
Blue⫹red
5.7–6.3
6.25
Silver
6.9–7.8
7.5
Color
a
Thera-Band values are provided by the manufacturer.
of the peak EMG amplitude in the rehabilitation exercises. Participants were instructed to gradually increase muscle contraction force toward maximum over a period of 2 seconds, sustain the MVIC for 3 seconds, and slowly release the force. Verbal encouragement was given during all trials. Exercise Equipment Thera-Band elastic tubing* of different resistances (red, green, blue, black, and silver) were used. Handleto-handle length of the elastic tubing was individually adjusted according to the following formula: Individual height minus 10 cm for the lateral raise and external rotation exercises. For the wrist extension exercise, the length of the tubing was set to half of the table height. During pilot testing, we found rather large differences in muscle activation and perceived loading between black and gray resistances. Thus, an intermediate resistance consisting of combined blue and red tubing was made for the further experiments. Thus, a total of 6 resistance levels were used. The material properties of Thera-Band elastic tubing have been described previously.20,31 Correspondingly, standard iron dumbbells of 2, 3, 4, 5, 6.25, and 7.5 kg were used. A comparison of * The Hygenic Corp, 1245 Home Ave, Akron, OH 44310-2575.
540
f
Physical Therapy
Volume 90
resistance in kilograms between the dumbbells and elastic tubing used in the present study is given in Table 1. Exercise Description All exercises were performed in a slowly controlled manner—lifting (⬃1.5 seconds) and lowering (⬃1.5 seconds) without sudden jerks or acceleration—for 3 consecutive repetitions. The order of exercises and loadings was randomized for each participant, and each set of exercise was initiated every 1.5 minutes. Participants were familiarized with the exercises on a separate day prior to testing. Three common rehabilitation exercises were chosen; one with a large range of motion (lateral raise), one with a small range of motion (wrist extension), and one involving joint rotation (external rotation). In the lateral raise exercise (Fig. 1, left), the participants stood erect holding the dumbbells or tubing handles to the side and abducted the shoulder joints until the upper arms were slightly above horizontal. The elbows were in a static, slightly flexed position (⬃5°) during the entire range of motion. During this exercise, the elastic tubing was stretched to slightly more than twice its resting length (⬃125%–150% more than resting length).
Number 4
In the wrist extension exercise (Fig. 1, middle), the participants rested their forearm on a table while holding the dumbbell or tubing handle in the same hand using a pronated grip. The elastic tubing was prestretched to twice its resting length. The starting position was from a flexed wrist. The participants then performed a wrist extension through a full range of motion. In the external rotation exercise using elastic tubing (Fig. 1, upper right), the participants stood erect while holding the elbow at 90 degrees, close to the side. The starting position was with the forearm in front of the body, and the elastic tubing was parallel to the frontal plane. The elastic tubing was attached to a door handle and prestretched to twice its resting length. The participants then performed an external rotation through a full range of motion. In the external rotation exercise using a dumbbell (Fig. 1, lower right), the participants lay on the nondominant side of the body while holding the dominant elbow at a 90-degree angle, close to the side. The starting position was with the forearm in front of the body. The participants then performed an external rotation through a full range of motion. Perceived Loading Immediately after each set of exercise, the Borg CR10 scale29 was used to rate perceived loading of the respective muscle groups during the exercise. The meaning of the scale was carefully explained to each individual prior to testing. EMG Signal Sampling and Analysis Electromyography signals were recorded from the mid-portion of the splenius capitis, upper trapezius, medial deltoid, infraspinatus, and extensor digitorum communis muscles. A April 2010
Electromyographic Comparison of Exercises With Dumbbells and Elastic Tubing
Figure 1. Illustration of the resistance exercises with elastic tubing (top) and dumbbells (bottom). The exercises are lateral raise (left), wrist extension (middle), and external rotation (right) exercises.
bipolar surface EMG configuration (Neuroline 720 01-K†) and an interelectrode distance of 2 cm were used. Before affixing the electrodes, the skin of the respective area was prepared with scrubbing gel (Acqua gel‡) to effectively lower the impedance to less than 10 k⍀. Electrode placement followed the SENIAM recommendations.32 The EMG electrodes were connected directly to small preamplifiers located near the recording site. The raw EMG signals were led through shielded wires to instrumental differentiation ampli-
fiers, with a bandwidth of 10 to 500 Hz and a common mode rejection ratio better than 100 dB, sampled at 1,000 Hz using a 16-bit A/Dconverter (DAQ Card-Al-16XE-50§) and recorded on computer via a laboratory interface (CED 1401, Spike2 software㛳). Representative samplings of raw EMG signals from one of the participants during the lateral raise exercise with elastic tubing and dumbbells, respectively, are shown in Figure 2.
§ †
Medicotest A/S, Rugmarken 10, 3650 Ølstykke, Denmark. ‡ Meditec SRL, Via Micheli 9 S. Polo Di Torrile, 43056 Torrile, Parma, Italy.
April 2010
National Instruments Corp, 11500 N Mopac Expwy, Austin, TX 78759-3504 㛳 Cambridge Electronic Design Ltd, Unit 4, Science Park, Milton Rd, Cambridge, CB4 0FE United Kingdom.
During later analysis, all raw EMG signals obtained during MVICs and during the dumbbell and elastic tubing exercises were digitally filtered, consisting of: (1) high-pass filtering at 10 Hz33 and (2) a moving rootmean-square (RMS) filter of 500 milliseconds. For each individual muscle, peak RMS EMG amplitude of the 3 repetitions performed at each level was determined, and the average value of these 3 repetitions was normalized to the absolute maximum EMG amplitude obtained during maximal voluntary isometric contraction and exercise testing. High levels of muscle activation were defined in the present study as normalized EMG amplitude above 60%.10,16
Volume 90
Number 4
Physical Therapy f
541
Electromyographic Comparison of Exercises With Dumbbells and Elastic Tubing
Figure 2. Raw recording of electromyography (EMG) signals during 3 repetitions of lateral raise exercise with elastic tubing (left) and dumbbells (right) in the trapezius (top), medial deltoid (middle), and splenius capitis (bottom) muscles. The root-mean-square EMG recording is overlaid (yellow tracing) on the raw EMG recording.
542
f
Physical Therapy
Volume 90
Number 4
April 2010
April 2010
a
2.7⫾0.3
3.8⫾0.4
4.6ⴞ0.5
7.0⫾0.7
8.2⫾0.7
Green/3 kg
Blue/4 kg
Black/5 kg
Blue⫹red/ 6.25 kg
Silver/7.5 kg
1.7⫾0.2
Volume 90
Silver/7.5 kg
Number 4
Medial Deltoid Muscle
Splenius Capitis Muscle
8.3⫾0.7
Silver/7.5 kg
2.0⫾0.2
8.8⫾0.7
7.6⫾0.7
7.2⫾0.7
5.5ⴞ0.6
3.3⫾0.3
18⫾3.7
15⫾2.7
13ⴞ2.4
9⫾1.4
7⫾1.1
5⫾0.7
8ⴞ1.3
8⫾1.5
6⫾0.9
6⫾1.1
5⫾1.0
5⫾0.8
75⫾3.6
78⫾3.3
83ⴞ3.1
75⫾3.4
63⫾3.6
53⫾3.2
16⫾2.4
14⫾2.7
14⫾2.5
12ⴞ2.1
8⫾1.1
5⫾0.7
16⫾2.2
10ⴞ2.0
7⫾1.2
7⫾1.2
6⫾1.1
5⫾0.9
84⫾4.0
83⫾3.0
84ⴞ3.0
72⫾3.2
60⫾3.2
46⫾3.2
35⫾4.5
30⫾4.1
21ⴞ3.5
10⫾1.7
6⫾0.9
5⫾0.9
7ⴞ1.3
6⫾1.0
5⫾0.7
4⫾0.8
4⫾0.6
3⫾0.4
73⫾2.7
77⫾3.2
82ⴞ3.0
81⫾2.6
72⫾3.0
66⫾3.1
45⫾4.3
41⫾4.4
35⫾3.4
29ⴞ4.2
14⫾2.7
9⫾1.3
12⫾1.7
7ⴞ0.8
6⫾1.2
5⫾0.9
4⫾1.2
3⫾0.5
80⫾2.0
83⫾3.2
88ⴞ2.5
80⫾2.5
70⫾2.8
63⫾2.8
27⫾2.8
19⫾2.0
13ⴞ1.9
10⫾1.8
9⫾1.7
8⫾2.1
16ⴞ2.2
13⫾2.1
11⫾2.3
8⫾1.2
7⫾1.1
6⫾0.8
62⫾3.9
63⫾4.9
58ⴞ4.6
47⫾4.2
35⫾4.0
24⫾3.0
Values are expressed as mean ⫾ standard error. The group median resistance at “max” is marked in bold for each exercise.
7.2⫾0.8
Blue⫹red/ 6.25 kg
3.7⫾0.4
5.2ⴞ0.5
Blue/4 kg
Black/5 kg
1.5⫾0.2
2.5⫾0.2
Red/2 kg
8.3⫾0.7
6.5ⴞ0.7
Green/3 kg
External rotation
5.5⫾0.5
6.5ⴞ0.6
5.0⫾0.6
3.7⫾0.4
2.5⫾0.3
1.9⫾0.2
8.5⫾0.6
6.9⫾0.6
5.2ⴞ0.4
4.2⫾0.4
2.9⫾0.3
1.7⫾0.2
5.5⫾0.6
Black/5 kg
Blue⫹red/ 6.25 kg
2.3⫾0.2
3.5⫾0.4
Green/3 kg
Blue/4 kg
1.7⫾0.3
Red/2 kg
Wrist extension
Trapezius Muscle
Extensor Digitorum Communis Muscle
Infraspinatius Muscle
52⫾4.1 56⫾4.2
35⫾3.7
43ⴞ2.9
32⫾3.2
27⫾3.1
22⫾2.9
78ⴞ3.6
76⫾4.3
70⫾3.8
69⫾4.3
61⫾5.2
55⫾4.6
59⫾4.0
56⫾3.8
50ⴞ2.5
38⫾2.8
29⫾1.8
27⫾2.6
33⫾3.3
30⫾4.4
25ⴞ2.2
20⫾2.2
17⫾2.7
21⫾3.0
16ⴞ2.4
11⫾2.0
8⫾1.5
6⫾0.9
6⫾0.7
69⫾4.3
66⫾3.6
62ⴞ5.3
44⫾3.9
31⫾3.9
21⫾2.8
63⫾5.2
62⫾5.3
58⫾4.5
49ⴞ5.1
35⫾4.3
24⫾2.4
80⫾4.5
81ⴞ4.0
80⫾4.5
74⫾4.7
66⫾4.5
56⫾6.0
69⫾4.7
65⫾4.4
61ⴞ5.4
45⫾3.7
32⫾3.0
22⫾2.3
75⫾4.2
71⫾4.1
73ⴞ3.6
59⫾4.1
44⫾4.7
36⫾4.4
24ⴞ2.6
19⫾2.5
14⫾1.7
11⫾1.5
10⫾1.6
5⫾0.5
49⫾4.7
45⫾4.7
36ⴞ4.0
31⫾4.0
26⫾4.2
19⫾3.8
72⫾3.1
70⫾3.9
69⫾3.9
68ⴞ2.4
55⫾3.4
41⫾4.5
32⫾2.5
22ⴞ2.3
19⫾2.5
12⫾1.4
8⫾1.1
4⫾0.5
51⫾4.6
45⫾5.2
42ⴞ4.9
32⫾4.5
23⫾4.0
17⫾4.3
Elastic Tubing Dumbbells Elastic Tubing Dumbbells Elastic Tubing Dumbbells Elastic Tubing Dumbbells Elastic Tubing Dumbbells Elastic Tubing Dumbbells
Red/2 kg
Lateral raise
Exercise
Perceived Loading
Normalized EMG Amplitude (%)
Perceived Loading (Borg CR10 Scale) and Muscle Activation (Normalized Electromyography [EMG] Amplitude) for the Splenius Capitis, Trapezius, Medial Deltoid, Infraspinatus, and Extensor Digitorum Communis Muscles at Each Absolute Resistance Level With Elastic Tubing (Red-Silver) and Dumbbells (2–7.5 kg) During Lateral Raise, Wrist Extension, and External Rotation Exercisesa
Table 2.
Electromyographic Comparison of Exercises With Dumbbells and Elastic Tubing
Physical Therapy f
543
Electromyographic Comparison of Exercises With Dumbbells and Elastic Tubing Data Analysis A 2-way (2 ⫻ 6) repeated-measures analysis of variance (ANOVA) (SAS version 9#) was used to determine whether differences existed between dumbbells and elastic tubing. Factors included in the model were type (dumbbells and elastic tubing) and resistance (6 resistance levels for dumbbells and elastic tubing, respectively), as well as type ⫻ resistance interaction. Dependent variables were perceived loading (Borg CR10 scale) and muscle activation (normalized EMG amplitude for each of the 5 muscles) (ie, 6 possible outcomes for each of the 3 exercises). To avoid mass significance due to multiple primary analyses, only 8 preplanned ANOVAs were performed—Borg CR10 scale for all 3 exercises; EMG activity of the upper trapezius, splenius, and medial deltoid muscles for the lateral raise exercise; EMG activity of extensor digitorum communis muscle for the wrist extension exercise; and EMG activity of the infraspinatus muscle during shoulder external rotation. Furthermore, the critical P value of the primary analyses was set to .01. When a significant main effect was found, post hoc comparisons were made to locate differences. Results are reported for both absolute (Tab. 2) and relative (Fig. 3) levels of resistance. Absolute levels of resistance refer to the color of the tubing and weight of the dumbbell. For determination of the relative level of resistance, the highest voluntary resistance of each participant—the 3 repetitions maximum (RM)—was set to level 0 (denoted “max” in Fig. 3). Likewise, each decrement and increment, respectively, in relative resistance level corresponded to a lower (denoted “submax” in Fig. 3) and higher (denoted “supramax” in Fig. 3) resistance in either dumbbell
weight or color of the tubing. To avoid mass significance of post hoc tests, the critical P value was set to .01, and all values are reported as group means ⫾ standard error unless otherwise stated. Finally, Spearman correlation coefficients were calculated to determine the relationship among: (1) perceived loading and EMG activity, (2) actual loading and EMG activity, and (3) actual and perceived loading (submaximal and maximal loadings) (Tab. 3). The strength of the relationship was defined as very weak (r⫽.0 –.2), weak (r⫽.2–.4), moderate (r⫽.4 –.7), strong (r⫽.7–.9), or very strong (r⫽.9 –1.0). A difference of less than 10% in normalized EMG amplitude between dumbbells and elastic tubing was considered clinically insignificant. This value was based on general strength training literature, where recommendations often are given in increments of 10 percentage points (eg, “it is recommended that novice to intermediate individuals train with 60 –70% of 1 RM”16(p690)). A priori power analysis showed that 16 participants in this paired design were sufficient to obtain a statistical power of 80% at a minimal relevant difference of 10% and a type I error probability of 1%, assuming a standard deviation of 10% based on previous research in our laboratory.10
Results
Group mean ⫾ standard error values at absolute resistance levels are presented in Table 2, and group mean ⫾ standard error values at relative resistance levels are shown in Figure 3. Note that “absolute resistance” refers to the color of the tubing and weight of the dumbbell, whereas “relative resistance” refers to the resistance level relative to “max.” Thus, the group mean values of Table 2 and Figure 3 can differ.
# SAS Institute Inc, PI Box 8000, Cary, NC 27513
544
f
Physical Therapy
Volume 90
Number 4
Perceived Loading (Borg CR10 Scale) A priori hypothesis testing of main effects showed a significant resistance effect for perceived loading in the lateral raise (F⫽32, P⬍.0001), wrist extension (F⫽49, P⬍.0001), and external rotation (F⫽38, P⬍ .0001) exercises (ie, perceived loading increased with increasing resistance during all 3 exercises with both dumbbells and elastic tubing). Furthermore, a trend toward a type effect was observed for external rotation only (F⫽6.2, P⫽.04) (ie, perceived loading during elastic tubing tended to be lower compared with dumbbells [⌬ Borg CR10 scale score⫽0.98⫾0.40, P⫽.04]). The type ⫻ resistance interaction was not significant for any of the examined exercises. Perceived loading at the individual “max” level with dumbbells versus elastic tubing, respectively, was not significantly different (5.5⫾0.6 versus 5.2⫾0.7 for lateral raise, 6.4⫾0.7 versus 6.5⫾0.8 for wrist extension, and 5.2⫾0.6 versus 5.2⫾0.7 for external rotation) (Fig. 3A). Muscle Activation (EMG) A priori hypothesis testing of main effects showed a significant resistance effect for EMG activity of the upper trapezius muscle (F⫽20, P⬍ .0001), medial deltoid muscle (F⫽ 10, P⬍.0001), and splenius capitis muscle (F⫽37, P⬍.0001) during the lateral raise exercise; for EMG activity of the extensor digitorum communis muscle (F⫽19, P⬍.0001) during the wrist extension exercise; and for EMG activity of the infraspinatus muscle (F⫽38, P⬍.0001) during the external rotation exercise. Thus, normalized EMG amplitude of the prime muscles generally increased with increasing absolute resistance (Tab. 1) and relative resistance (Figs. 3B–F) for both dumbbells and elastic tubing during all 3 exercises. There were no significant type effects for any of the prime muscles (ie, no difApril 2010
Electromyographic Comparison of Exercises With Dumbbells and Elastic Tubing LAT - Elastic tubing LAT - Dumbbells WR E - Elastic tubing WR E - Dumbbells EXR - Elastic tubing EXR - Dumbbells
A
Normalized EMG Amplitude (%)
Perceived Loading
Borg CR10 Scale Score
10 9 8 7 6 5 4 3 2 1 0 -5
-4 -3 -2 submax
-1
0 max
1
B
Trapezius Muscle
100 90 80 70 60 50 40 30 20 10 0 -5
2 3 4 supramax
-4 -3 -2 submax
Relative Resistance Level
Medial Deltoid Muscle
100 90 80 70 60 50 40 30 20 10 0 -5
-4
-3 -2 -1 0 1 submax max Relative Resistance Level
D
80 70 60 50 40 30 20 10 0 2 3 4 supramax
Splenius Capitis Muscle
80 70 60 50 40 30 20 10 0 -5
-4
F Normalized EMG Amplitude (%)
Normalized EMG Amplitude (%)
90
-3 -2 -1 0 1 submax max Relative Resistance Level
2 3 4 supramax
90
E Extensor Digitorum Communis Muscle
-4
1
100
2 3 4 supramax
100
-5
0 max
Relative Resistance Level
Normalized EMG Amplitude (%)
Normalized EMG Amplitude (%)
C
-1
-3 -2 -1 0 1 2 3 4 submax max supramax Relative Resistance Level
Infraspinatus Muscle
100 90 80 70 60 50 40 30 20 10 0 -5
-4
-3 -2 -1 0 1 2 3 4 submax max supramax Relative Resistance Level
Figure 3. Perceived loading rated on the Borg CR10 scale (A) and normalized electromyography (EMG) amplitude of the trapezius (B), medial deltoid (C), splenius capitis (D), extensor digitorum communis (E), and infraspinatus (F) muscles during the different exercises and relative loadings with dumbbells (open marks) and elastic tubing (blue marks). LAT⫽lateral raise exercise, WRE⫽wrist extension exercise, EXR⫽external rotation exercise, submax⫽submaximal, max⫽maximal, and supramax⫽supramaximal.
April 2010
Volume 90
Number 4
Physical Therapy f
545
Electromyographic Comparison of Exercises With Dumbbells and Elastic Tubing Table 3. Spearman Correlation Coefficient Among Perceived Loading, Actual Loading, and Normalized Electromyography (EMG) Amplitude of the Prime Muscles During Lateral Raise, Wrist Extension, and External Rotation Exercisesa Lateral Raise
Measure
a
Wrist Extension
Trapezius Muscle
Medial Deltoid Muscle
Actual Loading
Extensor Digitorum Communis Muscle
.76⫾.09
.65⫾.12
.83⫾.09
.59⫾.10
.92⫾.02
.89⫾.05
.58⫾.18
.96⫾.02
External Rotation
Actual Loading
Infraspinatus Muscle
Actual Loading
.79⫾.09
.85⫾.07
.92⫾.04
.99⫾.01
Perceived loading
Elastic tubing Dumbbells
.83⫾.05
.65⫾.15
Actual loading
Elastic tubing
.93⫾.04
.75⫾.07
.70⫾.10
.93⫾.05
Dumbbells
.94⫾.03
.78⫾.12
.71⫾.16
.83⫾.11
Values are expressed as mean ⫾ standard error.
ference between dumbbells and elastic resistance). The type ⫻ resistance interaction was not significant for any of the examined exercises. Normalized EMG amplitude at the individual “max” level with dumbbells versus elastic tubing, respectively, was 86%⫾2.4% versus 86%⫾1.8% for the upper trapezius muscle (Fig. 3B), 87%⫾1.9% versus 85%⫾3.0% for the medial deltoid muscle (Fig. 3C), and 59%⫾4.6% versus 64%⫾5.4% for the splenius capitis muscle during the lateral raise exercise (Fig. 3D). During the wrist rotation exercise, the normalized EMG amplitude was 83%⫾3.6% versus 76%⫾3.8% for the extensor digitorum communis muscle (Fig. 3E) and 63%⫾3.1% versus 68%⫾3.5% for the infraspinatus muscle during external rotation exercise (Fig. 3F). Relationship Among Main Variables There was a moderate to very strong relationship among perceived loading (Borg CR10 scale), actual loading, and normalized EMG amplitude of the prime muscles (Tab. 2).
Discussion The main finding of this study was the comparable high level of muscle activation during resistance exercise with elastic tubing and dumbbells, 546
f
Physical Therapy
Volume 90
indicating that both types of exercise can be used equally during rehabilitation. Perceived loading and the level of muscle activation increased with increased external resistance. The practical relevance of these results is discussed below. Relevance for Neck and Shoulder Rehabilitation Approximately half of female office workers reports frequent trouble in the neck and shoulder area,34 which often is paralleled by tenderness and tightness of the upper trapezius muscle.35 Exercises to target the upper trapezius muscle are essential in rehabilitation of work-related neck and shoulder muscle pain. We have previously reported marked reductions of pain symptoms in women with trapezius muscle myalgia in response to high-intensity specific strength training.4,10 In that study, several exercises—lateral raise and shrugs— yielded high levels of trapezius muscle activation. However, the lateral raise exercise required only one fifth of the nominal load used during the shrug exercise, making it more practical. The present study elaborated on these findings by showing similarly high levels of trapezius muscle activation using elastic tubing compared with dumbbells. High levels of trapezius muscle activation were seen at relative resis-
Number 4
tance level ⫺2 and higher (Fig. 3B). For example, if one individual can perform a 3 RM (“max”) with black elastic tubing, then both blue (level ⫺1) and green (level ⫺2) resistances induce sufficiently high levels of trapezius muscle activation. Based on these results, we suggest individual neck and shoulder rehabilitation protocols to be initiated at 2 levels below the 3 RM, corresponding to a perceived loading of approximately 3 on the Borg CR10 scale, and then gradually progress toward higher loads. Interestingly, supramaximal loads did not appear to further facilitate activation of the trapezius muscle (Fig. 3B), likely due to a shorter range of motion. Thus, the exercises should be executed in a controlled manner through a full range of motion. The splenius capitis muscle is one of the neck muscles involved in extension and rotation of the cervical spine. This muscle can be assessed by surface EMG activity in the posterior triangle of the neck in the space between the upper trapezius and sternocleidomastoid muscles. Isometric resistance training of the neck muscles has been shown to be effective in decreasing nonspecific neck pain.36 The lateral raise exercise— both with dumbbells and elastic tubing—induced fairly high levels of splenius capitis muscle activity April 2010
Electromyographic Comparison of Exercises With Dumbbells and Elastic Tubing (59%– 64% at 3 RM, Fig. 3D), despite a static neck position during movement of the shoulder joint. This requirement for static neck stabilization when performing exercises of the shoulder girdle at high intensities indicates that specific isometric neck exercises used in a previous study36 may be redundant. Future research involving indwelling electrodes should be performed to determine the level of muscle activation in the deep neck muscles during highintensity resistance exercises involving shoulder girdle movement compared with specific isometric neck exercises. Relevance for Rehabilitation of Forearm Muscle Pain Intensive use of a computer mouse and keyboard has been established as a risk factor for development of forearm pain.37 The extensor digitorum communis muscle is one of the major forearm muscles activated during computer work38,39 and is highly susceptible to fatigue.40 It is likely that the etiology of computerrelated tenderness of the forearm muscles is similar to that of trapezius muscle myalgia (ie, overload of lowthreshold motor units due to prolonged repetitive and monotonous work tasks41,42), indicating that these muscles may respond positively to specific resistance training. The present study showed equally high levels of extensor digitorum communis muscle activation during wrist extension exercises using dumbbells and elastic tubing (76%– 83% at 3 RM, Fig. 3E). High levels of muscle activation were obtained at relative resistance level ⫺4 and higher, corresponding to a perceived loading of 2 to 3 on the Borg CR10 scale, which may be a starting point for rehabilitation of tender forearm muscles. Relevance for Rotator Cuff Injuries Rotator cuff injuries are frequent in athletes in sports involving throwApril 2010
ing, physical rehabilitation—including high-intensity resistance training—is recommended as the primary treatment before surgery is considered.5 Although most of the rotator cuff muscles are covered by superficial muscles, the infraspinatus muscle can be assessed with surface EMG at the point below the posterior deltoid muscle lateral to the trapezius muscle. In the present study, relevant high levels of infraspinatus muscle activity were obtained with both dumbbells and elastic resistance (63%– 68% at 3 RM, Fig. 3F). Similar levels of infraspinatus muscle activity during external rotation using a specially built pulley system with an attached weight have been reported previously.43 High levels of muscle activation were obtained only at relative resistance level 0 and at supramaximal loadings, corresponding to a perceived loading of 5 and higher on the Borg CR10 scale. Thus, compared with the 2 other exercises of the present study, relatively high loadings are needed to obtain a high level of muscle activation during shoulder external rotation. Although very high intensities are not recommended during the initial phases of rehabilitation, such intensities may be necessary during later stages to ensure high levels of muscle activation. Perceived Loading Although EMG is commonly used in scientific experiments to analyze the level of muscle activation during specific exercises, only a few therapists have this opportunity. In addition, RM tests can be highly unpleasant and may be inappropriate in clinical practice. The present study showed that perceived loading rated on the Borg CR10 scale can be a helpful tool in determining the desired training intensity. For most of the investigated exercises, the 3 RM corresponded to 5 to 6 on the Borg CR10 scale. It should be noted that perceived loading was rated after 3 rep-
etitions and that several consecutive repetitions leading to muscular fatigue may cause different ratings. Thus, for comparison with the present results, perceived loading should be rated in the nonfatigued state after only a few repetitions of the full set. Methodological Considerations The present study showed clear similarities between dumbbells and elastic tubing with regard to muscle activation and perceived loading during graded resistance exercise. Although not specifically investigated in this study, some differences also may exist. Whereas dumbbells provide isotonic resistance, elastic resistance increases linearly with elongation of the material.20 Nevertheless, joint torque curves of elastic resistance training mimic isotonic training (eg, torque is increased similarly during shoulder abduction from 0° to 90° due to elongation of the material and increased lever arm length, respectively).31 Furthermore, it should be noted that all exercises were performed in a controlled manner and that differences between elastic tubing and dumbbells may exist during more explosive movements. Whereas the inertia of the dumbbell results in increased total moment of force during accelerative movements, the inertia of the elastic tubing is negligible. Thus, the results of the present study apply only for exercises performed according to general recommendations of basic strength training and rehabilitation (ie, in a controlled manner without sudden jerks or acceleration). Although some accommodation comes from prestretching of the elastic tubing, as few as 20 repetitions appear to stabilize the material.20 Because all elastic tubings were stretched several times during pilot testing in the present study, it is unlikely that the
Volume 90
Number 4
Physical Therapy f
547
Electromyographic Comparison of Exercises With Dumbbells and Elastic Tubing material properties were changed during actual testing. With the use of surface EMG, there is an inherent risk of cross talk from neighboring muscles. Even though a relatively short inter-electrode distance of 2 cm was used, it is possible that EMG recordings from the splenius and infraspinatus muscles may have been affected by surrounding muscles to some extent. To our knowledge, the optimal interelectrode distance for minimizing cross talk while retaining signal amplitude has not been determined for these particular muscles.
Conclusion Comparable high levels of muscle activation were obtained during resistance exercises with dumbbells and elastic tubing, indicating that therapists can choose either type in clinical practice. The Borg CR10 can be a useful aid in estimating the proper intensity of individual rehabilitation protocols. L.L. Andersen and M.K. Zebis provided concept/idea/research design and project management. All authors provided writing. C.H. Andersen and I.B.T. Bjørnlund provided data collection and participants. L.L. Andersen and C.H. Andersen provided data analysis. L.L. Andersen and O.M. Poulsen provided fund procurement. O.S. Mortensen and O.M. Poulsen provided facilities/equipment, institutional liaisons, and consultation (including review of manuscript before submission). This study was supported by The Danish Rheumatism Association (grant R68-A993) and The Hygenic Corporation (Akron, Ohio). Warm thanks to physical therapist students Annelie Svensson, Majbritt Preuss, and Brian Rasmussen for their valuable technical assistance. This article was received May 24, 2009, and was accepted October 25, 2009. DOI: 10.2522/ptj.20090167
References 1 DeLorme TL. Restoration of muscle power by heavy resistance exercises. J Bone Joint Surg Am. 1945;27:645– 667.
548
f
Physical Therapy
Volume 90
2 Taylor NF, Dodd KJ, Damiano DL. Progressive resistance exercise in physical therapy: a summary of systematic reviews. Phys Ther. 2005;85:1208 –1223. 3 Ylinen J. Physical exercises and functional rehabilitation for the management of chronic neck pain. Eura Medicophys. 2007;43:119 –132. 4 Andersen LL, Kjaer M, Sogaard K, et al. Effect of two contrasting types of physical exercise on chronic neck muscle pain. Arthritis Rheum. 2008;59:84 –91. 5 Braun S, Kokmeyer D, Millett PJ. Shoulder injuries in the throwing athlete. J Bone Joint Surg Am. 2009;91:966 –978. 6 Niederbracht Y, Shim AL, Sloniger MA, et al. Effects of a shoulder injury prevention strength training program on eccentric external rotator muscle strength and glenohumeral joint imbalance in female overhead activity athletes. J Strength Cond Res. 2008;22:140 –145. 7 Magnussen RA, Dunn WR, Thomson AB. Nonoperative treatment of midportion Achilles tendinopathy: a systematic review. Clin J Sport Med. 2009;19:54 – 64. 8 Pak S, Patten C. Strengthening to promote functional recovery poststroke: an evidencebased review. Top Stroke Rehabil. 2008; 15:177–199. 9 Suetta C, Andersen JL, Dalgas U, et al. Resistance training induces qualitative changes in muscle morphology, muscle architecture, and muscle function in elderly postoperative patients. J Appl Physiol. 2008; 105:180 –186. 10 Andersen LL, Kjaer M, Andersen CH, et al. Muscle activation during selected strength exercises in women with chronic neck muscle pain. Phys Ther. 2008;88:703–711. 11 Ballantyne BT, O’Hare SJ, Paschall JL, et al. Electromyographic activity of selected shoulder muscles in commonly used therapeutic exercises. Phys Ther. 1993;73: 668 – 677. 12 Bull ML, Freitas V, Vitti M, Rosa GJ. Electromyographic validation of the trapezius and serratus anterior muscles in the rowing and frontal-lateral cross, dumbbells exercises. Electromyogr Clin Neurophysiol. 2002;42:79 – 84. 13 Decker MJ, Hintermeister RA, Faber KJ, Hawkins RJ. Serratus anterior muscle activity during selected rehabilitation exercises. Am J Sports Med. 1999;27:784 – 791. 14 Ekstrom RA, Donatelli RA, Soderberg GL. Surface electromyographic analysis of exercises for the trapezius and serratus anterior muscles. J Orthop Sports Phys Ther. 2003;33:247–258. 15 Hintermeister RA, Lange GW, Schultheis JM, et al. Electromyographic activity and applied load during shoulder rehabilitation exercises using elastic resistance. Am J Sports Med. 1998;26:210 –220. 16 Ratamees NA, Alvar BA, Evetoch TK, et al. American College of Sports Medicine position stand: Progression models in resistance training for healthy adults. Med Sci Sports Exerc. 2009;41:687–708.
Number 4
17 Andersen LL, Magnusson SP, Nielsen M, et al. Neuromuscular activation in conventional therapeutic exercises and heavy resistance exercises: implications for rehabilitation. Phys Ther. 2006;86:683– 697. 18 Ribeiro F, Teixeira F, Brochado G, Oliveira J. Impact of low cost strength training of dorsi- and plantar flexors on balance and functional mobility in institutionalized elderly people. Geriatr Gerontol Int. 2009; 9:75– 80. 19 Colado JC, Triplett NT. Effects of a shortterm resistance program using elastic bands versus weight machines for sedentary middle-aged women. J Strength Cond Res. 2008;22:1441–1448. 20 Patterson RM, Stegink Jansen CW, Hogan HA, Nassif MD. Material properties of Thera-Band Tubing. Phys Ther. 2001;81: 1437–1445. 21 Burnett AF, Coleman JL, Netto KJ. An electromyographic comparison of neck conditioning exercises in healthy controls. J Strength Cond Res. 2008;22:447– 454. 22 Netto KJ, Burnett AF, Coleman JL. Neck exercises compared to muscle activation during aerial combat maneuvers. Aviat Space Environ Med. 2007;78:478 – 484. 23 Matheson JW, Kernozek TW, Fater DC, Davies GJ. Electromyographic activity and applied load during seated quadriceps exercises. Med Sci Sports Exerc. 2001;33: 1713–1725. 24 Ratamess NA, Faigenbaum AD, Hoffman JR, Kang J. Self-selected resistance training intensity in healthy women: the influence of a personal trainer. J Strength Cond Res. 2008;22:103–111. 25 Andersen LL, Jorgensen MB, Blangsted AK, et al. Randomized controlled intervention trial to relieve and prevent neck/shoulder pain. Med Sci Sports Exerc. 2008;40:983– 990. 26 Focht BC. Perceived exertion and training load during self-selected and imposedintensity resistance exercise in untrained women. J Strength Cond Res. 2007;21: 183–187. 27 Glass SC, Stanton DR. Self-selected resistance training intensity in novice weightlifters. J Strength Cond Res. 2004;18: 324 –327. 28 Fleck SJ, Kraemer WJ. Designing Resistance Training Programs. 3rd ed. Champaign, IL: Human Kinetics; 2003. 29 Borg G. Borg’s Perceived Exertion and Pain Scales. Champaign, IL: Human Kinetics; 1998. 30 Laursen B, Jensen BR, Nemeth G, Sjogaard G. A model predicting individual shoulder muscle forces based on relationship between electromyographic and 3D external forces in static position. J Biomech. 1998; 31:731–739. 31 Hughes CJ, Hurd K, Jones A, Sprigle S. Resistance properties of Thera-Band tubing during shoulder abduction exercise. J Orthop Sports Phys Ther. 1999;29: 413– 420. 32 SENIAM Web site. Available at: http://www. seniam.org. Accessed May 1, 2009.
April 2010
Electromyographic Comparison of Exercises With Dumbbells and Elastic Tubing 33 Winter DA. Biomechanics and Motor Control of Human Movement. 2nd ed. New York, NY: John Wiley & Sons Inc; 1990:11–50. 34 Blangsted AK, Sogaard K, Hansen EA, et al. One-year randomized controlled trial with different physical-activity programs to reduce musculoskeletal symptoms in the neck and shoulders among office workers. Scand J Work Environ Health. 2008;34: 55– 65. 35 Juul-Kristensen B, Kadefors R, Hansen K, et al. Clinical signs and physical function in neck and upper extremities among elderly female computer users: the NEW study. Eur J Appl Physiol. 2006;96: 136 –145. 36 Ylinen J, Takala EP, Nykanen M, et al. Active neck muscle training in the treatment of chronic neck pain in women: a randomized controlled trial. JAMA. 2003;289: 2509 –2516.
April 2010
37 Kryger AI, Andersen JH, Lassen CF, et al. Does computer use pose an occupational hazard for forearm pain; from the NUDATA study. Occup Environ Med. 2003;60:e14. 38 Thorn S, Forsman M, Hallbeck S. A comparison of muscular activity during single and double mouse clicks. Eur J Appl Physiol. 2005;94:158 –167. 39 Sogaard K, Sjogaard G, Finsen L, et al. Motor unit activity during stereotyped finger tasks and computer mouse work. J Electromyogr Kinesiol. 2001;11: 197–206. 40 Lin MI, Liang HW, Lin KH, Hwang YH. Electromyographical assessment on muscular fatigue: an elaboration upon repetitive typing activity. J Electromyogr Kinesiol. 2004;14:661– 669.
41 Ha¨gg GM. Human muscle fibre abnormalities related to occupational load. Eur J Appl Physiol. 2000;83:159 –165. 42 Andersen LL, Suetta C, Andersen JL, et al. Increased proportion of megafibers in chronically painful muscles. Pain. 2008; 139:588 –593. 43 Dark A, Ginn KA, Halaki M. Shoulder muscle recruitment patterns during commonly used rotator cuff exercises: an electromyographic study. Phys Ther. 2007;87: 1039 –1046.
Volume 90
Number 4
Physical Therapy f
549
Research Report
S. Buatois, PhD, is Project Manager, Centre d’Etudes et de Formation sur le Vieillissement and Department of Geriatrics, University Hospital of Nancy.
A Simple Clinical Scale to Stratify Risk of Recurrent Falls in CommunityDwelling Adults Aged 65 Years and Older Severine Buatois, Christine Perret-Guillaume, Rene Gueguen, Patrick Miget, Guy Vanc¸on, Philippe Perrin, Athanase Benetos
C. Perret-Guillaume, MD, PhD, is Senior Geriatrician, Centre d’Etudes et de Formation sur le Vieillissement and Department of Geriatrics, University Hospital of Nancy.
Background. Correct identification of people at risk for recurrent falls facilitates the establishment of preventive and rehabilitative strategies in older adults.
R. Gueguen, PhD, is Senior Statistician, Centre de Me´decine Pre´ventive, Vandoeuvre-les-Nancy, France.
clinical scale to stratify risk for recurrent falls in community-dwelling elderly people based on easily obtained social and clinical items and (2) to evaluate the added value of 3 clinical balance tests in predicting this risk.
P. Miget, MD, is Geriatrician, Centre de Me´decine Pre´ventive, Vandoeuvre-les-Nancy, France.
Design. This was a prospective measurement study.
G. Vanc¸on, MD, is Rehabilitation Physician, Office d’Hygie`ne Sociale de Meurthe et Moselle, Centre Florentin, Nancy, France. P. Perrin, MD, PhD, is Otorhinolaryngologist, National Institute for Health and Medical Research, U 954, Faculty of Medicine, Vandoeuvre-les-Nancy, France, and Professor, Department of Oto Rhino Laryngology, University Hospital of Nancy. A. Benetos, MD, PhD, is Professor of Medicine and Geriatrics and Director, Centre d’Etudes et de Formation sur le Vieillissement and Department of Geriatrics, University Hospital of Nancy, 54500 Vandoeuvre-le`s-Nancy, France. Address all correspondence to Dr Benetos at:
[email protected]. [Buatois S, Perret-Guillaume C, Gueguen R, et al. A simple clinical scale to stratify risk of recurrent falls in community-dwelling adults aged 65 years and older. Phys Ther. 2010;90:550 –560.] © 2010 American Physical Therapy Association
Objective. The purposes of this study were: (1) to develop and validate a simple
Methods. A population of 1,618 community-dwelling people over 65 years of age underwent a health checkup, including performance of 3 clinical balance tests: the One-Leg-Balance Test, the Timed “Up & Go” Test, and the Five-Times-Sit-to-Stand Test. Falls were recorded using a self-administered questionnaire that was completed a mean (SD) of 25⫾5 months after the visit. Participants were randomly divided into either group A (n⫽999), which was used to develop the scale, or group B (n⫽619), which was used to prospectively validate the scale.
Results. Logistic regression analysis identified 4 variables that independently pre-
dicted recurrent falls in group A: history of falls, living alone, taking ⱖ4 medications per day, and female sex. Thereafter, 3 risk categories of recurrent falls (low, moderate, and high) were determined. Predicted probability of recurrent falls increased from 4.1% to 30.1% between the first and third categories. This scale subsequently was validated with great accuracy in group B. Only the Five-Times-Sit-to-Stand Test provided added value in the estimation of risk for recurrent falls, especially for the participants who were at moderate risk, in whom failure on the test (duration of ⬎15 seconds) doubled the risk.
Limitations. Falls were assessed only once, and length of follow-up was heterogeneous (18 –36 months). Conclusions. Clinicians could easily classify older patients in low-, moderate-, or high-risk groups of recurrent falls by using 4 easy-to-obtain items. The Five-TimesSit-to-Stand Test provides added value to stratify risk for falls in people at moderate risk.
Post a Rapid Response or find The Bottom Line: www.ptjournal.org 550
f
Physical Therapy
Volume 90
Number 4
April 2010
Assessment of Risk of Recurrent Falls in Elderly People
A
ccording to a French survey of the National Institute for Prevention and Health Education in 2005, 24% of the population aged 65 to 75 years reported one or more falls in the year preceding the survey.1 This finding is consistent with results reported in other countries showing that falls occurred each year in more than one third of the population aged 65 years and older.2– 4 Moreover, recent French statistics have shown that more than three quarters of deaths from falls occurred in people aged 75 years and older. The rate of mortality increased even further over the age of 75 years, namely 71.6 deaths per 100,000 people between 75 and 84 years of age and 442 deaths per 100,000 people over 85 years of age.5 Physical therapists have an important role to play in both the prevention of falls in elderly people and the multidisciplinary process because of their expertise with balance and musculoskeletal issues. In primary care, health care professionals need to assess intrinsic and extrinsic risk factors for falls and identify individuals who are at high risk for falls, especially those prone to recurrent falls.6,7 Several intrinsic clinical factors, such as consumption of psychoactive drugs,3 number of medications,3,8,9 female sex,9,10 living alone,8,10 history of falls,2– 4,8,11,12 muscle weakness,2– 4,12,13 abnormalities of balance and gait,2– 4,11,13,14 impaired cognition,4 and depression,3,15 individually and cumulatively have been found to be important predictors for falls. By taking into account these factors, it is possible to evaluate the risk for falls in elderly people. Moreover, several clinical balance tests, such as the Timed “Up & Go” Test (TUG),16,17 the One-Leg Balance Test (OLB),18 the Five-Times-Sit-to-Stand Test (FTSS),19 the Berg Balance Scale,20 and the April 2010
Tinetti Balance Subscale,21 have been proposed for identifying the risk for falls in elderly people. Indeed, some prospective studies have shown an association between the results of these tests and the risk for falls.8,12,18,22–25 In recent years in France, the use of some of these tests for people over 65 years of age has increased among general practitioners and physical therapists and in public health primary care centers. Although large-scale utilization of these tests for primary care could be useful, such widespread use could prove to be both time-consuming and costly, and, therefore, indications for testing should be specified. Despite a large number of studies assessing the risk for falls, there currently are no prospective data defining a simple clinical tool that could be used for fall risk stratification in community-dwelling elderly people or delineating the subpopulations in which the use of such clinical balance tests would be useful. Such an algorithm should first be able to establish risk stratification based on easily measurable items obtained during a standard medical visit and subsequently to target the subgroups of people in whom clinical balance tests may provide further information for this stratification of risk for falls. Thus, the aims of this prospective study were: (1) to develop and validate a simple clinical scale to stratify the risk for recurrent falls in a community-dwelling population who were healthy and aged 65 years and older through the use of easily obtainable social and clinical items and (2) to subsequently test, in the different risk subgroups, the added value of 3 commonly used balance clinical tests (ie, FTSS, OLB, and TUG) in predicting the risk for recurrent falls.
Method Participants This population-based study investigated 2,735 consecutive volunteers (1,357 women and 1,378 men) aged 65 years and older (mean [SD] age⫽ 70.3 [4.5] years) who were consulting for a senior medical checkup between January 1, 2004, and June 30, 2005, at the Centre de Me´decine Pre´ventive (CMP) in Nancy, France. The CMP is a medical center linked to the French national health care system (Se´curite´ Sociale) that provides free medical examinations for all affiliated working and retired people and their families. It is one of the largest medical centers of its kind in France, conducting approximately 30,000 examinations per year of residents living in the region of Lorraine. The aim of the senior medical checkup is to detect signs of physical, mental, or psychological frailty. This population-based study did not have any specific inclusion or exclusion criteria other than age (ⱖ65 years). Those individuals who consulted for a health checkup were primarily community-dwelling individuals, apparently healthy, and motivated to be followed up. The CMP received authorization from the Comite´ National d’Informatique et des Liberte´s to conduct these analyses. All individuals included in this study gave their informed consent at the time of medical examination. Initial Assessment During the health checkup, data for several sociodemographic character-
Available With This Article at ptjournal.apta.org • Audio Abstracts Podcast This article was published ahead of print on March 4, 2010, at ptjournal.apta.org.
Volume 90
Number 4
Physical Therapy f
551
Assessment of Risk of Recurrent Falls in Elderly People istics were collected in the study population, including age, living status, weight, height, and practice of physical activities. Other clinical data also were collected. Participants were queried regarding their medical history and answered questions concerning history of falls in the previous year, alcohol consumption, and current pharmacological treatments, especially the use of psychotropic drugs (benzodiazepines, neuroleptics, and antidepressants). All participants underwent a cognitive evaluation, including administration of the Mini-Mental State Examination. Health status was assessed with a test used by all Health Examination Centers in France. The question “Given your age, please indicate a score between 0 and 10, reflecting your health as you feel” was used to assess self-perceived health status. Individuals were required to answer this question using a visual analog scale rated from 0 (poor) to 10 (very good), as per the recommendation of the French National Center of the Organization of Health Examination Centers (Centre Technique d’Appui et de Formation des Centre d’Examen de Sante´ [CETAF]).
than 12 seconds was considered failure of the test. For the FTSS, participants were instructed to stand up from a chair 5 times as quickly as possible without pushing off.8,19 The seat height for the chair was 45 cm. The threshold for the FTSS was based on the results of our previous study in the same population, in which we demonstrated that the optimal cutoff for performing the FTSS in predicting recurrent “fallers” was 15 seconds.8 Consequently, a duration longer than 15 seconds was considered failure of this test. These tests were chosen because they are reliable and valid clinical tools commonly used to assess functional mobility or postural stability in community-dwelling older adults, require only limited space, and can be administered quickly.
During the medical examination, the participants performed the following clinical balance tests: OLB, TUG, and FTSS. For the OLB, participants were instructed to remain upright on one leg without support for at least 5 seconds.18 The threshold of 5 seconds was chosen in accordance with the Tinetti Balance Subscale,21 which considers a person to have normal balance if he or she is able to stand on one leg without support for 5 seconds. A shorter duration than 5 seconds was considered failure of this test. For the TUG, participants were observed and timed while rising from an armchair, walking 3 m, turning, walking back, and sitting down again (normal duration ⬍12 seconds).16,17 In accordance with Bischoff et al,16 a longer duration
A fall was defined as an event resulting in a person coming to rest unintentionally on the ground or other lower level, and not as the result of a major intrinsic event (eg, stroke, syncope) or overwhelming hazard.4 An overwhelming hazard was defined as a hazard that could have resulted in a fall by the youngest, healthiest people.4 Using these definitions, 2 of the co-investigators (S.B., A.B.), who were blinded to the participants’ data, separately reviewed the fall events in order to classify them into 1 of 3 groups: “nonfallers” (no fall), single fallers (one fall), and recurrent fallers (2 or more falls). In 87% of cases (1,703 participants), the classification given by each investigator was the same, and in the remaining 13% of cases, the final classification
552
f
Physical Therapy
Volume 90
Follow-up Procedure In December 2006, a questionnaire was mailed to all participants that included questions regarding how many times they had experienced a fall since their visit and the circumstances and consequences of the falls (mean [SD] follow-up period⫽25 [5] months, range⫽18 –36 months).
Number 4
was made after discussion and agreement between the 2 co-investigators. Because studies have shown that a single fall can be considered an accidental event and, consequently, is less difficult to predict than recurrent falls,4,8,12 the objective of this study was to assess the risk for recurrent falls. Consequently, nonfallers and single fallers were combined into one group (control group) and compared with the group of recurrent fallers. Follow-up Data Information concerning falls was obtained from 1,958 of the 2,735 volunteers (72% of the total sample). Out of the 1,958 participants, 340 participants had at least one missing item from the following variables: history of falls, living alone, FTSS score, TUG score, and OLB score. Therefore, statistical analyses were performed on the 1,618 participants for whom all data were present (59% of the total sample, mean [SD] age⫽70 [4] years; 797 women and 821 men). According to the self-questionnaire, 333 participants (21%) reported one or more falls during the follow-up period. One fall was reported by 182 participants (11%), and 2 or more falls were reported by 151 participants (9%). Among the 333 fallers, 229 (69%) had no injuries or a minor injury (ie, scratch or superficial wound), and 104 (31%) had a major injury (ie, fracture, hematoma, joint dislocation, head injury, severe laceration, or serious soft tissue injury). No significant difference in demographic, clinical, or balance tests was found between injurious fallers (participants who had major injuries) and noninjurious fallers (participants who had no injury or minor injuries). To verify the representativeness of the sample, the sociodemographic and clinical characteristics of participants who responded to the questionnaire were compared with those April 2010
Assessment of Risk of Recurrent Falls in Elderly People Table 1. Baseline Characteristics of Responders and Nonresponders Variable
Responders
Age (y), mean (SD)
Nonresponders
t Value or 2 Valuea
Pb
70.1 (4.4)
70.7 (4.6)
3.1
.001
Female sex, n (%)
978 (50%)
400 (51%)
0.5
.47
Living alone, n (%)
462 (24%)
241 (32%)
17.6
⬍.0001
Body mass index (kg/m2), mean (SD)
27.1 (5.9)
28.4 (7.8)
4.5
⬍.0001
Mini-Mental Status Examination score, mean (SD)
27.8 (2.1)
27.0 (2.7)
7.5
⬍.0001
6.9 (1.7)
6.5 (1.8)
6.1
⬍.0001
11.7 (11.5)
10.3 (11.7)
2.7
.007
3.2 (2.6)
3.5 (2.9)
2.4
.01
Health perception score (0–10), mean (SD) Alcohol consumption (g/d), mean (SD) No. of drugs taken, mean (SD) Medications (ⱖ4 drugs per day), n (%)
776 (40%)
343 (44%)
4.7
.03
Use of psychotropic drugs, n (%)
228 (12%)
111 (14%)
3.6
.06
History of falls in the previous year, n (%)
372 (21%)
176 (27%)
10.6
.001
Five-Times-Sit-to-Stand Test score (⬎15 s), n (%)
675 (37%)
293 (43%)
7.6
.005
Timed “Up & Go” Test score (⬎12 s), n (%)
149 (8%)
101 (14%)
23.4
⬍.0001
One-Leg Balance Test score (⬍5 s), n (%)
192 (10%)
126 (18%)
25.6
⬍.0001
a For qualitative variables expressed as n (%), comparisons between the 2 groups were performed using a chi-square test. For quantitative variables expressed as mean (SD), comparisons between the 2 groups were performed using a t test. b Statistical significance was accepted at a level of Pⱕ.05.
who did not respond (Tab. 1). Although statistical analysis revealed significant differences, responders and nonresponders had fairly similar values compared for age (70.1⫾4.4 years versus 70.7⫾4.6 years) and number of drugs used (3.2⫾2.6 versus 3.5⫾2.9). The proportion of women also was similar in both groups. However, among participants who did not respond, the health perception score was lower, more participants lived alone and had a history of falls. These individuals also had significantly higher failure rates for the 3 clinical balance tests compared with those who responded. Randomization Procedure In order to develop and validate our model for stratifying the risk for recurrent falls, we chose to divide the cohort into 2 groups comprising two thirds and one third of the participants, respectively. A number was randomly assigned to each participant, and participants were classified in ascending order of these numbers. The first two thirds then were placed April 2010
in group A (n⫽999), and the remaining third were placed in group B (n⫽619). The largest group (group A) was chosen to develop the risk model, and the smallest group (group B) was chosen for the validation of the model. Both groups presented the same sociodemographic and clinical characteristics (data not shown). Data Analysis Comparison of baseline demographics and clinical characteristics, as well as clinical balance tests results, between recurrent fallers and controls in group A was conducted using a 2-sample t test and the chi-square test. Statistical significance was accepted at a level of Pⱕ.05. All variables revealing differences at the level of P⬍.10 in univariate analysis between recurrent fallers and controls in group A, except for the clinical balance tests results, were entered into multivariate logistic regression analyses in order to identify independent risk factors of recurrent
falls. A logistic regression model for the prediction of recurrent falls, with adjusted odds ratios of the variables and 95% confidence intervals, was obtained. To facilitate the use of the model in clinical practice, risk scores were obtained by multiplication of the regression coefficients of the present predictors with a factor of 5, rounded off to the nearest integer. Group A was stratified into 3 risk categories (low, moderate, and high) according to the predicted probability of recurrent falls (P), calculated from the logistic probability model following the formula26: P ⫽ 1/(1 ⫹ exp(⫺(0 ⫹ 11 ⫹  2 2 ⫹  3 3 . . . ⫹ kk))) ⫻ 100, where 0 is a constant (called the “intercept”) and 1, 2, 3, and so on are regression coefficients of the predictor variables 1, 2, and 3, respectively. For each category, the mean of all individual values of the
Volume 90
Number 4
Physical Therapy f
553
Assessment of Risk of Recurrent Falls in Elderly People Table 2. Baseline Characteristics of Controls (Nonfallers and Single Fallers) and Recurrent Fallers in Group A
Variable Age (y), mean (SD)
Controls (nⴝ903)
Recurrent Fallers (nⴝ96)
t Value or 2 Valuea
Pb
69.8 (4.2)
71.5 (4.7)
3.6
.001
Female sex, n (%)
424 (47%)
67 (70%)
18.1
⬍.0001
Living alone, n (%)
189 (21%)
37 (38%)
15.4
⬍.0001
Body mass index (kg/m2), mean (SD)
27.0 (5.7)
28.1 (4.0)
1.9
.05
Mini-Mental Status Examination score, mean (SD)
28.0 (2.0)
27.9 (2.3)
0.4
.69
Health perception score (0–10), mean (SD) Alcohol consumption (g/d), mean (SD) Medications (ⱖ4 drugs per day), n (%) Use of psychotropic drugs, n (%)
7.0 (1.6)
6.3 (1.9)
4.2
⬍.0001
12.2 (12.6)
11.8 (12.9)
0.3
.74
334 (37%)
52 (54%)
10.8
.001
91 (10%)
19 (20%)
8.3
.004
History of falls in the previous year, n (%)
160 (18%)
53 (55%)
72.7
⬍.0001
Five-Times-Sit-to-Stand Test score (⬎15 s), n (%)
321 (35%)
58 (60%)
22.8
⬍.0001
Timed “Up & Go” Test score (⬎12 s), n (%)
67 (7%)
12 (12%)
3.1
.08
One-Leg Balance Test score (⬍5 s), n (%)
88 (10%)
16 (17%)
4.4
.03
Duration of follow-up (mo), mean (SD)
25.6 (5.1)
24.9 (5.2)
1.1
.25
a
For qualitative variables expressed as n (%), comparisons between the 2 groups were performed using a chi-square test. For quantitative variables expressed as mean (SD), comparisons between the 2 groups were performed using a t test. b Statistical significance was accepted at a level of Pⱕ.05.
predicted probabilities was calculated and termed “risk estimated.”
the NCSS 2000 statistical software package.*
The scale for the stratification of participants according to the risk for recurrent falls, elaborated in group A, then was validated in group B. Following this procedure, participants in group B were divided into the 3 risk categories (low, moderate, and high), and the mean of the number of recurrent fallers was calculated for each category and termed “risk observed.” Thereafter, comparisons between the “risk estimated” in group A and the “risk observed” in group B were performed.
Role of the Funding Source This study was supported by grants from the Communaute´ Urbaine du Grand Nancy, France. The Communaute´ Urbaine du Grand Nancy had no role in the design, methods, participant recruitment, data collection and analysis, or preparation of the manuscript.
Finally, comparisons of the percentages of recurrent fallers according to the results of the 3 clinical balance tests in the 3 risk categories were conducted using the chi-square test in group B and in the entire cohort, comprising groups A and B. Statistical analyses were performed using
Results Overall, our analyses of the characteristics of this study population showed that none of the participants had any major physical or mental impairment, presented any signs of loss of autonomy, or regularly performed activities of daily living without assistance. Table 2 summarizes the main sociodemographic and clinical characteristics and the results of the 3 clinical balance tests for the control * NCSS, 329 North 1000 East, Kaysville, UT 84037.
554
f
Physical Therapy
Volume 90
Number 4
participants (nonfallers and single fallers) and recurrent fallers in group A. Univariate analysis revealed that older age, female sex, living alone, number of drugs used, use of psychotropic drugs, positive history of falls, and failure in the 3 clinical balance tests were more frequent among recurrent fallers compared with the control participants. Health perception score was lower and body mass index was higher in recurrent fallers compared with the other participants. No difference was observed in terms of Mini-Mental State Examination scores or alcohol consumption. Table 3 presents the risk model for the prediction of recurrent falls obtained by multiple logistic regressions, including all variables revealing differences at the P⬍.10 level in the univariate analysis (age, sex, living status, body mass index, health perception score, medications, history of falls, use of psychotropic drugs), except for the clinical bal-
April 2010
Assessment of Risk of Recurrent Falls in Elderly People Table 3. Risk Model for the Prediction of Recurrent Falls Obtained by Multiple Logistic Regression in Group Aa
Variable
Regression Coefficient (Standard Error)
P
Odds Ratio
95% Confidence Interval
Risk Scoreb
Predicted Probability of Recurrent Fallsc (%)
History of falls (yes)
1.55 (0.24)
⬍.0001
4.72
(3.00–7.43)
8
12.9
Living alone (yes)
0.56 (0.24)
.021
1.75
(1.08–2.82)
3
5.2
Medications (ⱖ4 drugs per day)
0.51 (0.23)
.025
1.66
(1.06–2.60)
3
5.0
0.48 (0.25)
.052
1.62
(0.99–2.65)
2
4.9
0
3.0
Sex (female) Intercept
⫺3.46 (0.24)
Final model (n⫽999), dependent variable: recurrent fallers (n⫽96) versus controls (n⫽903). Pseudo R ⫽.072. All significant variables revealing differences at the level of P⬍.10 in univariate analysis between recurrent fallers and controls in group A, except for the clinical balance tests results, were entered into multivariate logistic regression analyses. b Risk scores were obtained by multiplication of the regression coefficients of the present predictors with a factor of 5, rounded off to the nearest integer. c Predicted probabilities were calculated from the logistic probability model. a
Following the above analysis, the study population was divided into 3 risk categories: low (score⫽0 – 4), moderate (score⫽5–10), and high (score⫽11–16), representing 58%, 29%, and 13% of group A, respectively. The choice of scores 5 and 11 as thresholds for changing risk category was determined according to the predicted probability of recurrent falls. Individuals with a score of ⬍5 had a predicted probability of recurrent falls of ⬍5%, whereas those with a score of ⱖ11 had a April 2010
predicted probability of ⬎20%. According to this classification, the mean predicted probability of recurrent falls increased from 4.1% to 30.1% between the first and third categories (Fig. 2). Moreover, Figure 2 illustrates the validation of our scale for the stratification according to the risk for recurrent falls, elaborated in group A (hatched column) and subsequently verified in the second population group, group B (black column). The accuracy of the model was deemed excellent.
In order to assess the added value of the 3 balance clinical tests in the prediction of the risk for recurrent falls, the percentage of recurrent fallers in each risk category was calculated according to the results of the 3 clinical balance tests in group B and in the entire cohort (Tab. 4). The OLB and TUG scores did not significantly influence the percentage of recurrent falls in any of the 3 risk categories. In group B, the percentage of recurrent fallers was not significantly different between partici-
300 256 250
No. of Participants
ance tests results. The analysis identified 4 independent determinants for recurrent falls: positive history of falls, living alone, taking ⱖ4 medications per day, and female sex. No difference in the duration of followup was observed in the subgroups according to these 4 variables. Using the regression coefficients of the present model, a risk score was established as follows: positive history of falls represented a score of 8, living alone and taking ⱖ4 medications per day represented a score of 3 each, and female sex represented a score of 2. The overall score ranged from 0, in absence of any predictors, to 16, if all predictors were present. The frequency distribution of individual scores, from 0 to 16, is presented in Figure 1.
2
200 155
168
150
134 86
100
67 52 50
30
25
20
6
0 0
2
3
5
6
8
10
11
13
14
16
Risk Score
Figure 1. Frequency distribution of individual risk scores from 0 to 16.
Volume 90
Number 4
Physical Therapy f
555
Assessment of Risk of Recurrent Falls in Elderly People Risk Estimated Risk Observed 35
additional information in the assessment of the risk for recurrent falls in any of the risk categories.
Risk for Recurrent Falls (%)
30.1% 30
27.5%
25 20 15
11.5%
11.1%
10 4.1%
5
4.0%
0 Score 0–4 Low-Risk Category
Score 5–10 Moderate-Risk Category
Score 11–16 High-Risk Category
Figure 2. Comparison of risk for recurrent falls estimated according to the scale elaborated in group A with that observed in group B in the 3 risk categories.
pants with an FTSS score of ⬎15 seconds and those with an FTSS score of ⱕ15 seconds in the low-risk group and in the high-risk group. By contrast, in the moderate-risk group (score⫽5–10), the participants who needed more than 15 seconds to complete the FTSS presented twice as many recurrent falls compared with participants who performed the test in less than 15 seconds. The same results relative to the FTSS also were confirmed in the entire cohort as a whole.
Three main results can be derived from this study: 1. Taking into account 4 easily measurable items (history of falls, living alone, number of medications, and female sex), we elaborated a scale able to stratify individuals into 3 risk categories of recurrent falls, with a risk ratio of 7.5 between the first and third categories. This scale was validated with great accuracy in a second population with similar characteristics.
Discussion The aim of this prospective study was to propose an evaluation and stratification of the risk for recurrent falls in elderly people aged over 65 years living in the community, based on clinical items that are easily obtained by health care professionals, and to subsequently target subgroups of participants in whom balance clinical tests may provide further information for the stratification of risk for recurrent falls.
556
f
Physical Therapy
Volume 90
2. The FTSS provided added value in the estimation of risk for recurrent falls, but only in the subgroup of participants at moderate risk. In this group, participants with an FTSS score of ⬎15 seconds presented twice as many recurrent falls compared with participants who performed the test in less than 15 seconds. 3. The other 2 clinical balance tests (TUG and OLB) failed to provide
Number 4
Various types of assessment tools have been proposed for screening populations at high risk for falls and to reduce such risk by targeting specific factors. These tools are based on functional mobility assessment (eg, gait, strength [force-generating capacity], balance control) or on assessment of multiple factors known to be associated with fall risk (eg, psychological status, mobility dysfunction, acute and chronic illnesses, sensory deficits, medication use, history of falls).3,27–32 Depending on their complexity and purpose, these tools can be used by physical therapists, physicians, or nurses to assess older adults in community, home-support, long-term care, and acute care settings. The originality of the present study resides in the establishment of an assessment scale of risk for recurrent falls by combining 4 specific predictors, taking into account the results of multivariate logistic regression, which can be easily used by health care professionals across all clinical practice settings. Findings from other studies of community-dwelling elderly people have confirmed the validity of these factors separately.2,3,10 –12,31,32 Indeed, history of falls over the previous year is a high predictor often reported in risk tool assessment and screening recommendation for the prevention of falls in older people.2– 4,6 – 8,11,12,32,33 Furthermore, in a recent study, Kharicha et al10 showed that after adjustment for age, sex, income, and educational attainment, living alone remained associated with multiple falls. This finding was confirmed in the present study, as living alone was found to be the second strongest predictor of recurrent falls in our risk model. It also has been shown that elderly women have a higher risk of falling and balApril 2010
Assessment of Risk of Recurrent Falls in Elderly People Table 4. Number and Percentage of Recurrent Fallers in the 3 Risk Category Groups According to the Results of the Clinical Balance Tests in Group B and in Entire Cohort Clinical Balance Test Group B (nⴝ619)
No. of Recurrent Fallers/No. of Participants per Ranking (% Recurrent Fallers) Score 0–4 (nⴝ352)
Score 5–10 (nⴝ198)
Score 11–16 (nⴝ69)
10/243 (4%)
9/119 (8%)
10/36 (28%)
4/109 (4%)
13/79 (16%)a
9/33 (27%)
13/335 (4%)
20/183 (11%)
16/59 (27%)
1/17 (6%)
2/15 (13%)
3/10 (30%)
14/333 (4%)
20/181 (11%)
16/62 (26%)
0/19 (0%)
2/17 (12%)
3/7 (43%)
Score 0–4 (nⴝ931)
Score 5–10 (nⴝ490)
Five-Times-Sit-to-Stand Test Normal (ⱕ15 s) Abnormal (⬎15 s) Timed “Up & Go” Test Normal (ⱕ12 s) Abnormal (⬎12 s) One-Leg Balance Test Normal (ⱖ5 s) Abnormal (⬍5 s) Entire Cohort (nⴝ1,618)
Score 11–16 (nⴝ197)
Five-Times-Sit-to-Stand Test Normal (ⱕ15 s)
20/629 (3%)
21/292 (7%)
26/97 (27%)
Abnormal (⬎15 s)
17/302 (6%)
35/198 (18%)b
32/100 (32%)
35/887 (4%)
53/446 (12%)
45/164 (27%)
2/44 (4%)
3/44 (7%)
13/33 (39%)
34/880 (4%)
50/430 (12%)
46/161 (29%)
3/51 (6%)
6/60 (10%)
12/36 (33%)
Timed “Up & Go” Test Normal (ⱕ12 s) Abnormal (⬎12 s) One-Leg Balance Test Normal (ⱖ5 s) Abnormal (⬍5 s)
a Significant difference (P⫽.05). Participants with Five-Times-Sit-to-Stand Test scores of ⬎15 s presented twice as many recurrent falls compared with participants with Five-Times-Sit-to-Stand Test scores of ⱕ15 s in group B. b Significant difference (P⫽.003). Participants with Five-Times-Sit-to-Stand Test scores of ⬎15 s presented more than twice as many recurrent falls compared with participants with Five-Times-Sit-to-Stand Test scores of ⱕ15 s in entire cohort.
ance disorders than men of the same age.9 Older adults taking more than 3 medications per day also were found to be at increased risk for recurrent falls.2,34 Our data confirm that the fall prevention programs for elderly people must be oriented toward the following objectives: to improve balance and strength, to reduce the number of drugs used,35 to promote physical activity (such as Tai Chi intervention),36 and to promote home safety and safe behaviors. In the present study, knowledge of these 4 parameters appeared to be sufficient in classifying participants into 1 of 3 risk categories: (1) the low-risk category, in which the preApril 2010
dicted probability of recurrent falls is less than 5%; (2) the moderate-risk category; or (3) the high-risk category, in which the predicted probability of recurrent falls is higher than 20%. Predicted probability of recurrent falls increased from 4.1% to 30.1% between the first and third categories. That is, this scale was able to stratify participants into 3 risk categories for recurrent falls, with a risk ratio of 7.5 between the first and third categories. The strength of this study is that this scale, derived initially in a first group, subsequently was successfully validated with great precision in a second sample, who presented the same characteristics as the first study group.
In addition, this study assessed the added value of 3 clinical balance tests in estimating the risk for recurrent falls in subgroups of participants according to their risk score measured by our clinical scale. The results showed that only the FTSS was useful and presented an added value to hone the estimation of the risk for recurrent falls, whereas the OLB and TUG showed no added predictive value. Two prospective studies have shown that the inability to rise from a chair was associated with risk for falls in older people.2,12 In a recent prospective study of a large cohort of community-dwelling women aged 69 years or older, the inability to rise from a chair 5 times
Volume 90
Number 4
Physical Therapy f
557
Assessment of Risk of Recurrent Falls in Elderly People was 1 of the 3 components of a frailty index (the SOF [Study of Osteoporotic Fractures] Index), which was associated with the risk for recurrent falls.37 These data confirm our previous study results,8 which demonstrated that among the 3 balance tests, only a slower FTSS time (⬎15 seconds), which may reflect muscular weakness of the lower limbs or postural balance disorders, was a predictor of recurrent falls after adjustment for other independent factors (ie, history of falls, living alone, female sex, and number of medications). Moreover, the results of the current study further underscore that it may not be necessary to systematically assess the FTSS in all older study participants, but mainly those exhibiting a moderate clinical risk for recurrent falls, as assessed with the clinical scale established herein. For the participants in either the low-risk or high-risk category, the FTSS did not appear to provide any additive value for prediction of the risk for recurrent falls because, in these 2 groups, the risk remains low and high, respectively, independently of performance on the balance test. By contrast, in participants at moderate risk, failure on the FTSS doubled the risk for recurrent falls. Nonetheless, these results were found in small subgroups; consequently, further studies on larger groups will be needed to confirm the present data. In a recent prospective study of a cohort of older women with some disability, Lamb et al32 constructed 2 algorithms for prediction of falls, one based on self-report items alone and the other incorporating a selection of performance tests, especially those related to muscle strength. Their study showed the importance of performance tests, especially of knee extensor strength, in women at low risk for falls. Our data confirm that muscle strength of the lower limbs (which is one of the main determi558
f
Physical Therapy
Volume 90
nants of the FTSS) is a major factor for the prediction of the risk for falls in older adults.38 In this context, physical therapists have an important role in the assessment of muscle strength and treatment of muscle weakness in older adults at risk for falls. Our results also suggest that the predictive value of the OLB and TUG for risk for recurrent falls was not powerful enough compared with the FTSS for our sample of communitydwelling elderly people. According to previous prospective studies, the predictive value of the OLB and TUG for fall risk remains unclear. In a prospective study, Vellas et al18 demonstrated that impaired one-leg balance was the only significant independent predictor of injurious falls, but not of all falls. In another prospective study, the TUG did not predict risk for falls in a group of older adults who were healthy.27 Nevertheless, in older women with vertebral fractures, TUG scores were associated with risk for recurrent falls, but Sitto-Stand Test scores were not.25 Lastly, in a recent prospective study, Lin et al24 showed that the TUG, OLB, Functional Reach Test, and Tinetti Balance Subscale exhibited excellent test-retest reliability and discriminant validity, but poor responsiveness in prediction of falls in people aged 65 years and older. Our study showed that only 8% of the participants had abnormal TUG scores and 10% had abnormal OLB scores, whereas more than 35% had abnormal FTSS scores, both in the total population and in groups A and B. These results indicate that normal scores for the TUG and OLB are too easy to achieve for our sample or that the thresholds of 12 seconds for the TUG and 5 seconds for the OLB are not sufficiently sensitive for a general population of communitydwelling people aged over 65 years in apparently good health. These re-
Number 4
sults most likely can be generalized to those individuals with few balance or mobility impairments. For these individuals, it would be interesting to increase the difficulty of these 2 tests, for example, by providing a mental task during the TUG or increasing the duration of the OLB, and then to evaluate the predictive value of these clinical balance tests for risk for recurrent falls. Several limitations should be discussed. This study may be somewhat biased because 28% of the sample who did not respond were clinically more frail and consequently: (1) group fall rates might have been higher than previously reported if these individuals were included in the study and (2) the results of this study may not be valid for frailer populations. Another limitation of this study is the fact that the registration of falls was assessed only once, during a follow-up of 18 to 36 months, which might explain why only 21% of the participants reported one or more falls. In contrast, in the published literature, the percentage of people aged over 65 years who reported falls approached 30% and over.2– 4 A follow-up involving sending postcard questionnaires on a more frequent basis (eg, every 4 months) would certainly result in a higher rate of recording of falls. Moreover, participants were volunteers living in the community with a mean age of 70 years and in an apparently good state of health, and our findings may not be applicable to other populations. Finally, grouping people having experienced several falls in a short period of time with those who had 2 falls over the entire followup period under the same heading of recurrent fallers may be somewhat erroneous; however, in the present study, we decided to clearly distinguish the single fallers and to regroup those with more than one April 2010
Assessment of Risk of Recurrent Falls in Elderly People fall over the full period of the followup. Additional prospective studies are needed to confirm the predictive validity of the scale in other populations.
Implications for Clinical Practice We believe that the results of this study may have a significant impact on primary care clinical practice in the field of fall risk evaluation. First, with the help of this scale, physical therapists and other health care professionals could easily classify elderly, community-dwelling patients in low-, moderate-, or high-risk groups of recurrent falls by using 4 easy-toobtain items (ie, positive history of falls, living alone, taking ⱖ4 medications, and female sex). Second, among the different clinical tests available, the FTSS could fine-tune the estimation of the risk for recurrent falls, especially for those individuals at moderate risk. We could suggest use of this clinical scale for screening and use of a larger multifactorial risk assessment for the high-risk individuals and for those at moderate risk who fail the FTSS with low scores. This strategy could circumvent the need for proposing that time-consuming and costly examinations be performed in all individuals. Nevertheless, further studies in community-dwelling individuals should be conducted to ascertain the validity, effectiveness, and feasibility of this clinical scale, as well as the added value of different clinical balance tests in assessing the risk for recurrent falls. All authors contributed to the design and the organization of the study. Dr Buatois contributed to participant recruitment, analysis and interpretation of data, and writing the manuscript. Dr Benetos was the chief investigator and contributed to analysis and interpretation of data and writing the manuscript. Dr Buatois, Dr Perret-Guillaume, and Dr Benetos prepared the first draft of the manuscript, and Dr Gueguen, Dr Vanc¸on, and Dr Perrin provided critical intellectual
April 2010
input to subsequent drafts. Dr Gueguen advised on statistical techniques and contributed to interpretation of the data. Dr Miget contributed to participant recruitment and realization of the clinical balance tests. All authors approved the final version. This study was conducted with the help of the Clinical Investigation Centre of the University Hospital of Nancy. The authors thank the medical staff, the nurses, and the directors of the Centre de Me´decine Pre´ventive, Vandoeuvre-les-Nancy, France, and all participants. They also thank Mr Pierre Pothier for language review and stimulating discussions. The study was approved by the Comite´ National d’Informatique et des Liberte´s. This study was supported by grants from the Communaute´ Urbaine du Grand Nancy, France. This article was received May 14, 2009, and was accepted December 6, 2009. DOI: 10.2522/ptj.20090158
References 1 Guilbert P, Gautier A. Barome`tre Sante´ 2005: Premiers Re´sultats. Paris, France: Institut National de Pre´vention et d’E´ducation Pour la Sante´; 2006. 2 Campbell AJ, Borrie MJ, Spears GF. Risk factors for falls in a community-based prospective study of people 70 years and older. J Gerontol. 1989;44:M112–M117. 3 Stalenhoef PA, Diederiks JP, Knottnerus JA, et al. A risk model for the prediction of recurrent falls in community-dwelling elderly: a prospective cohort study. J Clin Epidemiol. 2002;55:1088 –1094. 4 Tinetti ME, Speechley M, Ginter SF. Risk factors for falls among elderly persons living in the community. N Engl J Med. 1988; 319:1701–1707. 5 Ermanel C, The´lot B, Jougla E, Pavillon G. Fatal home and leisure accidents in metropolitan France, 2000 –2004. Bull Epide´miol Hebd. 2007;37–38:318 –322. 6 Recommandations de Bonnes Pratiques Professionnelles: Evaluation et Prise en Charge des Personnes Age´es Faisant des Chutes Re´pe´te´es. Paris, France: Haute Autorite´ de Sante´, Socie´te´ Franc¸aise de Ge´riatrie et Ge´rontologie; 2009. 7 Good Practice Guide: Prevention of Falls in the Elderly Living at Home. Paris, France: Institut National de Pre´vention et d’Education Pour la Sante´, Re´seau Francophone de Pre´vention des Traumatismes et de Promotion de la Se´curite´; 2008. 8 Buatois S, Miljkovic D, Manckoundia P, et al. Five-Times-Sit-to-Stand test is a predictor of recurrent falls in healthy community-living subjects aged 65 and older. J Am Geriatr Soc. 2008;56:1575–1577.
9 Campbell AJ, Spears GF, Borrie MJ. Examination by logistic regression modelling of the variables which increase the relative risk of elderly women falling compared to elderly men. J Clin Epidemiol. 1990;43: 1415–1420. 10 Kharicha K, Iliffe S, Harari D, et al. Health risk appraisal in older people, 1: are older people living alone an “at-risk” group? Br J Gen Pract. 2007;57:271–276. 11 Ganz DA, Bao Y, Shekelle PG, et al. Will my patient fall? JAMA. 2007;297:77– 86. 12 Nevitt MC, Cummings SR, Kidd S, et al. Risk factors for recurrent nonsyncopal falls: a prospective study. JAMA. 1989; 261:2663–2668. 13 Lord SR, Ward JA, Williams P, et al. Physiological factors associated with falls in older community-dwelling women. J Am Geriatr Soc. 1994;42:1110 –1117. 14 Buatois S, Gueguen R, Gauchard GC, et al. Posturography and risk of recurrent falls in healthy non-institutionalized persons aged over 65. Gerontology. 2006;52: 345–352. 15 Whooley MA, Kip KE, Cauley JA, et al; for the Study of Osteoporotic Fractures Research Group. Depression, falls, and risk of fracture in older women. Arch Intern Med. 1999;159:484 – 490. 16 Bischoff HA, Stahelin HB, Monsch AU, et al. Identifying a cut-off point for normal mobility: a comparison of the Timed “Up and Go” Test in community-dwelling and institutionalised elderly women. Age Ageing. 2003;32:315–320. 17 Podsiadlo D, Richardson S. The timed “Up & Go”: a test of basic functional mobility for frail elderly persons. J Am Geriatr Soc. 1991;39:142–148. 18 Vellas BJ, Wayne SJ, Romero L, et al. Oneleg balance is an important predictor of injurious falls in older persons. J Am Geriatr Soc. 1997;45:735–738. 19 Whitney SL, Wrisley DM, Marchetti GF, et al. Clinical measurement of sit-to-stand performance in people with balance disorders: validity of data for the Five-TimesSit-to-Stand Test. Phys Ther. 2005;85: 1034 –1045. 20 Muir SW, Berg K, Chesworth B, et al. Use of the Berg Balance Scale for predicting multiple falls in community-dwelling elderly people: a prospective study. Phys Ther. 2008;88:449 – 459. 21 Tinetti ME. Performance-oriented assessment of mobility problems in elderly patients. J Am Geriatr Soc. 1986;34:119 –26. 22 Bogle Thorbahn LD, Newton RA. Use of the Berg Balance Test to predict falls in elderly persons. Phys Ther. 1996;76: 576 –583. 23 Kristensen MT, Foss NB, Kehlet H. Timed “Up & Go” Test as a predictor of falls within 6 months after hip fracture surgery. Phys Ther. 2007;87:24 –30. 24 Lin MR, Hwang HF, Hu MH, et al. Psychometric comparisons of the timed up and go, one-leg stand, functional reach, and Tinetti balance measures in communitydwelling older people. J Am Geriatr Soc. 2004;52:1343–1348.
Volume 90
Number 4
Physical Therapy f
559
Assessment of Risk of Recurrent Falls in Elderly People 25 Morris R, Harwood RH, Baker R et al. A comparison of different balance tests in the prediction of falls in older women with vertebral fractures: a cohort study. Age Ageing. 2007;36:78 – 83. 26 Hilbe JM. Logistic Regression Models. Boca Raton, FL: CRC Press; 2009. 27 Boulgarides LK, McGinty SM, Willett JA, et al. Use of clinical and impairment-based tests to predict falls by communitydwelling older adults. Phys Ther. 2003;83: 328 –339. 28 Lord SR, Menz HB, Tiedemann A. A physiological profile approach to falls risk assessment and prevention. Phys Ther. 2003;83:237–252. 29 Milisen K, Staelens N, Schwendimann R, et al. Fall prediction in inpatients by bedside nurses using the St. Thomas’s Risk Assessment Tool in Falling Elderly Inpatients (STRATIFY) instrument: a multicenter study. J Am Geriatr Soc. 2007;55: 725–733.
560
f
Physical Therapy
Volume 90
30 Scott V, Votova K, Scanlan A, et al. Multifactorial and functional mobility assessment tools for fall risk among older adults in community, home-support, long-term and acute care settings. Age Ageing. 2007; 36:130 –139. 31 Tromp AM, Pluijm SM, Smit JH, et al. Fall-risk screening test: a prospective study on predictors for falls in communitydwelling elderly. J Clin Epidemiol. 2001;54: 837– 844. 32 Lamb SE, McCabe C, Becker C, et al. The optimal sequence and selection of screening test items to predict fall risk in older disabled women: the Women’s Health and Aging Study. J Gerontol A Biol Sci Med Sci. 2008;63:1082–1088. 33 American Geriatrics Society, British Geriatrics Society, and American Academy of Orthopaedic Surgeons Panel on Falls Prevention. Guideline for the prevention of falls in older persons. J Am Geriatr Soc. 2001;49:664 – 672. 34 Leipzig RM, Cumming RG, Tinetti ME. Drugs and falls in older people: a systematic review and meta-analysis, I: psychotropic drugs. J Am Geriatr Soc. 1999;47:30 –39.
Number 4
35 Tinetti ME, McAvay G, Claus E. Does multiple risk factor reduction explain the reduction in fall rate in the Yale FICSIT Trial? Frailty and Injuries Cooperative Studies of Intervention Techniques. Am J Epidemiol. 1996;144:389 –399. 36 Lin MR, Hwang HF, Wang YW, et al. Community-based tai chi and its effect on injurious falls, balance, gait, and fear of falling in older people. Phys Ther. 2006; 86:1189 –1201. 37 Ensrud KE, Ewing SK, Taylor BC, et al. Comparison of 2 frailty indexes for prediction of falls, disability, fractures, and death in older women. Arch Intern Med. 2008; 168:382–389. 38 Moreland JD, Richardson JA, Goldsmith CH, Clase CM. Muscle weakness and falls in older adults: a systematic review and meta-analysis. J Am Geriatr Soc. 2004; 52:1121–1129.
April 2010
Research Report Kinematics of Rising From a Chair: Image-Based Analysis of the Sagittal Hip-Spine Movement Pattern in Elderly People Who Are Healthy Mohammad R. Fotoohabadi, Elizabeth A. Tully, Mary P. Galea
Background. Rehabilitation of elderly patients with sit-to-stand (STS) dysfunction includes retraining coordinated movement among participating body segments. Although trunk position is considered important, spinal movement has not been measured. Objective. The aim of this study was to describe the sagittal thoracolumbar kinematics and hip-lumbar interaction during the STS task in elderly people who were healthy in order to guide physical therapists in developing treatment strategies.
Design. This was an observational study. Methods. Ten retroreflective markers were attached to the midline thoracolumbar spine, pelvis, and right lower limb of 41 elderly people who were healthy. A 2-dimensional video analysis system was used to measure sagittal thoracic, lumbar, hip, and knee joint angles during the STS task. Maximal available flexion-extension angles in these joints and regions also were determined.
Results. Prior to buttocks lift-off, forward trunk lean comprised concurrent hip and lumbar flexion and thoracic extension. Hip flexion dominated, with a hip/lumbar ratio of 4.7:1 and a thoracic/lumbar ratio of 1.7:1. The hip and lumbar spine contributed 90% and 23% of their maximal available flexion angle, respectively, and the thoracic spine contributed 86% of its maximal extension range of movement. After lift-off, the hips and lumbar spine extended (ratio of 5.2:1), and the thoracic spine flexed (thoracic/lumbar ratio of 0.4:1). At lift-off, the hips and knees were similarly flexed (96°) and then locked together in a linear pattern of extension. Following lift-off, there was a brief transition phase (5% of STS duration) in which, although the hips, knees, and lumbar spine were extending, the trunk continued to flex forward a few degrees.
M.R. Fotoohabadi, PT, MScMedEd, PhD, is Assistant Professor and Senior Lecturer and Head, Education Development Office, Faculty of Rehabilitation Sciences, Shiraz University of Medical Sciences, Shiraz, Fars, Iran. E.A. Tully, BAppSci (Physio), PhD, is Senior Lecturer, School of Physiotherapy, Faculty of Medicine Dentistry and Health Sciences, University of Melbourne, 200 Berkeley St, Carlton, Victoria 3010, Australia. Address all correspondence to Dr Tully at: e.tully@ unimelb.edu.au. M.P. Galea, BAppSci (Physio), BA, PhD, is Professor of Clinical Physiotherapy and Director, Rehabilitation Sciences Research Centre, University of Melbourne, and Austin Health, Parkville, Victoria, Australia. [Fotoohabadi MR, Tully EA, Galea MP. Kinematics of rising from a chair: image-based analysis of the sagittal hip-spine movement pattern in elderly people who are healthy. Phys Ther. 2010;90:561– 571.] © 2010 American Physical Therapy Association
Limitations. Results may differ in elderly people who are less active. Conclusions. The revised model for image-based analysis demonstrated concurrent hip and thoracolumbar movement during the STS task. Close to full available hip flexion and thoracic extension were needed for optimal STS performance.
Post a Rapid Response or find The Bottom Line: www.ptjournal.org April 2010
Volume 90
Number 4
Physical Therapy f
561
Kinematics of Rising From a Chair
D
uring activities of daily living, such as standing up from a sitting position and walking, the thoracolumbar spine (trunk) and lower limbs act as an interdependent kinematic chain of joints. The concept of a series of related joints providing coordinated movement during functional activities and of movement at one joint being affected by and affecting movement at adjacent joints is familiar to clinicians.
The activity of rising from a sitting position is an essential prerequisite for walking and, therefore, functional independence.1–5 It has been reported that people who have difficulty rising to a standing position have greater likelihood of falling during ambulation6 – 8 and to need help with daily activities.9,10 Inability to stand up has been linked to death in elderly people.11,12 The sit-to-stand (STS) task is a complex activity, involving movement of all body segments from head to foot. The task requires sufficient joint mobility, lower-limb strength (forcegenerating capacity), and balance to enable the center of mass to be transferred forward and upward from the stable seated position to erect standing on a small base of support, the feet.12–14 For optimal performance, each joint or body part must move the correct amount in the right direction and at the appropriate time. Rehabilitation of patients with STS dysfunction includes retraining of this movement interaction.2,15,16 However, the ability of the therapist
Available With This Article at ptjournal.apta.org • Audio Abstracts Podcast This article was published ahead of print on February 18, 2010, at ptjournal.apta.org.
562
f
Physical Therapy
Volume 90
to facilitate the STS movement effectively depends on a background understanding of the interaction of all contributing joints and body segments in people who are healthy (ie, the movement pattern). A consequence of the aging process is a decline in the attributes required for successfully completing the STS movement. Deficits include lowerlimb muscle weakness9,14,17–20 and decreased balance.21–24 The kinematics of this important task have been well documented with respect to duration, velocity, and acceleration of body segments,25–27 as well as the kinematics and kinetics of the lowerlimb joints.28 –30 Although trunk angle has been measured with respect to an external reference (eg, horizontal or vertical plane) and with respect to the pelvis,17,18 the sagittal contribution of the thoracic and lumbar regions to trunk movement has been largely ignored. This failure to provide information regarding the thoracolumbar kinematics during the STS task is due to the model of measurement used in previous 2dimensional (2D) image-based studies in which the spine was considered a “rigid unit” defined by skin reference markers located on proximal and distal ends of the trunk segment, such as markers on the spinous process of the first thoracic vertebra (T1) and first sacral spine (S1),31 the acromion and mid-iliac crest,25,32 the scapula spine and sacroiliac joints,33 and the lateral glenohumeral joint and greater trochanter.27,34,35 Even recent 3-dimensional (3D) studies13,36 failed to include the spine in their STS analysis, using the rigid trunk marker placement of earlier gait studies.37 As a result, it has been concluded that the pelvis and spine act together as a functional unit during the STS task, with minimal movement in the spine or between the spine and pelvis.16(p131)
Number 4
Although recently the sagittal contribution of the lumbar spine38 and of the thoracic spine and lumbar spine39 during the STS task in young individuals has been described, the sagittal hip-spine movement pattern during the STS task in elderly people who are healthy appears not to have been investigated. Previous image-based studies have reported the range of knee and ankle movement required for optimal STS performance; however, the measurement of hip angle has largely been inaccurate due to the model of reference marker placement. Markers on the shoulder, greater trochanter, and knee have been used to measure both trunk angle17,24,40 and hip angle.27,35,41 As a valid hip angle requires markers to be located on the body segments immediately adjacent to the hip joint (ie, pelvis and femur), inclusion of the joints of the thoracolumbar spine in the shouldergreater trochanter-knee model has confounded the sagittal hip measurement. In addition, no study has determined the proportion of full available flexion-extension angles in the hip, lumbar spine, and thoracic spine used by elderly people during the STS task. Knowledge of the amount of movement required by the participating joints during STS performance enables the therapist to predict the effect of stiffness in interfering with or causing compensatory movements in other joints or regions and to include appropriate mobilizing exercises in the treatment plan. Therefore, the aim of this study was to use a revised model of reference marker placement and a 2D video motion analysis system to describe the sagittal kinematics of the thoracic spine, lumbar spine, and hip and knee joints during the STS task in elderly people who were healthy.
April 2010
Kinematics of Rising From a Chair
Method Participants As in a previous study by this team,39 a sample of convenience of 41 community-dwelling elderly people (22 female and 19 male) were recruited through newspaper advertisement. The participants had a mean (SD) age of 69.9 (5.3) years, height of 1.67 (0.09) m, and body mass index of 26.0 (3.5) kg/m2. Informed consent was obtained from all participants, and the project had ethics approval from the Human Research Ethics Committee of the University of Melbourne. The participants were deemed to be healthy if they had no identifiable movement dysfunction; no history of significant spinal, hip, or knee pathology; and no presence of vertebrofemoral pain requiring treatment during the preceding 6 months. The Western Ontario and McMaster Universities Arthritis Index (WOMAC) Questionnaire (Likert scale)42 was used to determine the presence of any pain (5 items), stiffness (2 items), or functional difficulty (17 items) in this group. Study Design/Instrumentation A 2D Peak Motus video analysis system (PEAK)* was used to evaluate the STS transfer from a seat set at 100% of knee height (thigh horizontal). Based on a previous protocol,39 a single camera was positioned at a distance of 6.5 m, perpendicular to the sagittal plane. Six reflective spherical markers with black bases were attached over the midline thoracolumbar spine and pelvic landmarks, and 4 flat circular markers were placed over the right lateral aspect of the lower limb in an area that minimized skin movement on the lateral thigh (Fig. 1). Good testretest reliability for skin marker placement was established in a
* Peak Performance Technologies Inc, 7388 S Revere Pkwy, Englewood, CO 80112.
April 2010
Figure 1. (Left) Marker placement on body landmarks. (Right) Diagram illustrating method of calculation of thoracic, lumbar, hip, and knee flexion-extension angles (for definition of zero reference position, refer to text). PSIS⫽posterior superior iliac spine, ASIS⫽anterior superior iliac spine, 2/3Th⫽proximal thigh, SC⫽supracondylar, LTC⫽lateral tibial condyle, LM⫽10 cm above lateral malleolus.
standing position (intraclass correlation coefficient [1,1]⫽.80 –.93). Figure 1 illustrates angle definitions and calculation. Zero thoracic spine flexion or extension occurred when straight lines joining markers on the 1st and 3rd thoracic spinous processes (T1-T3) and markers on the 11th thoracic and 1st lumbar spinous processes (T11-L1) intersected at 0 degrees. Zero lumbar and hip angles were defined when a straight line joining the T11 and L1 markers or the proximal thigh (2/3Th) and supracondylar (SC) markers was per-
pendicular to the pelvic plane. The pelvic plane comprises a straight line joining posterior and anterior superior iliac spines (PSIS and ASIS). Zero knee flexion or extension occurred when straight lines joining the 2/3Th and SC markers and the LTC (lateral tibial condyle) and LM (10 cm above lateral malleolus) markers intersected at zero. Any spinal angle anterior to the zero position was called “x” degrees of flexion, whereas angles posterior to the zero reference represented “y” degrees of extension.
Volume 90
Number 4
Physical Therapy f
563
Kinematics of Rising From a Chair With arms folded, the participants performed 3 STS trials at a selfselected pace. Prior to videotaping, participants were encouraged to adopt their most comfortable anteroposterior (bare) foot position. They also performed tests for full available thoracic spine flexion (Fig. 2A), lumbar spine flexion (Fig. 2B), hip flexion (Fig. 2C), and thoracic spine extension (Fig. 2D). Before performing 3 spinal flexibility trials, participants were given 3 practice trials to familiarize themselves with the specific movement. One (best) videotaped image of STS performance (judged from the videotape by the smoothness of the movement and the lack of out-of-plane motion) and the last of each of the spinal flexibility trials were automatically digitized for each participant using the 2D PEAK software program. The data then were converted to angles and smoothed using a fourth-order Butterworth (high cutoff) filter43 at an optimum cutoff frequency determined by the software.44 The accuracy and reliability of PEAK for uniplanar measurement of joint angles have been established previously.45 All STS angular data then were imported into the Microsoft Excel program† and used in conjunction with Kaleidagraph version 3.8‡ to normalize the data to 100% movement duration. In this study, 3 events were defined: (1) the start of the STS movement as the point of 10% increase in the horizontal (x) displacement of the T1 marker; (2) buttocks lift-off (LO) at the point of a 10% increase in the vertical (y) displacement of the proximal thigh (2/3Th) marker, as described by Mourey and colleagues29,32; and (3) the end of the † Microsoft Corp, One Microsoft Way, Redmond, WA 98052-6399. ‡ Synergy Software, 2457 Perkiomen Ave, Reading, PA 19606.
564
f
Physical Therapy
Volume 90
STS movement at the point of no further hip extension. For the purpose of analysis, the STS movement then was divided into a pre-LO phase and a post-LO phase. Data Analysis Means and standard deviations for the excursions (end angle minus start angle), ranges used in the STS task (maximum angle minus minimum angle), angles at LO, and percentage of maximum joint and segment angles used during STS performance were calculated. The hip/lumbar ratio was calculated for each phase by dividing the range of hip movement by the lumbar spine range of movement. The thoracic/ lumbar ratio was calculated in a similar manner.
Results The results of the WOMAC Questionnaire confirmed that the participants had no pain (mean [SD] score⫽0 [0] out of 20), no significant stiffness (mean [SD] score⫽1 [1] out of 8), and no functional difficulty (mean [SD] score⫽2 [5] out of 68), suggesting that this group comprised elderly people who were healthy. Movement Pattern Figure 3 demonstrates the participants’ angular movement pattern throughout the STS task. The figure shows that in the pre-LO phase, concurrent hip and lumbar flexion combined with thoracic extension comprised forward trunk lean. This interaction reversed in the post-LO phase, as the hips and knees and the lumbar spine extended while the thoracic spine flexed. Lift-off occurred at 30.1% of STS duration and corresponded to an averaged 3.6° knee extension from a sitting flexion angle of 100.2 degrees. At LO, the hip and knee joints were similarly flexed (96.0° [8.2°] and 96.6° [8.1°], respectively), after which these joints were locked to-
Number 4
gether in a linear pattern of extension to raise the body into the standing position. The sequence involved in change of movement direction around LO was as follows. The knee commenced extending at 20% of STS duration. The lumbar spine began to extend (23%) before the hip (27%), and the thoracic spine changed from extension to flexion at LO (30%). At LO, trunk forward lean was 50.8 degrees (9.4°) with respect to the horizontal plane. After LO, trunk forward lean was maintained, reaching a maximal angle of 49.8 degrees (9.1°) with respect to the horizontal (1° more flexion) at 35% of STS duration. Thus, there was a brief transition or balancing period occupying 5% of the total movement duration, where although the hips and knees were extending so that the buttocks were lifting from the chair, the trunk remained flexed forward. Angular Changes The excursion and range of movement of the sagittal thoracic spine and lumbar spine, hip and knee joints, and trunk slope (relative to the horizontal plane) during the STS task are shown in Table 1. The hip moved 19.4 degrees from the start position (77.1°) to its maximal flexion angle (96.5°) immediately prior to LO and was the dominant joint responsible for bringing the trunk forward. Although the lumbar spine moved through a range of 20.1 degrees from a sitting position to a standing position, it contributed only 4.1 degrees of flexion in the pre-LO phase, from a start angle of 2.7 degrees in the sitting position. Thus, the hip/lumbar ratio in the pre-LO phase was 4.7:1; that is, for every 4.7 degrees of hip flexion, there was 1 degree of lumbar flexion. The thoracic spine also extended through a relatively small range so that the thoracic flexion angle reduced from the initial 37.1 degrees in the sitting position to 30 degrees at LO. During the pre-LO April 2010
Kinematics of Rising From a Chair
Figure 2. Tests for maximal available (A) thoracic spine flexion, (B) lumbar spine flexion, (C) hip flexion, and (D) thoracic spine extension.
April 2010
Volume 90
Number 4
Physical Therapy f
565
Kinematics of Rising From a Chair phase, every 1 degree of lumbar flexion was accompanied by 1.7 degrees of thoracic extension. During the post-LO phase, there was significantly more range of movement for the lumbar spine (19.2°); however, there also was a greater increase in hip joint range of movement (98.8°), so that the hip/lumbar ratio was 5.2:1. The thoracic/lumbar ratio during the post-LO phase showed that for every 1 degree of lumbar extension, there was 0.4 degree of thoracic flexion (Tab. 2).
Figure 3. Mean sagittal thoracic and lumbar spine, hip and knee joints, and trunk slope angles plotted against normalized test duration throughout the sit-to-stand task from horizontal seat height in elderly adults who were healthy (N⫽41). A downward slope indicates extension, and an upward slope indicates flexion, except in the case of trunk slope. LO⫽buttocks lift-off, 2D⫽2-dimentional.
Table 3 indicates the percentage of full available flexion-extension angle of the joints and regions used during the STS task. During the STS task, 90% of maximal available hip joint flexion was used during the pre-LO phase. However, the proportion of maximal lumbar flexion angle used was only 23%. Maximal available thoracic extension was 24.3 (11.3) degrees of flexion. Therefore, the maximum range of available thoracic extension is the arithmetic difference between maximal available thoracic flexion (66.4°) and extension (24.3°); that is, 42.1 (9.6) degrees. However, the maximum thoracic extension angle used during the STS task (equal to minimum STS thoracic
Table 1. Excursion and Range of Movement (in Degrees) of Participating Joints and Regions During the Sit-to-Stand (STS) Task in Elderly Adults Who Were Healthy (N⫽41)a Start of STS Task
At Buttocks Lift-off
End of STS Task
Maximum
Minimum
Range
Mean (SD), 95% CI
Mean (SD), 95% CI
Mean (SD), 95% CI
Mean (SD), 95% CI
Mean (SD), 95% CI
Mean (SD), 95% CI
Thoracic spine
37.1 (10.3), 34.0 to 40.2
30.0 (10.6), 26.7 to 33.3
38.1 (9.8), 35.1 to 41.1
38.1 (9.9), 35.1 to 41.1
30.0 (11.1), 26.6 to 33.4
8.1 (2.8), 7.2 to 9.0
Lumbar spine
2.7 (8.8), 0.0 to 5.4
5.9 (8.9), 3.2 to 8.6
⫺13.3 (7.0), ⫺15.4 to ⫺11.2
6.8 (10.0), 3.7 to 9.9
⫺13.3 (7.0), ⫺15.4 to ⫺11.2
20.1 (8.5), 17.5 to 22.7
Hip joint
77.1 (8.2), 74.6 to 79.6
96.0 (8.2), 93.5 to 98.5
⫺2.8 (6.9), ⫺4.9 to ⫺0.7
96.5 (8.8), 93.8 to 99.2
⫺2.8 (6.9), ⫺4.9 to ⫺0.7
99.3 (9.2), 96.5 to 102.1
Knee joint
100.2 (8.1), 97.7 to 102.7
96.6 (8.1), 94.1 to 99.1
1.2 (5.8), ⫺0.6 to 3.0
100.2 (8.1), 97.7 to 102.7
1.2 (5.8), ⫺0.6 to 3.0
99.0 (9.9), 96.0 to 102.0
Trunk slope (to horizontal)
81.2 (4.5), 79.8 to 82.6
50.8 (9.4), 47.9 to 53.7
84.7 (3.3), 83.7 to 85.7
84.8 (3.3), 83.8 to 85.8
49.8 (9.1), 47.0 to 52.6
35.0 (8.6), 32.4 to 37.6
Region/Joint
a
Positive values denote flexion movement; negative values denote extension movement. 95% CI⫽95% confidence interval.
566
f
Physical Therapy
Volume 90
Number 4
April 2010
Kinematics of Rising From a Chair Table 2. Hip/Lumbar and Thoracic/Lumbar Ratios and 95% Confidence Interval (CI) for Mean (SD) Values During the Sit-to-Stand Movement in Elderly Adults Who Were Healthy (N⫽41) Range (°)
a
Phase
Hip Joint
Lumbar
Pre-LOa
19.4 (6.3)
4.1 (5.3)
Post-LO
98.8 (9.8)
19.2 (8.7)
Hip/Lumbar Ratio
Range (°)
Thoracic/ Lumbar Ratio
Thoracic
Lumbar
4.7:1 95% CI⫽4.3–5.1
7.1 (4.5)
4.1 (5.3)
1.7:1 95% CI⫽1.3–2.1
5.2:1 95% CI⫽4.7–5.7
8.1 (4.4)
19.2 (8.7)
0.4:1 95% CI⫽0.1–0.7
LO⫽buttocks lift-off.
flexion) was 30.0 (11.1) degrees of flexion, so that in this group there was only 5.7 degrees left for any further thoracic extension. Therefore, the maximum thoracic extension used during STS corresponded to 86.5% (9.8%) of available thoracic extension. Duration The mean (SD) time to stand up from a sitting position on a horizontal (thigh) seat, with arms folded across the chest and at a self-selected speed, was 2.0 (0.4) seconds.
Discussion This study used a revised model of marker location to describe for the first time the coordinated sagittal movement pattern between the hips and knees and the lumbar spine and thoracic spine during the STS task in community-dwelling elderly people who were healthy. In addition, the proportion of full available flexionextension angle used by the joints and regions during the STS task was determined. The use of a 2D method was supported by Baer and Ash-
burn33 and Shum et al,38 who concluded that out-of-sagittal-plane movements during the STS task were insignificant in individuals. Hip-Spine Movement Interaction The finding of concurrent hip and lumbar flexion during the pre-LO phase contradicts the statement in a clinical text that “flexion of the extended trunk at the hips”16(p143) is critical for effective STS performance and the belief that flexing the lumbar spine is a “trick” or compensatory movement.16 Although hip joint flexion remained the dominant factor in bringing the body mass forward, a small degree of lumbar flexion was nevertheless a component of trunk forward lean. These findings highlight errors in current beliefs regarding the spinal contribution to the STS movement that have resulted from use of a rigid trunk model in image-based studies with markers on proximal and distal ends of the trunk segment.25,27,31,32 These incorrect concepts regarding the movement pattern have the potential to adversely influence the assessment and
retraining of patients with STS dysfunction. The model used in this study also provided a more accurate measurement of the true hip or pelvicfemoral angle. Use of the shouldergreater trochanter-knee marker placement has largely been responsible for the variation in sagittal hip joint angles reported in STS studies.19,20,27,46,47 Close to full available hip flexion (90%) was used by our participants during the pre-LO phase, indicating that hip limitation would pose a major problem in the kinematic chain. Possible compensation would include increased lumbar flexion or, in the case of very limited hip movement, thoracic flexion instead of extension to bring the trunk sufficiently forward and at an appropriate velocity. It is not clear why our elderly group used only 23% of their maximal available lumbar flexion angle during the pre-LO phase; however, sufficient range of movement remained to allow lumbar compensation for a stiff hip joint, if needed.
Table 3. Percentage of Maximal Available Mean Flexion and Extension Angles and 95% Confidence Interval (CI) for Thoracic Spine, Lumbar Spine, and Hip Joint Used During the Sit-to-Stand (STS) Task in Elderly Adults Who Were Healthy (N⫽41) Maximal Angle (°) Available
Maximal Angle (°) Used in STS Task
% Maximal Angle Used in STS Task
Region/Joint
Mean (SD)
Mean (SD)
Mean (SD)
Thoracic flexion
66.4 (7.9)
38.1 (9.9)
Lumbar flexion
29.6 (8.8)
Hip flexion
April 2010
107.2 (10.6)
6.8 (10.0) 96.5 (8.8)
95% CI
P
57.4 (11.2)
53.9–60.9
.03
23.0 (10.7)
17.7–28.3
.01
90.0 (6.4)
87.8–92.2
.01
Volume 90
Number 4
Physical Therapy f
567
Kinematics of Rising From a Chair Our results also indicate that diminished mobility of the thoracic spine, which moves in the opposite direction to the lumbar spine and requires close to full (86.5%) extension in the pre-LO phase, may be a problem in some individuals. A key finding of the study by Ikeda et al17 was that the older participants did not extend their heads as the trunk was flexing forward, so that the older group was facing down at LO. In our group, the hips provided a large flexion angle (96.5°); however, the small flexion contribution (4.1°) from the lumbar spine may have put more demand on the adjacent thoracic spine to bring the center of the upper body mass forward over the feet. Thus, as the hips and lumbar spine flexed, the thoracic spine extended only 8.1 (2.8) degrees, despite some further (5.7°) thoracic extension being available. A review of the videotapes revealed that our participants did not appear to be looking down at LO. However, for people with larger thoracic curvatures (eg, elderly women with postmenopausal kyphosis), the lower cervical spine may not have the range of extension to fully compensate for a lack of thoracic extension. The role of vision in aiding postural control has been confirmed in previous studies.48,49 The ability to look directly forward at LO enables people to orient themselves with the vertical and horizontal features of the environment. This orientation provides visual feedback that is important in elderly people to help control the velocity of the center of mass and thus maintain control of the movement and balance during the STS task.29 It follows that elderly people with increased kyphosis or thoracic stiffness may require thoracic mobilizing exercises to enable optimal STS performance. It is difficult to compare the timing of LO in the present study with that in previous studies, as researchers have used different definitions for 568
f
Physical Therapy
Volume 90
the start and end of the STS movement and the LO event. However, the sequence of change in direction of the knee first and then the hip joints before LO is similar to that reported by other authors.4,27 In addition, our study demonstrated that the onset of knee extension at 20% of STS duration was closely followed by lumbar extension (23%) and then hip joint extension (27%). However, the thoracic spine did not change from extension to flexion until LO at 30% of STS duration. Carr and Shepherd16 recommended that for optimum STS performance, the body mass should move forward and upward without a pause between the 2 movements. However, our participants demonstrated a brief transitional or balancing phase, where the trunk remained tipped forward following LO for 5% of STS duration. This phase appears to have been due to the knees extending approximately 3.5 degrees, whereas the lumbar spine and hips had only extended 1 to 2 degrees in this time. In addition, as the thoracic spine changed from extension to flexion at LO, this change may have added a small amount to trunk forward lean immediately post-LO. This brief balancing phase prior to the commencement of trunk extension appears not to have been reported previously, probably due to the use of different models for measurement of joint angles. It has been noted that trunk movement must be sufficient to propel the upper body mass forward50; however, this trunk movement must be limited to prevent the possibility of falling forward at LO.29 Furthermore, the importance of postural stability around LO by controlling the center of mass in relation to the foot support area has been emphasized.16 Thus, the priority of the elderly group in gaining stability before attempting to rise46 may explain this small delay in trunk extension.
Number 4
There was a relatively large variation of spinal angular displacement during the STS task, as indicated by the standard deviations in the movement data. It is not unusual to see a large degree of variation in the erect thoracolumbar posture in people who are healthy. Stagnara et al51 suggested that the differences in spinal curvatures among participants in their study was so great that the average curve values were of little use as normative data. This variation may be exaggerated in elderly adults with interacting effects of age. It is possible that habitual postural differences may be associated with inter-subject variations in the thoracolumbar start posture in the sitting position and reflected in differences in the mobility of the spinal regions. Compared with the young participants in our previous study,39 the elderly participants in the present study had a different start posture in the sitting position, with increased thoracic kyphosis (37.1° versus 32.2°) and a much straighter lumbar spine (2.7° versus 14.5° of flexion). The movement pattern was similar in both groups, with concurrent lumbar flexion and thoracic extension accompanying hip flexion during the pre-LO phase and a reversal of these movements in the post-LO phase. The sequence of change in movement direction of the joints and regions prior to LO also was similar in both groups. Although our young and old groups had a similar maximum hip flexion angle (98.9° versus 96°), the smaller contribution of lumbar flexion in the elderly group (4.1° versus 7.0°) resulted in a larger pre-LO hip-lumbar ratio (4.7:1 versus 3.1:1). The older participants also extended their thoracic spine less than the younger participants (8.1° versus 14.6° of flexion) in the pre-LO phase. Using 3D motion analysis, Farquhar et al52 reported the lower-limb kinematics during the STS task in a conApril 2010
Kinematics of Rising From a Chair trol group of people of similar age (62 [6.3] years) who were healthy. Using a pelvic-femoral angle, their graph (Fig. 1)52(p570) indicated a similar occurrence of maximal hip flexion immediately before LO. However, in comparison with our group, the sitting hip flexion angle was only 60 degrees compared with 77 degrees, increasing to a maximum of 76 degrees versus 96.5 degrees in the pre-LO phase. The maximum knee flexion angles (79°– 89°) versus 100 degrees in our group indicated that Farquhar and colleagues’ participants were sitting on a higher chair, which possibly explains the different values obtained. These authors did not attempt to measure trunk slope or spinal movement. Shum and colleagues38,53 appear to be the only other authors who have measured sagittal spinal angles during the STS task, although in a different age group (41.7 [8.2] years). Using an electromagnetic tracking device with sensors on the L1 spinous process, posterior sacrum, and lateral thigh, they demonstrated a similar hip-spine interaction in controls who were healthy that supported our results. Lumbar flexion led the hips during the pre-LO phase, and the hips extended more rapidly than the lumbar spine in the post-LO phase. Although agreeing with the pattern of joint interaction, Shum et al38 reported different values for hip and lumbar angles. Their sitting hip flexion angles were approximately 45 (9) degrees compared with 76.9 (7.3) degrees in our study, which again could be partially explained by the slightly higher seat height (110% versus 100%) in our study. The sitting lumbar angles of their middleaged group were similar to those of our young group39 (14° versus 14.5°), with both groups showing considerably more lumbar flexion than the elderly participants in the present study (2.7°). Maximal hip flexion angles for the middle-aged April 2010
group versus our young39 and old groups were 89 (11), 98.9 (6.5), and 96.5 degrees, respectively, which again may be explained by the higher seat, which required less trunk forward lean to rise. The major differences occurred in the maximal lumbar flexion angles used during the STS task, which were 41.8 (8) degrees in the middle-aged group38 versus 21.5 (9.2) degrees and 6.8 (10) degrees, respectively, for our young39 and old groups. These findings mean that during the pre-LO phase, our young group39 flexed the lumbar spine only 7 degrees as opposed to 27 degrees in Shum and colleagues’38 middle-aged group. The explanation for the differences among studies is likely found in the use of different measurement tools. For example, it is not clear whether the electromagnetic device, which measures the tilt of the sacrum as opposed to the plane of the pelvis (PSIS-ASIS), would provide the same angle values. In their radiographic study of 109 people with low back pain (aged 21– 83 years), Lord et al54 demonstrated that lumbar lordosis (from L1 to S1) averaged 49 degrees in a standing position and 34 degrees in a sitting position, which indicates that our surface-based studies have all underestimated the true bony lumbar posture. It also has been demonstrated that skin movement on the pelvis results in a small underestimation of maximal lumbar flexion angles (trunk toward thighs), more so in elderly people (average of 6.5°).55 Nevertheless, despite the value differences observed, the overall pattern of hip-lumbar spine movement interaction was similar in these STS studies. Hip-Knee Interaction Although not the focus of this investigation, the essential role of hip and knee extension in raising the body mass to a standing position must be acknowledged. Without effective
hip and knee joint extension postLO, people may be forced to sit back down in the chair.32,56 However, unexpected findings in this group of elderly participants rising from a thigh-horizontal seat height were the similar angle (96.0°) of hip and knee flexion at LO and the subsequent locking together of these joints in a linear pattern of extension. Presumably, this hip-knee movement interaction occurs to take the head upward in the shortest possible path, as indicated by the close to vertical displacement of the body’s center of mass in the extension phase demonstrated by Roebroeck et al.26 Although lower-limb kinematics have been studied by many researchers,7,18,40,41 there are 2 major reasons why this close interaction between hip and knee joints appears not to have been reported: First, many researchers have placed the feet in a standardized position that may have prevented the natural kinematics of STS movement from being demonstrated. For example, Vander Linden et al7 and Schenkman et al18 used a dorsiflexion-forward leg angle of 18 degrees with respect to the vertical, which affects the knee angle at LO in people of different heights. In contrast, our participants practiced the STS task 2 to 3 times prior to videotaping and were encouraged to adopt their most comfortable anteroposterior foot position. Thus, we contend that the movement kinematics were close to natural for the STS task in our group. Second, use of the conventional model of marker placement by previous researchers (acromion, greater trochanter, and knee) resulted a hip flexion angle contaminated not only by the inclusion of spinal movement but also by possible protraction of the scapula and acromion (frequently observed as individuals attempt to rise), so that measurement error prevented the recording of true locking together of hip and knee joint extension.
Volume 90
Number 4
Physical Therapy f
569
Kinematics of Rising From a Chair In summary, although most previous authors ignored the spinal contribution to STS performance, the results of the present study demonstrate that in elderly people who are healthy, concurrent hip and lumbar flexion and thoracic extension position the upper body (trunk) segment in an appropriate forward lean posture during the pre-LO phase, and that this movement interaction reverses the post-LO phase. We acknowledge that, as with previous STS studies, this is a surface-based analysis that may not truly represent the movement of the underlying skeleton. In addition, our sample of elderly people appears to have been biased toward the top end of physical health, as indicated by the WOMAC results. Thus, the findings may not be generalizable to all elderly people, such as those with obesity or greater levels of stiffness. Nevertheless, the results of this study provide knowledge of the sagittal kinematics of the hip-spine interaction during the STS task in elderly people who are healthy. This knowledge has practical implications for physical therapists providing both prophylactic and rehabilitation programs. Physical therapists are encouraged to use this knowledge to provide correct facilitation of this important daily activity. For example, encouraging patients to practice anterior pelvic tilting (lumbar extension) in a sitting position as a preparatory exercise for the STS task is based on incorrect assumptions. Attention to retraining of correct movement patterns of the spinal segments is important because of their potential to affect trunk movement during the STS task, as well as the movement of other segments in the kinematic chain such as hip or knee joints or cervical spine. It also seems advisable for elderly patients or those with pathology to practice spinal mobilizing exercises, in partic-
570
f
Physical Therapy
Volume 90
ular thoracic extension, to enable an optimal STS performance. For future research, we recommend that a standardized pelvifemoral hip angle be used to provide more valid data and permit comparisons across kinematic studies. Further development of our 2D model to include the cervical spine and sagittal head tilt is in progress. This model will enable a detailed description of the hip-spine interaction during the STS task from different seat heights or during tasks such as sitting down from a standing position or standing to walk. The effect of limited sagittal movement in any of the component joints or of an intervention such as mobilizing exercises on the STS kinematics also should be investigated. All authors provided concept/idea/research design, data analysis, project management, and consultation (including review of manuscript before submission). Dr Fotoohabadi and Dr Tully provided writing. Dr Fotoohabadi provided data collection. Dr Tully provided participants and facilities/ equipment. This research was carried out in partial fulfillment of Dr Fotoohabadi’s PhD degree requirements at the University of Melbourne. This project had ethics approval from the Human Research Ethics Committee of the University of Melbourne. A platform presentation of this research was given at the Movement Analysis 2005: Building Bridges Conference; February 3–5, 2005; University of Auckland (Tamaki Campus), New Zealand. This article was received March 19, 2009, and was accepted October 7, 2009. DOI: 10.2522/ptj.20090093
References 1 Burdett RG, Habasevich R, Pisciotta J, Simon SR. Biomechanical comparison of rising from two types of chairs. Phys Ther. 1985;65:1177–1183. 2 Eriksrud O, Bohannon RW. Relationship of knee extension force to independence in sit-to-stand performance in patients receiving acute rehabilitation. Phys Ther. 2003; 83:544 –551.
Number 4
3 Guralnik JM, Ferrucci L. Assessing the building blocks of function: utilizing measures of functional limitation. Am J Prev Med. 2003;25(3 suppl 2):112–121. 4 Hughes MA, Myers BS, Schenkman ML. The role of strength in rising from a chair in the functionally impaired elderly. J Biomech. 1996;29:1509 –1513. 5 Tiedemann A, Shimada H, Sherrington C, et al. The comparative ability of eight functional mobility tests for predicting falls in community-dwelling older people. Age Ageing. 2008;37:430 – 435. 6 Dehail P, Bestaven E, Muller F, et al. Kinematic and electromyographic analysis of rising from a chair during a “Sit-to-Walk” task in elderly subjects: role of strength. Clin Biomech (Bristol, Avon). 2007;22:1096 –1103. 7 Vander Linden DW, Brunt D, McCulloch MU. Variant and invariant characteristics of the sit-to-stand task in healthy elderly adults. Arch Phys Med Rehabil. 1994;75: 653– 660. 8 Yamada T, Demura S. Relationships between ground reaction force parameters during a sit-to-stand movement and physical activity and falling risk of the elderly and a comparison of the movement characteristics between the young and the elderly. Arch Gerontol Geriatr. 2009;48:73– 77. 9 Puthoff ML, Nielsen DH. Relationships among impairments in lower-extremity strength and power, functional limitations, and disability in older adults. Phys Ther. 2007;87:1334 –1347. 10 Shepherd RB, Carr JH. Reflections on physiotherapy and the emerging science of rehabilitation. Aust J Physiother. 1994;40: 39 – 47. 11 Hirvensalo M, Rantanen T, Heikkinen E. Mobility difficulties and physical activity as predictors of mortality and loss of independence in the community-living older population. J Am Geriatr Soc. 2000;48: 493– 498. 12 Janssen G, Bussmann B, Stam H. Determinants of the sit-to stand movement: a review. Phys Ther. 2002;82:866 – 879. 13 Galli M, Cimolin V, Crivellini M, Campanini I. Quantitative analysis of sit to stand movement: experimental set-up definition and application to healthy and hemiplegic adults. Gait Posture. 2008;28:80 – 85. 14 Lord SR, Murray SM, Chapman K, et al. Sit-to-stand performance depends on sensation, speed, balance, and psychological status in addition to strength in older people. J Gerontol A Biol Sci Med Sci. 2002; 57:M539 –M543. 15 Bohannon RW. Knee extension strength and body weight determine sit-to-stand independence after stroke. Physiother Theory Pract. 2007;23:291–297. 16 Carr JH, Shepherd RB. Standing up and sitting down. In: Carr JH, Shepherd RB, eds. Stroke Rehabilitation: Guidelines for Exercise and Training to Optimize Motor Skill. Edinburgh, Scotland: ButterworthHeinemann; 2003:31–144.
April 2010
Kinematics of Rising From a Chair 17 Ikeda E, Schenkman M, Riley HW. Influence of age on dynamics of rising from a chair. Phys Ther. 1991;71:473– 481. 18 Schenkman M, Riley PO, Pieper C. Sit to stand from progressively lower seat heights: alterations in angular velocity. Clin Biomech (Bristol, Avon). 1996;11: 153–158. 19 Alexander NB, Schultz AB, Warwick DN. Rising from a chair: effects of age and functional ability on performance biomechanics. J Gerontol. 1991;46:M91–M98. 20 Gross MM, Stevenson PJ, Charette SL, et al. Effect of muscle strength and movement speed on the biomechanics of rising from a chair in healthy elderly and young women. Gait Posture. 1998;8:175–185. 21 Bernardi M, Rosponi A, Castellano V, et al. Determinants of sit-to-stand capability in the motor impaired elderly. J Electromyogr Kinesiol. 2004;14:401– 410. 22 Dubost V, Beauchet O, Manckoundia P, et al. Decreased trunk angular displacement during sitting down: an early feature of aging. Phys Ther. 2005;85:404 – 412. 23 Manckoundia P, Buatois S, Gueguen R, et al. Clinical determinants of failure in balance tests in elderly subjects. Arch Gerontol Geriatr. 2008;47:217–228. 24 Millington PJ, Myklebust BM, Shambes GM. Biomechanical analysis of sit to stand motion. Arch Phys Med Rehabil. 1992;73: 609 – 617. 25 Nuzik S, Lamb R, VanSant A, Hirt S. Sit-tostand movement pattern: a kinematic study. Phys Ther. 1986;66:1708 –1713. 26 Roebroeck M, Doorenbosch C, Harlaar J, et al. Biomechanics and muscular activity during sit-to-stand transfer. Clin Biomech. 1994;9:235–244. 27 Shepherd RB, Gentile AM. Initial trunk position and biomechanical consequences in standing up. Hum Mov Sci. 1994;13:817– 840. 28 Etnyre B, Thomas DQ. Event standardization of sit-to-stand movements. Phys Ther. 2007;87:1651–1666; discussion 1667–1668. 29 Mourey F, Grishin A, d’Athis P, et al. Standing up from a chair as a dynamic equilibrium task: a comparison between young and elderly subjects. J Gerontol A Biol Sci Med Sci. 2000;55:B425–B431. 30 Yamada T, Demura S. Influence of load burdens on lower limbs in each movement phase and the characteristics of sitto-stand movement. Sport Sci Health. 2007;2:8 –15.
April 2010
31 Park ES, Park CI, Lee HJ, et al. The characteristics of sit-to-stand transfer in young children with spastic cerebral palsy based on kinematic and kinetic data. Gait Posture. 2003;17:43– 49. 32 Mourey F, Pozzo T, Rouhier-Marcer I, Didier JP. A kinematic comparison between elderly and young subjects standing up from and sitting down in a chair. Age Ageing. 1998;27:137–146. 33 Baer GD, Ashburn AM. Trunk movements in older subjects during sit-to-stand. Arch Phys Med Rehabil. 1995;76:844 – 849. 34 Carr JH, Ow JE, Shepherd RB. Some biomechanical characteristics of standing up at three different speed: implications for functional training. Physiother Theory Pract. 2002;18:47–53. 35 Cahill BM, Carr JH, Adams R. Intersegmental co-ordination in sit-to-stand: an age cross-sectional study. Physiother Res Int. 1999;4:12–27. 36 Sibella F, Galli M, Romei M, et al. Biomechanical analysis of sit-to-stand movement in normal and obese subjects. Clin Biomech. 2003;18:745–750. 37 Davis RB, Ounpuu S, Tyberski DJ, Gage JR. A gait analysis data collection and reduction technique. Hum Mov Sci. 1991;10: 575–587. 38 Shum GL, Crosbie J, Lee RY. Effect of low back pain on the kinematics and joint coordination of the lumbar spine and hip during sit-to-stand and stand-to-sit. Spine. 2005;30:1998 –2004. 39 Tully EA, Fotoohabadi MR, Galea MP. Sagittal spine and lower limb movement during sit-to-stand in healthy young subjects. Gait Posture. 2005;22:338 –345. 40 Mak MKY, Levin O, Mizrahi J, Hui-Chan CWY. Joint torques during sit-to-stand in healthy subjects and people with Parkinson’s disease. Clin Biomech (Bristol, Avon). 2003;18:197–206. 41 Pai YC, Rogers MW. Control of body mass transfer as a function of speed of ascent in sit-to-stand. Med Sci Sports Exerc. 1990;22: 378 –384. 42 Bellamy N. WOMAC Osteoarthritis Index User Guide VI. Brisbane: Australia; 2003. 43 Winter DA. The Biomechanics and Motor Control of Human Movement. 3rd ed. New York, NY: John Wiley & Sons Inc; 2005. 44 Jackson KM. Fitting of mathematical functions to biomechanical data. IEEE Trans Biomed Eng. 1979;26:122–124.
45 Selfe J. Validity and reliability of measurements taken by the PEAK 5 motion analysis system. Med Eng Tech. 1998;22:220 – 225. 46 Schultz AB, Alexander NB, Ashton-Miller JA. Biomechanical analyses of rising from a chair. J Biomech. 1992;25:1383–1391. 47 Shepherd RB, Koh HP. Some biomechanical consequences of varying foot placement in sit-to-stand in young women. Scand J Rehabil Med. 1996;28:79 – 88. 48 Berencsi A, Ishihara M, Imanaka K. The functional role of central and peripheral vision in the control of posture. Hum Mov Sci. 2005;24:689 –709. 49 Nougier V, Bard C, Fleury M, Teasdale N. Contribution of central and peripheral vision to the regulation of stance: developmental aspects. J Exper Child Psychol. 1998;68:202–215. 50 Riley P, Schenkman M, Mann R, Hodge WA. Mechanics of a constrained chair-rise. J Biomech. 1991;24:77– 85. 51 Stagnara P, Mauro JC, de Gran G, et al. Reciprocal angulation of vertebral bodies in a sagittal plane: approach to references for the evaluation of kyphosis and lordosis. Spine. 1982;7:335–342. 52 Farquhar SJ, Reisman DS, Snyder-Mackler L. Persistence of altered movement patterns during a sit-to-stand task 1 year following unilateral total knee arthroplasty. Phys Ther. 2008;88:567–579. 53 Shum GL, Crosbie J, Lee RY. Threedimensional kinetics of the lumbar spine and hips in low back pain patients during sit-to-stand and stand-to-sit. Spine. 2007; 32:E211–E219. 54 Lord MJ, Small JM, Dinsay JM, Watkins RG. Lumbar lordosis: effects of sitting and standing. Spine. 199722:2571–2574. 55 Kuo YL, Tully EA, Galea MP. Skin movement errors in measurement of sagittal lumbar and hip angles in young and elderly subjects. Gait Posture. 2008;2:264 – 270. 56 Yu B, Holly-Crichlow N, Brichta P, et al. The effects of the lower extremity joint motions on the total body motion in sit-tostand movement. Clin Biomech (Bristol, Avon). 2000;15: 449 – 455.
Volume 90
Number 4
Physical Therapy f
571
Research Report
Construct Validity of Muscle Force Tests of the Rotator Cuff Muscles: An Electromyographic Investigation Rebecca L. Brookham, Linda McLean, Clark R. Dickerson R.L. Brookham, MSc, is a doctoral student in the Department of Kinesiology, University of Waterloo, Waterloo, Ontario, Canada. L. McLean, PhD, is Associate Professor, School of Rehabilitation Therapy, Queens University, Kingston, Ontario, Canada. C.R. Dickerson, PhD, is Assistant Professor, Department of Kinesiology, University of Waterloo, 200 University Ave W, Waterloo, Ontario, N2L 3G1 Canada. Address all correspondence to Dr Dickerson at:
[email protected]. [Brookham RL, McLean L, Dickerson CR. Construct validity of muscle force tests of the rotator cuff muscles: an electromyographic investigation. Phys Ther. 2010;90: 572–580.] © 2010 American Physical Therapy Association
Background. Manual muscle tests (MMTs) are used in clinical settings to evaluate the function and strength (force-generating capacity) of a specific muscle in a position at which the muscle is believed to be most isolated from other synergists and antagonists. Despite frequent use of MMTs, few electromyographic evaluations exist to confirm the ability of MMTs to isolate rotator cuff muscles. Objective. This study examined rotator cuff isolation during 29 shoulder muscle force tests (9 clinical and 20 generic tests).
Design. An experimental design was used in this study. Participants and Measurements. Electromyographic data were recorded from 4 rotator cuff muscles and 10 additional shoulder muscles of 12 male participants. Maximal isolation ratios (mean specific rotator cuff muscle activation to mean activation of the other 13 recorded muscles) defined which of these tests most isolated the rotator cuff muscles.
Results. Three rotator cuff muscles were maximally isolated (obtained highest isolation ratios) within their respective clinical test groups (lateral rotator test group for the infraspinatus and teres minor muscles and abduction test group for the supraspinatus muscle). The subscapularis muscle was maximally isolated equally as effectively within the generic ulnar force and clinical medial rotation groups. Similarly, the supraspinatus and teres minor muscles were isolated equally as effectively in some generic test groups as they were in their respective clinical test groups.
Limitations. Postural artifact in the wire electrodes caused exclusion of some channels from calculations. The grouping of muscle force tests based on test criteria (clinical or generic tests and muscle action) may have influenced which groups most isolated the muscle of interest. Conclusions. The results confirmed the appropriateness of 9 commonly used clinical tests for isolating rotator cuff muscles, but suggested that several other muscle force tests were equally appropriate for isolating these muscles.
Post a Rapid Response or find The Bottom Line: www.ptjournal.org 572
f
Physical Therapy
Volume 90
Number 4
April 2010
Construct Validity of Muscle Force Tests
M
uscle force tests that isolate the rotator cuff muscles allow for uncompromised interpretation of muscle function and strength (force-generating capacity). Identification of these muscle force tests will promote accurate assessment of weakness, which in turn could direct strategies for strengthening and injury prevention. Manual muscle tests (MMTs) are used to evaluate a muscle’s strength in a position in which it is believed to be most isolated from other muscle contributions.1 During an MMT, individuals exert maximal effort in defined postures against static manual resistance provided by a clinician. Because one cooperative function of the rotator cuff muscles is to maintain the humeral head within the glenoid cavity,2 true isolation (when all other surrounding muscles are inactive) is an unlikely state. Relative isolation of the rotator cuff is more physiologically realistic and is defined as occurring in a muscle force test for which the muscle of interest is most activated and when all other surrounding muscles are least activated (relative to each muscle’s maximum). The definition of relative isolation does not require maximal activation of the muscle of interest, but rather greater activation of that muscle relative to the mean activation of the other surrounding muscles. Few evaluations have confirmed the ability of MMTs to isolate rotator cuff muscles. Studies have identified muscle force tests that produce maximal activation of the rotator cuff muscles, but frequently omitted consideration of contributions from surrounding muscles and, therefore, did not confirm isolation.3–7 Kelly et al8 did consider contributions from surrounding synergistic muscles when performing an electromyographic (EMG) examination of the rotator cuff muscles (excluding the teres minor muscle) to identify isolation muscle force tests. Optimal MMTs were April 2010
determined based on 4 criteria: maximal activation of the cuff muscle, minimal activation from involved synergists, good test-retest reliability, and minimal positional pain provocation. Twenty-nine isometric muscle force tests were performed—27 core muscle force tests (3 exertion tests [elevation, lateral rotation, and medial rotation], 3 scapular elevation tests [0°, 45°, and 90°], and 3 humeral rotation tests [45°, 0°, and ⫺45°]) and 2 other tests (Gerber push-off test and Gerber push-off with force test). The authors8 concluded that the optimal isolation tests were: elevation at 90 degrees of scapular elevation and 45 degrees of lateral rotation for the supraspinatus muscle, lateral rotation at 0 degrees of scapular elevation and 45 degrees of medial rotation for the infraspinatus muscle, and Gerber push with force test for the subscapularis muscle. One limitation of the study by Kelly et al8 was that they gave no evidence for the assumed synergists of the rotator cuff, ignoring the possibility that shoulder muscle function changes with posture9 and, therefore, that rotator cuff synergists may change as posture changes. Furthermore, the study was limited to recording the EMG activity of 8 muscles, so it was possible that key synergists were not measured (eg, the EMG activity of the teres minor muscle was not recorded, and it has been found to act in synergy with the infraspinatus muscle in lateral rotation3,10). The Gerber push with force test was identified as the optimal MMT for the subscapularis muscle. This conclusion may be flawed because this test was assessed independently from the core muscle force tests; it was excluded from the analysis of variance (ANOVA), as it did not fit the format of the other tests. Because this test was excluded from the ANOVA and the integrated EMG activity was assessed only by
rank order, it is not known whether the lift-off test produced a significantly higher integrated EMG signal in the subscapularis muscle than in the other 28 muscle force tests. Jenp et al11 used a different technique to isolate the rotator cuff muscles: from 29 test postures, those muscle force tests that produced the largest EMG activity for the rotator cuff muscles of interest were identified as potential postures of isolation. Minimal activation of surrounding muscles (pectoralis major; anterior, middle, and posterior deltoid; and the other 3 rotator cuff muscles) then was assessed from only these potential postures to identify postures of isolation. The authors’ initial selection of postures that maximally activated the rotator cuff muscles may have eliminated other potential isolation postures because a muscle may not have to be in a state of maximal activation to be isolated. Further research is needed to evaluate the ability of MMTs to relatively isolate the rotator cuff muscles. The primary objective of this research was to evaluate rotator cuff relative isolation during 29 muscle force tests. We hypothesized that
Available With This Article at ptjournal.apta.org • eAppendix 1: Descriptions of the Muscle Force Test Groups (Part A), & Maximal Voluntary Contractor Testing Positions (Part B) • eAppendix 2: Mean (SD) Percentage of Muscle Activation for Each Muscle Force Test for All Participants • Audio Abstracts Podcast This article was published ahead of print on February 4, 2010, at ptjournal.apta.org.
Volume 90
Number 4
Physical Therapy f
573
Construct Validity of Muscle Force Tests muscle force tests commonly used by clinicians would be most effective in isolating the rotator cuff muscles and that these muscles would be most isolated in muscle force tests based on their respective primary actions (lateral rotation for the teres minor and infraspinatus muscles, medial rotation for the subscapularis muscle, and abduction for the supraspinatus muscle).
Method Participants Students were recruited using poster advertisements. Twelve right-hand– dominant male university students (mean age⫽20.7 years, range⫽18 – 29; mean height⫽180.6 cm, range⫽165.1–193.1; and mean weight⫽76.7 kg, range⫽49.0 – 88.2) participated in the experiment after providing informed consent. All participants exhibited full range of motion in the shoulder. Exclusion criteria included a history of upper-limb or low back injury within the previous 6 months and known neuromuscular, cardiovascular, or metabolic conditions that might affect the participants’ safety during muscle force tests. Electromyography After cleaning the skin surface with Betadine solution containing 10% povidone-iodine,* 4 bipolar intramuscular electrodes† were inserted into the supraspinatus, infraspinatus, and teres minor muscles (needle: 27 gauge, 30-mm length; placements similar to those described by Delagi and Perotto12) and into the subscapularis muscle (needle: 25 gauge, 50-mm length; placement similar to that described by Nemeth et al13) on the right side. The needles were removed, and the wires (44 gauge) remained in the muscle during testing. After cleaning the skin surface with * Purdue Products LP, One Stamford Forum, Stamford, CT 06901-3441. † CareFusion (formerly Viasys Healthcare Inc), 3750 Torrey View Ct, San Diego, CA 92130.
574
f
Physical Therapy
Volume 90
isopropyl alcohol, 10 silver-silver chloride bipolar surface electrodes (product #272)‡ were placed on the right side over the latissimus dorsi; triceps brachii (long head); biceps brachii; anterior, middle, and posterior fibers of deltoid; pectoralis major (sternal and clavicular insertions); and middle and upper trapezius muscles using the placements described by Cram and Kasman.14 Electrodes were interfaced with a Noraxon Telemyo 2400T G2 wireless transmitter.‡ The EMG signals were preamplified close to the source (gain of 500) and band-pass filtered (10 –1,500 Hz). The sampling rate was 4,000 Hz. Functional tests (maximal voluntary contraction [MVC] muscle force tests described in eAppendix 1, part B; available at ptjournal.apta.org) were performed to ensure proper electrode placement. Testing Protocol Prior to experimental trials, muscle-specific (2 each) MVCs were performed in recommended postures14 (eAppendix 1, part B; available at ptjournal.apta.org). Participants then performed 29 ramped, maximal-effort, isometric contractions with their right arm. These 29 muscle force tests were divided into 7 groups containing functionally similar muscle force tests (eAppendix 1, part A; available at ptjournal.apta. org). Nine of these muscle force tests were clinically used tests, which were divided into lateral rotation, medial rotation, and abduction muscle force test groups. Twenty additional generic muscle force tests were performed to assess further isolation possibilities and were organized into groups based on hand force direction (palmar, dorsal, radial, or ulnar). Resistance to muscle force tests was provided by a stationary steel beam. Muscle force tests ‡ Noraxon USA Inc, 13430 N Scottsdale Rd, Suite 104, Scottsdale, AZ 85254.
Number 4
were performed in a randomized order (within and between groups). Trials were 6 seconds in duration (participants were instructed to ramp up to maximal activation during the second to fourth seconds of each trial and then relax for the remaining time), and 2 minutes of rest was enforced between muscle force tests, as recommended by De Luca.15 EMG Analysis Electromyographic data were processed using Matlab 7.0.1 software.§ All raw EMG data were full-wave rectified and filtered using a single-pass, second-order, low-pass Butterworth filter (3-Hz cutoff frequency). Peak activation levels of filtered MVC trials were chosen as 100% activation, and experimental trials (29 muscle force tests) were normalized to these peaks by participant and muscle (expressed as %MVC). Mean muscle activations for each muscle for each muscle force test were calculated during the second to fourth seconds (during maximal activation) of each 6-second trial. Isolation Ratios To identify which muscle force tests most isolated the muscles of interest, we determined during what muscle force test there was a maximal amount of EMG activity in the muscle of interest, relative to a minimal EMG activation of all of the other recorded muscles. This was determined by an isolation ratio (IR) (equation 1), which was used to define muscular isolation for each rotator cuff muscle in each muscle force test. (1)
Isolation Ratio (IR) ⫽ (% MVC activity of rotator cuff muscle of interest/100) (兺%MVC of all other 13 recorded muscles/1,300)
§
The MathWorks Inc, 3 Apple Hill Dr, Natick, MA 01760-2098.
April 2010
Construct Validity of Muscle Force Tests To illustrate the meaning of an IR, consider IRs equal to zero, 1, and infinity: an IR of zero indicates that the rotator cuff muscle of interest was not activated (turned off); an IR of 1 indicates that the rotator cuff muscle of interest was activated equally as much as the mean activation of the other 13 recorded muscles; and an IR of infinity indicates that the rotator cuff muscle of interest was active when all the other 13 recorded muscles were not activated (turned off), representing true isolation. Furthermore, an IR greater than 1 (eg, 1.5) indicates the rotator cuff muscle of interest is activated more (eg, 1.5 times more) than the mean activation of the other 13 recorded muscles. Isolation ratios are affected most by other active muscles, such as synergistic muscles that contribute to the main action of the rotator cuff muscle of interest. Antagonistic muscles, which act in opposition to the main action of the rotator cuff muscles of interest, would be expected to be minimally active during rotator cuff MMTs. Therefore, antagonistic muscles would be expected to contribute very little to the mean activation of the muscles in the denominator of the IRs. Participants most often reached a maximal plateau between the second and fourth seconds of each trial. Therefore, mean IR values were calculated over this 2-second window. The sampled window was adjusted to accommodate for anomalies in the time-series EMG data of individual muscle force tests. Due to the sensitivity of intramuscular electrodes, visually identifiable artifacts (large, rapid spikes in amplitude) occasionally occurred within the 2-second window. Often it was possible to shift the window slightly to avoid the artifact (a total of 116 trials were adjusted). When this was not possible, data for that muscle on that trial were excluded from the analysis (excluded trials per total trials for the April 2010
infraspinatus, supraspinatus, teres minor, and subscapularis muscles were 14/348, 99/348, 68/348, and 54/348, respectively). Adjustments were made in the IR calculation for the other muscles to reflect these missing data (removal of the muscle with artifact and division by the correct number of remaining muscles in the equation). There were 175 total adjusted IRs, of which 77.7% removed only 1 muscle, 17.7% removed 2 muscles, and 4.6% removed 3 muscles from the denominator. Statistical Analysis Statistical analysis was performed using JMP IN 5.1.2 software.㛳 Four oneway, repeated-measures ANOVAs (one for each rotator cuff muscle) compared IRs across the muscle force test groups. The groups were first confirmed with a one-way ANOVA, which confirmed that there were no statistical differences (P⬍.05) among mean IRs for the muscle force tests within the functional groups. Homogenous variance was seen among groups when mean IR data were transformed to a natural logarithm. When the ANOVAs indicated significant differences among mean IRs, post hoc analyses (Tukey honestly significant difference test) were performed to determine significant differences among muscle force test groups (P⬍.05). A sample size estimate was not calculated a priori, but rather was driven by convenience and expense. Fatigue Analysis The first 2 muscle force tests performed by each participant were repeated at the end of testing. The EMG data were down-sampled to 2,048 Hz, fast Fourier analysis was performed, and mean and median power frequency values (MnPF, MdPF) were calculated. Paired t tests (one-tailed) assessed whether signif㛳
SAS Institute Inc, PO Box 8000, Cary, NC 27513.
icant changes in MdPF or MnPF were evident between the initial and final trials (P⬍.05). Percent differences in MdPF and MnPF values were calculated, as follows: (2) % Difference ⫽ Final Value ⫺ Initial Value Initial Value
冋
册
䡠 100
Role of the Funding Source This study was supported by funds from the Department of Kinesiology, University of Waterloo. The department had no role in the design, conduct, or reporting of this research.
Results Isolation of the Infraspinatus Muscle Differences in IRs for the defined muscle force test groups existed (P⬍.0001), and the mean maximal IR occurred in the clinical lateral rotation group (Fig. 1). There were significant differences among the IRs of the following groups: lateral rotation ⬎ medial rotation and ulnar, dorsal, and palmar force ⬎ radial force and abduction. Isolation of the Supraspinatus Muscle Differences in IRs for the defined muscle force test groups existed (P⬍.0001), and the mean maximal IR occurred within the clinical abduction muscle force test group (Fig. 2). There were significant differences among the IRs of the following groups: (1) abduction ⬎ medial rotation and palmar and ulnar force and (2) radial and dorsal force ⬎ ulnar force. Isolation of the Teres Minor Muscle Differences in IRs for the defined muscle force test groups existed (P⬍.0001), and the mean maximal IR occurred within the clinical lateral rotation muscle force test group (Fig. 3). There were significant differences between IRs in the follow-
Volume 90
Number 4
Physical Therapy f
575
Construct Validity of Muscle Force Tests ing groups: (1) lateral rotation and ulnar force ⬎ radial force and abduction and (2) medial rotation and palmar and dorsal force ⬎ abduction.
Figure 1. Mean isolation of the infraspinatus muscle using the isolation ratio (IR) among muscle force test groups. The clinical muscle force tests are within the medial rotation, lateral rotation, and abduction muscle force test groups (white bars). The generic muscle force tests are within the palmar, dorsal, radial, and ulnar force groups (blue bars). The results of the Tukey honestly significant difference test are shown; levels not connected by the same letter are significantly different. The error bars represent ⫺1 standard deviation. The highest mean (SD) IR (maximal isolation) was found in the lateral rotation group (2.29⫾1.17); the lowest mean IR was found in the abduction group (0.68⫾0.33).
Figure 2. Mean isolation of the supraspinatus muscle using the isolation ratio (IR) among muscle force test groups. The clinical muscle force tests are within the medial rotation, lateral rotation, and abduction muscle force test groups (white bars). The generic muscle force tests are within the palmar, dorsal, radial, and ulnar force groups (blue bars). The results of the Tukey honestly significant difference test are shown; levels not connected by the same letter are significantly different. The error bars represent ⫺1 standard deviation. The highest mean (SD) IR (maximal isolation) was found in the abduction group (1.84⫾1.27); the lowest highest mean IR was found in the ulnar group (0.96⫾0.60).
576
f
Physical Therapy
Volume 90
Number 4
Isolation of the Subscapularis Muscle Differences in IRs for the defined muscle force test groups existed (P⬍.0001), and the mean maximal IR occurred within the generic ulnar force muscle force test group (Fig. 4). There were significant differences between IRs in the following groups: (1) ulnar force ⬎ palmar force, lateral rotation, abduction, and dorsal and radial force; (2) medial rotation ⬎ lateral rotation, abduction, and dorsal and radial force; and (3) palmar force ⬎ abduction and dorsal and radial force. Mean Muscle Activation Mean percentage of muscle activation of all participants in each muscle force test is shown in eAppendix 2 (available at ptjournal.apta.org). On average, the maximum activation levels were: 83% for the infraspinatus muscle during prone lateral rotation (muscle force test 13), 74% for the supraspinatus muscle during the empty can test (muscle force test 1), 64% for the teres minor muscle during prone lateral rotation (muscle force test 13), and 58% for the subscapularis muscle during neutral arm posture with ulnar resistance (muscle force test 28). Fatigue Significant decreases in both MnPF and MdPF existed only in the infraspinatus muscle (P⫽.016 and P⫽.022, respectively), with mean percent decreases of 8.3% and 11.4%, respectively. Significant decreases in MdPF occurred for the biceps brachii muscle (P⫽.026), which decreased an average of 5.3%.
Discussion The purpose of this study was to identify muscle force tests that relaApril 2010
Construct Validity of Muscle Force Tests tively isolated the rotator cuff muscles. The supraspinatus, infraspinatus, and teres minor muscles were maximally relatively isolated (produced maximal IRs) within their respective functional clinical test groups (lateral rotation for the infraspinatus and teres minor muscles and abduction for the supraspinatus muscle), although other generic muscle force tests were found to be equally effective in isolating these muscles. The subscapularis muscle was relatively isolated within its respective clinical medial rotation group; however, this level of isolation was second to that of the ulnar force group, although not statistically different. These findings establish traditional MMTs as effective for the relative maximal isolation of individual rotator cuff muscles. Considerable population variability (as indicated by large data standard deviations) suggests that specific muscle force tests may not relatively isolate rotator cuff muscles for all individuals. Therefore, we recommend that the rotator cuff muscles be tested in a number of isolation muscle force tests to verify isolation and promote accurate muscle assessment. Isolation and Activation of the Infraspinatus Muscle The highest mean IR for the infraspinatus muscle occurred within the lateral rotation group (X⫽2.29, SD⫽0.53) and was significantly higher than mean IRs in all other groups. Therefore, the infraspinatus muscle was most isolated (activated 2.29 times more than the other 13 recorded muscles) within the clinical lateral rotation muscle force tests. These findings support the hypothesis that clinical tests in the lateral rotation group (prone and sitting lateral rotation) are appropriate in isolating and assessing the strength and function of the infraspinatus muscle. The findings matched expectations, as the infraspinatus April 2010
Figure 3. Mean isolation of the teres minor muscle using the isolation ratio (IR) among muscle force test groups. The clinical muscle force tests are within the medial rotation, lateral rotation, and abduction muscle force test groups (white bars). The generic muscle force tests are within the palmar, dorsal, radial, and ulnar force groups (blue bars). The results of the Tukey honestly significant difference test are shown; levels not connected by the same letter are significantly different. The error bars represent ⫺1 standard deviation. The highest mean (SD) IR (maximal isolation) was found in the lateral rotation group (1.76⫾0.94); the lowest mean IR was found in the abduction group (0.84⫾0.49).
Figure 4. Mean isolation of the subscapularis muscle using the isolation ratio (IR) among muscle force test groups. The clinical muscle force tests are within the medial rotation, lateral rotation, and abduction muscle force test groups (white bars). The generic muscle force tests are within the palmar, dorsal, radial, and ulnar force groups (blue bars). The results of the Tukey honestly significant difference test are shown; levels not connected by the same letter are significantly different. The error bars represent ⫺1 standard deviation. The highest mean (SD) IR (maximal isolation) was found in the ulnar force group (1.95⫾1.20); the lowest mean IR was found in the abduction group (0.39⫾0.38).
Volume 90
Number 4
Physical Therapy f
577
Construct Validity of Muscle Force Tests muscle is primarily a lateral rotator and thus should be significantly activated in lateral rotation muscle force tests. Previous findings also showed the infraspinatus muscle to be most isolated from synergists (supraspinatus and posterior deltoid muscles) during sitting lateral rotation.8 The infraspinatus muscle has been found to be significantly activated during side-lying lateral rotation (peak of 85% MVC),3 and previous literature10 also has shown the infraspinatus and teres minor muscles to be equally activated during prone and side-lying lateral rotation. In the present study, the maximal activation of the infraspinatus muscle was, on average, 83% and occurred during prone lateral rotation (muscle force test 13). These findings suggest that the infraspinatus muscle may be isolated in some positions during which it is maximally activated, which complies with previous methods of determining isolation muscle force tests.3,10,16 However, the sitting lateral rotation muscle force test, which was found to be equally effective in isolating the infraspinatus muscle, produced lower activations (average of 58% MVC), demonstrating that muscle force tests of maximal activation are not always indicative of maximum isolation. Isolation and Activation of the Supraspinatus Muscle The supraspinatus muscle was most isolated from the other 13 recorded muscles in the abduction, lateral rotation, radial force, and dorsal force groups. These results indicate that the clinical abduction muscle force tests (empty can, full can, Blackburn, and supraspinatus muscle neutral abduction muscle force tests) and muscle force tests within the radial force group (abduction and flexion), dorsal force group (abduction and horizontal abduction), and lateral rotation group (prone lateral rotation in an abducted humeral posture and sit578
f
Physical Therapy
Volume 90
ting lateral rotation) are effective in isolating the supraspinatus muscle from surrounding muscles and are appropriate to use in assessing the strength and function of the supraspinatus muscle. Because the supraspinatus muscle aids the deltoid muscle in shoulder abduction, it is not surprising that the supraspinatus muscle is isolated in muscle force tests of abduction. It was surprising to see the supraspinatus muscle isolated in muscle force tests of flexion and lateral rotation (within the radial force and lateral rotation groups, respectively). This isolation could be a result of grouping muscle force tests containing combinations of abduction and flexion or abduction and lateral rotation postures within the same group. Perhaps if muscle force tests contained only one primary posture and were grouped separately, a clearer distinction would be made in isolating the supraspinatus muscle in abduction muscle force tests. On average, the maximal activation of the supraspinatus muscle was 74%, which occurred during the empty can test (muscle force test 1). Other muscle force tests identified to equally isolate the supraspinatus muscle produced only submaximal activations (average of ⱖ29% MVC). Previous literature has recommended empty can and full can tests be used to assess the integrity of the supraspinatus muscle.8,17,18 Empty can and full can tests were found to be equivalent in accuracy for detecting supraspinatus muscle tears.18 These tests are very similar to the abduction (90°) muscle force test within the radial force group. Kelly et al8 considered the infraspinatus muscle as a synergist of the supraspinatus muscle and compared supraspinatus muscle isolation during 6 muscle force tests (including the empty can and full can tests) and concluded that the full can test best isolated the supraspinatus muscle.
Number 4
Results from the present study showed that there was no significant difference between IRs of the empty can and full can tests, indicating the comparable ability of both muscle force tests to isolate the supraspinatus muscle. Isolation and Activation of the Teres Minor Muscle The teres minor muscle was most isolated from the other 13 recorded muscles within the lateral rotation, medial rotation, palmar force, dorsal force, and ulnar force groups. These results confirm that clinical tests (prone and sitting lateral rotation) are appropriate in assessing the teres minor muscle. However, other muscle force tests within the medial rotation, ulnar force, palmar force, and dorsal force groups (medial rotation test; shoulder extension, adduction, and abduction tests; and horizontal abduction and adduction tests) are equally effective in isolating the teres minor muscle. The teres minor muscle is primarily a lateral rotator, so it was expected that it would be isolated in muscle force tests of lateral rotation. The teres minor muscle originates on the infraspinous fossa of the scapula and inserts on the greater tubercle of the humerus, so it also was not surprising for the teres minor muscle to be isolated in muscle force tests of extension and adduction. It was surprising, however, to find that the teres minor muscle also was isolated in muscle force tests of medial rotation and abduction, during which the muscle would not be expected to be very active. It is possible that the teres minor muscle is activated in these muscle force tests of medial rotation and abduction to help stabilize the glenohumeral joint and prevent anterior dislocation. The teres minor muscle was maximally activated an average of 64%, and this occurred during prone lateral rotation (muscle force test 13). April 2010
Construct Validity of Muscle Force Tests Previous literature10 showed the teres minor muscle to be significantly activated in prone and sitting lateral rotation muscle force tests similar to those of the present study. However, the teres minor muscle was not always maximally activated (average of ⱖ24% MVC) in other isolation muscle force tests. Isolation of the Subscapularis Muscle The subscapularis muscle was most isolated (1.95 times more activated) from the other 13 recorded muscles within the muscle force tests of the ulnar force group. However, these levels of isolation were not significantly different from those displayed during muscle force tests within the clinical medial rotation group (1.60). These results confirm that clinical tests (prone medial rotation, lift-off, and belly press muscle force tests) are appropriate in assessing the subscapularis muscle. However, other muscle force tests from the ulnar force group (shoulder extension and adduction) are equally effective in isolating the subscapularis muscle and can be used in the assessment of the strength and function of the subscapularis muscle. The present findings indicate no difference in isolation of the subscapularis muscle between the lift-off and belly press muscle force tests, and both muscle force tests have been shown to maximally activate the subscapularis muscle.7 However, some authors have concluded that the lift-off muscle force test was superior compared with the belly press muscle force test, as the subscapularis muscle was more isolated from the pectoralis major muscle4 and from the pectoralis major and latissimus dorsi muscles8 during the lift-off muscle force test. The highest activation of the subscapularis muscle was, on average, 58% and occurred during the neutral arm posture with ulnar resistance (muscle force test 28). Although maximal activation occurred April 2010
within an identified isolation muscle force test, other defined isolation muscle force tests produced submaximal activations (ⱖ34% MVC) of the subscapularis muscle. Assessment of Fatigue Although significant decreases in the frequency content of the EMG signal existed for the infraspinatus and biceps muscles, there was no indication (frequency decrease) that any of the other 12 muscles were fatigued. Recommended rest levels15 were provided, and the number of maximal muscle force tests in the present study was the same as those done in previous studies.8,11 We conclude that fatigue had minimal, if any, impact on the results. Limitations If channels were excluded from calculations due to motion artifact, or if motion artifact was missed and these data were included in the analysis, inaccuracy would enter the IR calculations, and thus the identification of isolation muscle force tests using that criterion may be incorrect. In addition, no data were collected to support the reliability of these IRs. The grouping of muscle force tests based on specific muscle force test criteria may have inflated or deflated group mean IRs, which could have influenced which groups most isolated the muscle of interest. Due to the instrumentation used in this study, the methods do not exactly mimic clinical application of the tests. There is limited generalizability of these results due to the small sample size (of participants who were asymptomatic).
Conclusion This study demonstrated that attempts to isolate the rotator cuff muscles fully are problematic. However, it also revealed that muscle force tests exist that relatively isolate the rotator cuff muscles and can detect large changes in rotator cuff
muscle activation. The rotator cuff muscles were maximally relatively isolated within the following groups: lateral rotation group (infraspinatus muscle); abduction, lateral rotation, and radial and dorsal force groups (supraspinatus muscle); lateral rotation, medial rotation, and palmar, dorsal, and ulnar force groups (teres minor muscle); and medial rotation and ulnar force groups (subscapularis muscle). We confirmed both study hypotheses. First, the clinical tests were generally most effective in achieving maximum isolation of the rotator cuff muscles, although other muscle force tests also were equally effective. Second, although rotator cuff muscles were relatively isolated within muscle force tests based on their primary action, in some instances, alternative muscle force tests (and actions) similarly isolated some muscles. The results confirmed the appropriateness of currently applied clinical tests in assessing individual rotator cuff muscle status and identified additional muscle force tests that similarly relatively isolated these muscles. These findings may provide clinicians with more confidence, qualitative accuracy, and flexibility in their assessment of musclespecific strength. Future studies of rotator cuff isolation should continue to investigate the contributions of these surrounding muscles during these and other muscle force tests and should include a larger population sample. All authors provided concept/idea/research design, writing, data collection and analysis, and consultation (including review of manuscript before submission). Ms Brookham and Dr Dickerson provided project management. Dr Dickerson provided fund procurement and facilities/equipment. Ms Brookham provided participants. Dr McLean and Dr Dickerson provided institutional liaisons.
Volume 90
Number 4
Physical Therapy f
579
Construct Validity of Muscle Force Tests Study approval was provided by the University of Waterloo Office of Research Ethics. This study was supported by funds from the Department of Kinesiology, University of Waterloo. This article was received January 23, 2009, and was accepted October 30, 2009. DOI: 10.2522/ptj.20090024
References 1 Daniels L, Worthingham C. Muscle Testing Techniques of Manual Examination. Philadelphia, PA: WB Saunders Co; 1986. 2 Moore L, Dalley AF. Clinically Oriented Anatomy. 4th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 1999. 3 Townsend H, Jobe FW, Pink M, Perry J. Electromyographic analysis of the glenohumeral muscles during a baseball rehabilitation program. Am J Sports Med, 1991; 19:264 –272. 4 Greis PE, Kuhn JE, Schultheis J, et al. Validation of the lift-off test and analysis of subscapularis activity during maximal internal rotation. Am J Sports Med. 1996;24: 589 –593. 5 Decker M, Tokish J, Ellis H, et al. Subscapularis muscle activity during selected rehabilitation exercises. Am J Sports Med. 2003;31:126 –134. 6 Suenaga N, Minami A, Fujisawa H. Electromyographic analysis of internal rotation motion of the shoulder in various arm positions. J Shoulder Elbow Surg. 2003;12: 501–505.
580
f
Physical Therapy
Volume 90
7 Tokish J, Decker M, Ellis H, et al. The bellypress test for the physical examination of the subscapularis muscle: electromyographic validation and comparison to the lift-off test. J Shoulder Elbow Surg. 2003; 12:427– 430. 8 Kelly BT, Kadrmas WR, Speer KP. The manual muscle examination for rotator cuff strength; an electromyographic investigation. Am J Sports Med. 1996;24:581– 588. 9 Liu J, Hughes RE, Smutz WP, et al. Roles of deltoid and rotator cuff muscles in shoulder elevation. Clin Biomech. 1997;12:32– 38. 10 Ballantyne BT, O’Hare SJ, Paschall JL, et al. Electromyographic activity of selected shoulder muscles in commonly used therapeutic exercises. Phys Ther. 1993;73: 668 – 682. 11 Jenp Y, Malanga GA, Growney ES, An K. Activation of the rotator cuff in generating isometric shoulder rotation torque. Am J Sports Med. 1996;24:477– 485. 12 Delagi E, Perotto A. Anatomic Guide for the Electromyographer. 2nd ed. Springfield, IL: Charles C Thomas, Publisher; 1980. 13 Nemeth G, Kronberg M, Brostrom L. EMG recordings from the subscapularis muscle: description of a technique. J Orthop Res. 1990;8:151–153. 14 Cram JR, Kasman GS. Introduction to Surface Electromyography. Gaithersburg, MD: Aspen Publishers Inc; 1998. 15 De Luca CJ. The use of surface electromyography in biomechanics. J Appl Biomech. 1997;13:135–163.
Number 4
16 Dark A, Ginn KA, Halaki M. Shoulder muscle recruitment patterns during commonly used rotator cuff exercises: an electromyography study. Phys Ther. 2007;87:1039 – 1046. 17 Jobe FW, Moynes DR. Delineation of diagnostic criteria and a rehabilitation program for rotator cuff injuries. Am J Sports Med. 1982;10:336 –339. 18 Itoi E, Tadato K, Sano A, et al. Which is more useful, the “full can test” or the “empty can test,” in detecting the torn supraspinatus tendon? Am J Sports Med. 1999;27:65– 68. 19 Malanga GA, Jenp Y-N, Growney ES, KaiNan A. EMG analysis of shoulder positioning in testing and strengthening the supraspinatus. Med Sci Sports Exerc. 1996; 28:661– 664. 20 Blackburn TA, McLeod WD, White B, Wofford L. EMG analysis of posterior rotator cuff exercise. Athl Train. 1990;25:40 – 45. 21 Clarkson MH, Gilewich GB. Musculoskeletal Assessment: Joint Range of Motion and Manual Muscle Strength. Baltimore, MD: Lippincott Williams & Wilkins; 1989. 22 Janda V. Muscle Function Testing. London, United Kingdom: Butterworths; 1983. 23 Gerber C, Krushell RJ. Isolated rupture of the tendon of the subscapularis muscle. J Bone Joint Surg Br. 1991;73:389 –394. 24 Gerber C, Hersche O, Farron A. Isolated rupture of the subscapularis tendon: results of operative repair. J Bone Joint Surg Am. 1996;78:1015–1023.
April 2010
Research Report Motor and Sensory Nerve Conduction Are Affected Differently by Ice Pack, Ice Massage, and Cold Water Immersion Esperanza Herrera, Maria C. Sandoval, Diana M. Camargo, Tania F. Salvini
Background. It is well known that reducing tissue temperature changes sensory and motor nerve conduction. However, few studies have compared the effect of different cold modalities on nerve conduction parameters. Objective. The purpose of this study was to compare the effects of ice pack, ice massage, and cold water immersion on the conduction parameters of the sural (sensorial) and tibial motor nerves.
Design. An experimental study was conducted in which the participants were randomly assigned to 1 of 3 intervention groups (n⫽12 per group). Independent variables were cold modality and pre- and post-cooling measurement time. Dependent variables were skin temperature and nerve conduction parameters.
Methods. Thirty-six people who were healthy, with a mean (SD) age of 20.5 (1.9) years, participated in the study. Each group received 1 of the 3 cold modalities, applied to the right calf region for 15 minutes. Skin temperature and nerve conduction parameters were measured before and immediately after cooling.
Results. All 3 modalities reduced skin temperature (mean⫽18.2°C). There also was a reduction in amplitude and an increase in latency and duration of the compound action potential. Ice massage, ice pack, and cold water immersion reduced sensory nerve conduction velocity (NCV) by 20.4, 16.7, and 22.6 m/s and motor NCV by 2.5, 2.1, and 8.3 m/s, respectively. Cold water immersion was the most effective modality in changing nerve conduction parameters. Limitations. The cooling area of the ice massage and ice pack was smaller than that of the cold water immersion. The examiner was not blinded to the treatment group. The population included only participants who were healthy and young.
E. Herrera, PT, MS, is a PhD student in the Program of Physiological Sciences, Federal University of Sa˜o Carlos, Sa˜o Carlos, Brazil, and Titular Professor, Department of Physical Therapy, Universidad Industrial de Santander, Ciudad Universitaria, Carrera 27 Calle 9, Bucaramanga, Santander, Colombia. Address all correspondence to Ms Herrera at:
[email protected]. M.C. Sandoval, PT, PhD, is Associate Professor, Department of Physical Therapy, Universidad Industrial de Santander. D.M. Camargo, MS, is Associate Professor, Department of Physical Therapy, Universidad Industrial de Santander. T.F. Salvini, PT, PhD, is Titular Professor, Department of Physical Therapy, Federal University of Sa˜o Carlos. [Herrera E, Sandoval MC, Camargo DM, Salvini TF. Motor and sensory nerve conduction are affected differently by ice pack, ice massage, and cold water immersion. Phys Ther. 2010;90:581–591.] © 2010 American Physical Therapy Association
Conclusions. All 3 modalities were effective in reducing skin temperature and changing sensory conduction at a physiological level that is sufficient to induce a hypoalgesic effect. The results suggest that cold water immersion, as applied in this study, is the most indicated modality for inducing therapeutic effects associated with the reduction of motor nerve conduction.
Post a Rapid Response or find The Bottom Line: www.ptjournal.org April 2010
Volume 90
Number 4
Physical Therapy f
581
Cold Modalities and Nerve Conduction
C
ryotherapy is the therapeutic application of a substance to remove body heat, resulting in diminished tissue temperature.1,2 It often is used in sports and rehabilitation settings during the immediate and rehabilitative phases of injury management.3 Reduced tissue temperature, blood flow, and cellular metabolism are some of the physiological effects of cryotherapy.2– 8 Cryotherapy also reduces nerve conduction velocity (NCV) in the sensory and motor nerves9,10 and has a controversial effect on muscle strength (force-generating capacity).11–13 These physiological changes lead to some therapeutic effects such as a reduction in pain and muscle spasm and the prevention of posttraumatic edema.1–13 Various modalities are frequently used to deliver cryotherapy treatment. The efficacy of cooling depends on the method, application time, and treatment area and the individual’s physical activity level immediately before or after the intervention.14 Overall, crushed ice pack, ice massage, and cold water immersion are considered the most effective clinical modalities for reducing tissue temperature.14 –17 The efficacy of the cryotherapy modalities has been assessed by comparing their capacity to decrease intramuscular,18 intra-articular,19 and skin temperature10,14 –17,20,21 and to maintain the temperature changes. Skin temperature measurement has been widely used because it is a simple and noninvasive procedure. Some authors,
Available With This Article at ptjournal.apta.org • Audio Abstracts Podcast This article was published ahead of print on February 25, 2010, at ptjournal.apta.org.
582
f
Physical Therapy
Volume 90
based on skin temperature measurements, have hypothesized that skin temperature changes are closely related to changes in subcutaneous and intramuscular temperature.10,15–17 However, the study by Jutte et al,22 which used a multiple regression model, showed that skin temperature was a weak predictor of intramuscular temperature because it explained only 21% of temperature variance within the muscle. The influence of subcutaneous and muscular tissue thickness on the cooling of deeper tissues also has been debated.23,24 A more precise way of analyzing the efficacy of cryotherapy modalities would be to compare their effects on deep tissues directly associated with clinical intervention using quantitative, direct, and reliable measurement. For example, nerve fibers are targeted for cryotherapy intervention to reduce muscle pain and spasm,3 and the changes attributed to cooling can be identified through nerve conduction studies (NCS) in which reliability has been established previously.25 Prior electrophysiological studies have determined a direct linear relationship between skin temperature and NCV and an inverse relationship with latency, amplitude, and duration of compound action potential.26 –30 Nevertheless, this relationship varies according to the type of nerve fiber. Sensory nerves can show a reduction of 1.4 to 2.6 m/s for every degree of skin temperature reduction, whereas motor NCV can decrease by 1.1 to 1.5 m/s/°C.1 There are other factors that affect the relationship between skin temperature and NCV, such as the depth of the nerve, the amount of surrounding subcutaneous tissue, age, range of temperature variation,27–30 and possibly the type of modality used to alter skin temperature.
Number 4
In the literature, there is a lack of studies comparing the effects of the different cold modalities on motor and sensory nerve conduction parameters. We found only one study31 that established a greater effect of cold packs compared with gel packs on reducing ulnar motor NCV. However, this study did not analyze the effect of these modalities on sensory nerve conduction. Therefore, it is important to compare the effectiveness of the different cryotherapy modalities on motor and sensory nerve conduction to provide physiological parameters that contribute to the indication of the most adequate modality according to the desired therapeutic effect. Considering that each cold modality has a different capacity to cool the skin and subcutaneous tissues and that nerve fiber conduction is affected by skin temperature changes, the hypothesis of this study was that cryotherapy protocols with different characteristics should have different effects on sensory and motor nerve conduction. The purpose of this study was to compare the effects of 3 commonly used therapeutic cold modalities (ice pack, ice massage, and cold water immersion) on the conduction parameters of the sural nerve and tibial motor nerve in participants who were healthy.
Method Research Design An experimental study was conducted with 3 randomly assigned intervention groups. The independent variables were cold modality type (ice pack, ice massage, and water immersion) and measurement time (pre- and post-cooling). The dependent variables were skin temperature (degrees Celsius) and nerve conduction parameters: NCV (meters per second), latency and duration (milliseconds), amplitude of compound muscle (millivolts), and sensory action potentials (microvolts). April 2010
Cold Modalities and Nerve Conduction Participants The participants were informed of the experimental procedures and the risks involved with the study and signed a consent form. Thirty-six participants who were healthy (18 women and 18 men) were enrolled in this study. The participants’ mean (SD) age, mass, height, and body mass index were 20.5 (1.9) years, 60.2 (8.4) kg, 1.63 (0.1) m, and 22.4 (1.6) kg/m2, respectively. The sample size for each cold modality group was determined through the application of the sampsi command of Stata 9.0 software.* The following design specifications were taken into account: ␣⫽.05; (1⫺ )⫽0.9; ratio⫽1:1; and method of calculation⫽analysis of covariance (ANCOVA) for repeated measurements, with a baseline measurement and a final measurement. The correlation between the initial and final measurements was r⫽.2. This method defined a sample of 10 to 12 participants for each cold modality group. All participants filled out a health questionnaire that indicated the presence of any of the following exclusion criteria: history of alcoholism or smoking, peripheral vascular or cardiovascular disease, diabetes, neurological or skeletal muscle disorders, recent trauma or injury to the right leg, local hot or cold insensitivity, cold adverse reactions, Raynaud phenomenon, and pregnancy. Additionally, the participants were asked to avoid eating and drinking any stimulants (eg, alcohol, caffeine, chocolate) 2 hours before the intervention and to not exercise for at least 4 hours before intervention. These exclusion criteria and recommendations were considered according to previous studies.10,31
Instruments Skin temperature was measured using an infrared thermometer (Raytek ST PRO†) that displays a precision of 1°C, high reliability (intraclass correlation coefficient⫽.97), validity (r⫽ .92), and responsiveness (change index⫽4.2). Nerve conduction measurements were acquired using a Nicolet Compass Meridian System‡ and standard surface electrodes from the same manufacturer. The selection of cold modalities was based on their high effectiveness in reducing skin temperature and their frequent application in the clinical setting.14 –17 The ice pack consisted of 279 g of crushed ice in a plastic bag of 18 ⫻ 8 cm without air. Ice massage was applied by using an ice block of 279 g with dimensions of 8 ⫻ 10 ⫻ 5 cm. Water immersion was conducted in an acrylic container of 20 ⫻ 35 ⫻ 30 cm, filled with water and crushed ice until the water temperature reached approximately 10°C, as reported previously.17,20 The temperature of this modality was measured throughout the intervention, showing an initial mean of 8.9 (1.0)°C and a final mean of 7.8 (1.2)°C. Procedure The participants were randomly assigned to 1 of 3 cold modality groups by using a computer-generated random number sequence.32 Furthermore, to minimize the influence of the circadian cycle on body temperature regulation, all participants received the cold modality at the same time (eg, 2– 6 PM). The intervention and measurement procedures were performed on the right calf of each participant. Given that the postcooling measurement had to be taken immediately after the cold modality application, the same room was used for the application of inter†
* StataCorp LP, 4905 Lakeway Dr, College Station, TX 77845.
April 2010
Raytek Corp, 1201 Shaffer Rd, Santa Cruz, CA 95061. ‡ Nicolet Biomedical Co, 5225 Verona Rd #2, Fitchburg, WI 53711-4497.
vention and for the measurement procedures. Room temperature was maintained at 24 (0.08)°C, and there were no significant variations during the tests (P⫽.29). Before the experimental protocol, the participants were asked whether they had followed the recommendations regarding stimulant intake and exercise. Their height and weight were recorded to calculate the body mass index. The participants wore T-shirts and shorts and, for acclimatization, assumed the prone position on the standard examining table for 15 minutes. During the acclimatization time, the treatment area to be cooled was determined and the electrodes for NCS were placed. Cold modalities. The cold modalities were applied for 15 consecutive minutes by the same trained physical therapist (M.C.S.). This duration is frequently used for treatments because it is sufficient to achieve therapeutic effects and it avoids complications from cold modalities.21,33 The ice massage and the ice pack were applied to a previously determined rectangular area (18 ⫻ 8 cm) on the calf (Fig. 1). The ice pack was applied directly to the skin and without compression. The ice massage was applied by continuous longitudinal displacements. For the cold water immersion, the participants remained seated while immersing the right leg as far as the top border of the rectangle determined for the previous modalities (Fig. 1). At the end of intervention, the leg was quickly dried without friction, and the participant returned to the prone position for the post-cooling measurement. All participants completed the experimental protocols without adverse reactions to the cold. Skin temperature measurement. Skin temperature was measured immediately before (pre-cooling) and after (post-cooling) the cold modal-
Volume 90
Number 4
Physical Therapy f
583
Cold Modalities and Nerve Conduction pre-cooling measurement, and the recording electrodes were not removed during the intervention, except in the participants who received cold water immersion. In this case, the recording electrodes were removed after the pre-cooling measurement and replaced at the sites previously marked for the postcooling measurement.
Figure 1. Cooling area. Ice pack and ice massage were applied to the same rectangular area defined according to the following procedure: (a) measurement of the length of the leg between the head of the fibula (1) and the lateral malleolus (2); (b) definition of the midpoint between the head of the fibula and the lateral malleolus (3); (c) projection of a perpendicular line to the posterior part of the leg, marking the midpoint of the calf (4); and (d) placement of the center of an acetate mold in the midpoint of the calf to mark a rectangle (18 ⫻ 8 cm) where the ice pack and ice massage would be applied. For the cold water immersion, the participants immersed their right leg in a cold water tank as far as the top border of the rectangle (5).
Figure 2. Stimulation and recording electrode sites for the sural nerve and tibial motor nerve conduction studies. (A) Sural nerve: antidromic technique was performed. (B) Tibial motor nerve: distal stimulation on the medial malleolus and proximal stimulation on the medial aspect of the knee crease (not shown). S⫽stimulation site, R⫽recording site, and G⫽ground electrode.
ity application. The temperature was measured at the center of the previously defined rectangle (Fig. 1) with an infrared thermometer placed in a perpendicular position and kept as close as possible to the skin without touching. Nerve conduction measurement. Compound action potentials resulting from stimulation of the posterior tibial motor and sural nerves were recorded twice, before and after cooling, according to standardized techniques described by Oh.34 These 584
f
Physical Therapy
Volume 90
nerves were selected because they are located within the treatment area. Furthermore, the posterior tibial nerve has a high quantity of motor fibers, and the sural nerve is a pure sensory nerve,26,34 allowing the assessment of the cooling effects in both motor and sensory fibers. Nerve conduction studies were obtained by the same examiner (E.H.). In order to reduce technical variations, the stimulation and recording sites were delimited with a permanent ink marker during the
Number 4
Before the NCS measurement, the participants were instructed to avoid leg movements. The sural nerve recordings were obtained with a bandwidth of 20 Hz to 3 kHz, a gain of 20 V per division, and a sweep speed of 1 millisecond per division. The surface bar recording electrode was placed immediately behind the lateral malleolus and the stimulating electrode, was placed about 14 cm proximal to the active recording electrode, just lateral to the midline of the width of the calf muscle34 (Fig. 2A). Stimuli were 100microsecond rectangular pulses, with amplitude adjusted slightly higher than needed to ensure a maximum response. The nerve signals were obtained by averaging 20 responses. The following sensory nerve parameters were measured: NCV, peak latency, peak-to-peak amplitude, and duration (onset to end of negative wave) of the compound sensory action potential. The tibial motor nerve recordings were obtained with a bandwidth of 2 Hz to 10 kHz, a gain of 2 mV per division, and a sweep speed of 2 milliseconds per division. The active disc recording electrode was placed over the abductor hallucis muscle, and the reference disc recording electrode was placed at the base of the big toe. The ground electrode was positioned on the calf muscle. The distal stimulation site was on the ankle immediately behind the medial malleolus, and the proximal stimulation site was on the knee on the medial aspect of the knee crease34 April 2010
Cold Modalities and Nerve Conduction (Fig. 2B). The following motor nerve parameters were measured: NCV for the nerve segment between ankle and knee, distal latency, amplitude, and duration of the negative wave of the compound muscle action potential. Intrarater reliability of sural and tibial motor NCS. Before data collection, we assessed intrarater reliability of the tibial motor and sural nerve recordings in 20 participants following the same recording techniques described above. The same examiner who performed the assessments of the current study tested each participant twice on 2 separate days with a minimum 8-day lapse between the measurements.25 Statistical Procedures Intrarater reliability of nerve conduction parameters was evaluated using the Bland-Altman method.35 Data were reported as mean difference (95% limits of agreement). For the present study, descriptive statistics were used to summarize the characteristics of the population, the skin temperature, and nerve conduction data, which are presented as mean (SD). The baseline characteristics of the cold modality group participants were compared using analysis of variance (ANOVA) and a chi-square test, depending on the scale of measurement of each variable.32 The measurements obtained before and after cooling were compared using a paired t test because the normal distribution of all variables was proven by the Shapiro-Wilk test.32,36 In addition, an ANCOVA37 compared the effects of the 3 modalities of skin temperature and the nerve conduction parameters using the ice massage group as reference. For the statistical analysis, the Stata 9.0 software was used, with a significance level of ␣⫽.05.
April 2010
Results Intrarater Reliability of NCS The intrarater analysis showed mean differences close to zero, and there was no evidence of systematic error. The mean differences (95% limits of agreement) for the sural nerve parameters were: latency⫽0.17 millisecond (⫺0.73, 1.07), NCV⫽⫺0.07 m/s (⫺4.48, 4.33), amplitude⫽⫺2.9 V (⫺20.73, 14.95), and duration⫽0.04 millisecond (⫺0.3, 0.37).25 Respective data for the tibial motor nerve parameters were: latency⫽0.23 millisecond (⫺1.10, 1.56), NCV⫽⫺0.32 m/s (⫺6.20, 5.53), amplitude⫽⫺0.1 mV (⫺4.30, 4.10), and duration⫽0.36 millisecond (⫺0.91, 1.63) (unpublished data). Effects of Cold Modalities on Skin Temperature and Nerve Conduction Parameters A total of 39 potential participants were assessed for eligibility; 2 did not meet inclusion criteria, and 1 was not assisted to the experimental session. Twelve participants were randomly allocated to each experimental group (Fig. 3). There were no significant differences in baseline characteristics among the cold modality group participants (P⬎.05) (Tab. 1). There was a decrease in skin temperature after the application of the 3 modalities (P⬍.0001) (Tab. 2). The ice massage caused a greater decrease in skin temperature compared with the ice pack (⫽3.03, P⫽.001) and the cold water immersion (⫽9.36, P⬍.0001). All 3 modalities induced an increase in latency and duration of the compound action potential of the sural and tibial motor nerves (P⬍.05). There also was a reduction in the amplitude of the potentials and the NCV (P⬍.05) (Tabs. 3 and 4). The effect of the cold water immersion on all motor nerve parameters, as well as on amplitude and duration of sural nerve potential, was different and greater compared with the effect of ice massage (Tab. 5). There
were no differences between the effects of the ice pack and ice massage on the motor and sensory conduction parameters (P⬎.05) (Tab. 5).
Discussion The 3 cold modalities resulted in significant changes in every sural nerve parameter, except cold water immersion in amplitude (Tab. 4). Mean differences among parameters determined before and after cooling were greater than those determined in the intrarater reliability analysis. The effects of ice massage and ice pack on the tibial motor nerve parameters were more subtle (Tab. 3). Although latency and duration differences were statistically significant for the effect of ice pack intervention on the tibial motor nerve, mean differences were lower or similar to those determined in the intrarater reliability analysis for this nerve. However, it is important to note that assessments after ice pack and ice massage protocols did not require the removal of electrodes, which usually is the main source of error in NCS. Mean differences in tibial motor nerve parameters from cold water immersion were greater than those determined in the intrarater reliability analysis. Therefore, we believe that the motor and sensory nerve conduction changes determined for each modality were a real consequence of cooling rather than error in measurement methods. The results of this study support the proposed hypothesis because the cold modalities applied have different effects on motor and sensory nerve conduction. The modality of cold water immersion, as applied in this study, had the greatest effect on the conduction parameters, especially of the tibial motor nerve (Tab. 5). The modalities of ice pack and ice massage, as applied in this study, differed substantially from the cold water immersion. First, the ice massage and ice pack were applied
Volume 90
Number 4
Physical Therapy f
585
Cold Modalities and Nerve Conduction
Figure 3. Flow of participants through the study.
to the same calf area (44 cm2), whereas the area/volume covered by cold water immersion was much greater, including the calf, ankle, and foot regions where the nerve becomes more superficial and thus
more susceptible to cooling. Second, these modalities also have thermodynamic differences: in the ice massage and ice pack modalities, heat exchange occurs by conduction, whereas cold water immersion in-
volves conduction and convection processes.1 Our results can be explained mainly by the differences in the area/volume, and this parameter of the cold modalities may contribute to a greater cooling effect on the
Table 1. Demographic Characteristics of the Participantsa Intervention Group Ice Massage (nⴝ12)
Ice Pack (nⴝ12)
Cold Water Immersion (nⴝ12)
P
19.7 (1.3)
20.7 (1.3)
20.9 (2.6)
.26
Female participants, n (%)
5 (41.7)
6 (50)
7 (58.3)
.72
Height (m)
1.61 (0.1)
1.64 (0.1)
1.65 (0.1)
.54
Variable Age (y)
a
Mass (kg)
58 (7.1)
60.4 (8.6)
62.1 (9.7)
.51
Body mass index (kg/m2)
22.2 (1.6)
22.3 (1.4)
22.6 (1.7)
.81
Data are presented as mean (SD), except for the number and percentage of female participants.
586
f
Physical Therapy
Volume 90
Number 4
April 2010
Cold Modalities and Nerve Conduction Table 2. Skin Temperature in Participants Submitted to Different Cold Modalitiesa Skin Temperature (°C)
a b c
Intervention Group
Pre-Cooling
Post-Cooling
Differenceb
Ice massage
31.58 (1.07)
3.98 (1.15)
⫺27.6 (1.32)c
Ice pack
31.12 (2.13)
6.68 (3.4)
⫺24.43 (2.87)c
Cold water immersion
31.55 (0.89)
13.32 (1.33)
⫺18.23 (1.46)c
Data are presented as mean (SD). Difference⫽post-cooling ⫺ pre-cooling. P⬍.0001.
subcutaneous tissues, including the peripheral nerve. Future studies are needed to compare the effects of the cold modalities on nerve conduction with different thermodynamic properties applied to an area of similar magnitude. Paradoxically, cold water immersion was the modality that caused the least skin temperature reduction (Tab. 2), possibly due to the fact that a greater area received the treatment, leading to a faster activation of the thermoregulatory responses that protect the body from abrupt temperature changes.38 Consequently, skin temperature was quickly stabilized and did not adequately reflect the effects of cooling on subcutaneous tissues.22 The cooling induced by the 3 modalities was effective in reducing the NCV and prolonging the latency and duration of the compound muscle and sensory action potentials (Tabs. 3 and 4). The effects of temperature reduction on nerve conduction parameters are well described in the literature27–30 and may result from the changes in the structure of the axonal membrane39 and from the conductance of the voltage-sensitive sodium and potassium channels.27 Therefore, the cold reduces the nerve membrane current, which lengthens the refractory periods following a stimulus; as a result, the duration of the nerve action potential increases and the rate of impulse transmission decreases. April 2010
In the scientific literature, the relationship between the amplitude of compound action potential and temperature remains a controversial issue. Some studies that analyzed the effect of temperature changes on conduction parameters showed a negative relationship,40,41 whereas other authors did not identify this relationship.27 In the present study, cold water immersion significantly reduced the amplitude of compound muscle action potential (Tab. 3). Similarly, the ice massage and the ice pack reduced the amplitude of sensory compound action potential (Tab. 4). Perhaps the differences between the results of the present study and those of previous studies27,40,41 are due to the differences in the skin temperature changes. In previous studies,27,40,41 skin temperature decreased only from 33.6°C to 22.5°C, whereas in the present study, the cooling induced by all modalities was greater (from 31.6°C to 4°C). The amplitude of the compound action potential represents the number of nerve fibers that responds to an appropriate electrical stimulus.34 Therefore, the reduction of this parameter after the cold modality application could suggest an increase in the activation threshold of some nerve fibers, as well as a block of the fibers that are more sensitive to cooling. Additionally, the increase in the duration of the compound action potential is an indicator of alteration in the discharge synchronization of nerve fibers.34
The physiological mechanisms of the hypoalgesic effect of cryotherapy have not yet been completely elucidated. Different hypotheses have been proposed: (1) closing of the pain gate, (2) counter-irritant effect that activates inhibitory control mechanisms, (3) increase in the activation threshold of nociceptors, and (4) participation of descending pathways of the central nervous system that modulate pain by releasing endogenous opiates. It also has been suggested that the hypoalgesic effect of cryotherapy could be related to an increase in pain threshold and pain tolerance associated with a decrease in NCV.9 We suggest that the inactivation of some nerve fibers, which is evident in the decrease in compound action potential amplitude, as well as the change in the synchronization response of these fibers could be other important physiological mechanisms for the hypoalgesic effect of cryotherapy. Studies are needed to investigate this hypothesis. Although the present study did not include specific pain measurements, the results for the 3 modalities suggest that the hypoalgesic effect of cryotherapy may be produced mainly by the reduction in sensory fiber conduction because the cooling effect on the conduction parameters was usually greater in the sensory nerve than in the motor nerve (Tabs. 3 and 4). Ice massage, ice pack, and cold water immersion reduced sural NCV by 37.9%, 31.9%,
Volume 90
Number 4
Physical Therapy f
587
588
f
Physical Therapy
Volume 90
Number 4
14.9 (4.00) [12.36, 17.44]
5.39 (0.50) [5.07, 5.70]
Amplitude (mV)
Duration (ms)
5.74 (0.58) [5.37, 6.11]
14.04 (3.98) [11.51, 16.57]
47.17 (3.11) [45.20, 49.14]
3.62 (0.46) [3.32, 3.91]
Post-cooling
Pre-cooling 3.82 (0.60) [3.44, 4.21] 49.58 (3.73) [47.21, 51.95] 16.72 (2.70) [15.00, 18.43] 5.63 (1.04) [4.96, 6.29]
0.15 (0.17)c [0.04, 0.26] ⫺2.50 (1.31)c [⫺3.34, ⫺1.67] ⫺0.86 (1.77) [⫺1.99, 0.27] 0.35 (0.26)c [0.18, 0.52]
Difference
b
39.29 (12.18) [31.55, 47.03]
1.26 (0.10) [1.20, 1.32]
Amplitude (V)
Duration (ms)
1.40 (0.17) [1.30, 1.50]
18.95 (11.30) [11.77, 26.13]
33.50 (5.76) [29.84, 37.16]
4.62 (1.21) [3.85, 5.39]
Post-cooling
b
Data are presented as mean (SD) [95% confidence interval. Difference⫽post-cooling ⫺ pre-cooling. c P⬍.05. d P⬍.0001.
a
53.92 (2.84) [52.11, 55.72]
2.93 (0.27) [2.76, 3.11]
Latency (ms)
Sensory nerve conduction velocity (m/s)
Pre-cooling
Nerve Conduction Parameter
Ice Massage
1.38 (0.15) [1.28, 1.47]
0.14 (0.14)c [0.05, 0.23]
52.42 (4.27) [49.70, 55.13]
⫺20.42 (5.96)c [⫺24.20, ⫺16.63]
39.97 (11.31) [32.78, 47.15]
3.17 (0.33) [2.96, 3.38]
1.68 (1.04)c [1.02, 2.35]
⫺20.34 (9.54)d [⫺26.40, ⫺14.28]
Pre-cooling
b
Difference
Parameters of Sural Nerve Conduction Before and After Cold Modality Applicationa
Table 4.
b
Data are presented as mean (SD) [95% confidence interval]. Difference⫽post-cooling ⫺ pre-cooling. c P⬍.05. d P⬍.0001.
a
49.67 (3.31) [47.56, 51.77]
3.47 (0.43) [3.19, 3.74]
Latency (ms)
Motor nerve conduction velocity (m/s)
Pre-cooling
Nerve Conduction Parameter
Ice Massage
Parameters of Tibial Motor Nerve Conduction Before and After Cold Modality Applicationa
Table 3.
1.53 (0.19) [1.40, 1.64]
24.86 (12.32) [17.03, 32.69]
35.67 (6.91) [31.28, 40.05]
4.50 (0.71) [4.05, 4.95]
Post-cooling
Ice Pack
6.00 (1.01) [5.36, 6.64]
16.12 (2.80) [16.12, 14.34]
47.50 (2.81) [45.71, 49.29]
4.02 (0.63) [3.61, 4.42]
Post-cooling
Ice Pack Difference
0.15 (0.11)c [0.08, 0.22]
⫺15.11 (11.10)c [⫺22.16, ⫺8.05]
⫺16.75 (5.53)c [⫺20.26, ⫺13.24]
1.33 (0.60)d [0.95, 1.71]
Difference
0.38 (0.16)d [0.27, 0.48]
⫺0.60 (1.04) [⫺1.26, 0.06]
⫺2.08 (1.56)d [⫺3.08, ⫺1.09]
0.19 (0.14)c [0.10, 0.28]
Pre-cooling
1.33 (0.14) [1.23, 1.42]
40.79 (14.85) [31.36, 50.22]
54.08 (4.56) [51.18, 56.98]
3.08 (0.38) [2.84, 3.32]
8.64 (1.49) [7.70, 9.59]
12.53 (3.06) [10.58, 14.47]
40.67 (2.84) [38.86, 42.47]
6.99 (0.93) [6.40, 7.58]
Post-cooling
Difference
3.18 (1.37)c [2.31, 4.05]
⫺3.09 (3.20)c [⫺5.13, ⫺1.06]
⫺8.33 (2.19)c [⫺9.72, ⫺6.94]
3.38 (0.73)d [2.92, 3.85]
2.68 (0.20) [2.55, 2.80]
44.18 (14.49) [34.97, 53.39]
31.50 (2.71) [29.78, 33.22]
5.39 (0.70) [4.95, 5.83]
Post-cooling
1.35 (0.17)d [1.24, 1.46]
3.39 (7.99) [⫺1.69, 8.47]
⫺22.58 (2.35)c [⫺24.08, ⫺21.09]
2.31 (0.38)d [2.07, 2.55]
Difference
Cold Water Immersion
5.46 (0.67) [5.03, 5.88]
15.63 (3.04) [13.69, 17.56]
49.00 (3.59) [46.72, 51.28]
3.61 (0.55) [3.26, 3.96]
Pre-cooling
Cold Water Immersion
Cold Modalities and Nerve Conduction
April 2010
Cold Modalities and Nerve Conduction Table 5. Effects of Cold Modalities on Nerve Conduction Parameters (Analysis of Covariance, Using the Ice Massage Group as Reference) Tibial Motor Nerve Ice Pack Parameter
Sural Nerve
Cold Water Immersion
Coefficient ()
Probability (P)
Coefficient ( )
Ice Pack
Probability (P)
Coefficient ( )
Cold Water Immersion
Probability (P)
Coefficient ( )
Probability (P)
Latency
0.03
.87
3.23
⬍.0001
⫺0.54
.08
0.50
.09
Nerve conduction velocity
0.39
.51
⫺6.02
⬍.0001
3.18
.12
⫺2.11
.29
Amplitude
0.65
.47
⫺2.08
.022
5.43
.15
24.16
⬍.0001
Duration
0.04
.92
2.84
⬍.0001
0.02
.73
1.22
⬍.0001
and 41.8%, respectively. In contrast, these modalities reduced tibial motor NCV by only 5.0%, 4.2%, and 17.0%, respectively. It is difficult to compare these results with those of previous studies9,31 because of the different intervention protocols and analyzed nerves. Algafly and George9 applied an ice pack for a mean time of 26 minutes and obtained a skin temperature of 10°C and a 33% reduction in sensory plantar NCV. McMeeken et al31 used an ice pack for 15 minutes and obtained a skin temperature of 5.6°C and an approximate reduction of 13% in ulnar motor NCV. Even though these studies measured different nerves, in a broad sense they corroborate the results of the present study, which demonstrated a greater cooling effect on the sensory fibers than on the motor fibers. The greater sensibility of the sensory fibers to cooling may be due to their more superficial location compared with the motor fibers, which would explain why the functional effects of cryotherapy on sensibility42 are more pronounced than the effects on muscle function.12 The depression of sensory and motor NCV derived from cooling modalities also may indicate the risks of deleterious effects associated with prolonged icing, such as skin burn and superficial nerve damage. Previous studies33,43 have shown cases of nerve palsy resulting from ice application near the subcutaneous course April 2010
of nerves. The consequent disability was transient (1– 4 days) or prolonged (4 – 6 months), with all patients eventually reaching full recovery. The application of cryotherapy is typically safe and beneficial if the protocol is appropriate and sufficiently monitored. However, clinicians must be aware of the location of major peripheral nerves, the thickness of the overlying subcutaneous fat, the method of application, the duration of tissue cooling, and the surface area covered.33,43 The results of the comparison of the effects of the 3 cold modalities on conduction parameters show that the cold water immersion protocol used in this study, although neither the most comfortable modality for the participant nor the easiest to apply, may be the most indicated for greater therapeutic effect mediated by the change in motor conduction (eg, in muscle spasm and spasticity [hypertonicity]). In contrast, the hypoalgesic effect could be induced by any of the 3 assessed modalities. Cold water immersion was more efficient in changing some parameters of sensory conduction, but the application of the 3 modalities lowered skin temperature to less than 13.6°C and reduced NCV by more than 10%. As suggested in previous studies,10,17 these changes could be associated with the hypoalgesic effect of cryotherapy.
The present study had some methodological limitations that restrict the generalization of the results. The cooling area of the ice massage and ice pack was small compared with that of the cold water immersion and possibly smaller than those used in the clinical setting. The study sample comprised only young participants who were healthy, and the responses might have been different in older adults and individuals with clinical disorders. The time used for each modality (15 minutes) also may have been insufficient to induce greater effects on the motor nerve, especially in the case of the ice pack and the ice massage, which were applied to restricted areas. Considering that the nerve conduction evaluations were taken immediately before and after the cold modality application, the examiner was not blinded to the treatment group. This fact may limit the internal validity of the study. Subsequent studies are needed to determine the functional relevance of changes in nerve conduction induced by the cold modalities on sensibility and muscle strength, as well as the clinical importance of these changes. The present study contributes to the literature because, to our knowledge, it is the first study comparing the effect of 3 modalities frequently used in clinical practice on the parameters of motor and sensory nerve conduction.
Volume 90
Number 4
Physical Therapy f
589
Cold Modalities and Nerve Conduction
Conclusions Ice massage, ice pack, and cold water immersion were effective in reducing skin temperature and changing most of the motor and sensory conduction parameters, with greater effects on the sensory nerve. Cold water immersion, as applied in this study, was the most effective modality in changing nerve conduction, especially in the tibial motor nerve. Our results can be considered clinically relevant and contribute to the informed choice of a cryotherapy modality based on the desired physiological and therapeutic effects. All authors provided concept/idea/research design, writing, and consultation (including review of manuscript before submission). Ms Herrera and Dr Sandoval provided data collection. Ms Herrera, Dr Sandoval, and Ms Camargo provided data analysis. Ms Herrera provided project management and participants. Dr Salvini provided facilities/equipment. The study protocol was approved by the Institutional Ethics Committee of Universidad Industrial de Santandar. The study followed the Declaration of Helsinki. Ms Herrera acknowledges Coordenac¸a˜o de Aperfeic¸oamento de Pessoal de Nı´vel Superior (CAPES, Brazil) for providing her doctoral grant. This article was received April 22, 2009, and was accepted December 6, 2009. DOI: 10.2522/ptj.20090131
References 1 Knight KL. Cryotherapy in Sport Injury Management. Champaign, IL: Human Kinetics; 1995. 2 Nadler SF, Weingand K, Kruse RJ. The physiological basis and clinical applications of cryotherapy and thermotherapy for the pain practitioner. Pain Physician. 2004;7:395–399. 3 Bleakley C, McDonough S, MacAuley D. Use of ice in the treatment of acute softtissue injury: a systematic review of randomized controlled trials. Am J Sports Med. 2004;32:251–261. 4 Oliveira NML, Rainero EP, Salvini TF. Three intermittent sessions of cryotherapy reduce the secondary muscle injury in skeletal muscle of rat. Journal Sports Science and Medicine. 2006;5:228 –234.
590
f
Physical Therapy
Volume 90
5 Hubbard TJ, Aronson SL, Denegar CR. Does cryotherapy hasten return to participation? A systematic review. J Athl Train. 2004;39:88 –94. 6 Olson JE, Stravino VD. A review of cryotherapy. Phys Ther. 1972;52:840 – 853. 7 Kowal MA. Review of physiological effects of cryotherapy. J Orthop Sports Phys Ther. 1983;5:66 –73. 8 Hocutt JE. Cryotherapy. Am Fam Physician. 1981;23:141–144. 9 Algafly AA, George KP. The effect of cryotherapy on nerve conduction velocity, pain threshold and pain tolerance. Br J Sports Med. 2007;41:365–369. 10 Chesterton LS, Foster NE, Ross L. Skin temperature response to cryotherapy. Arch Phys Med Rehabil. 2002;83:543–549. 11 Ruiz DH, Myrer JW, Durrant E, Fellingham GW. Cryotherapy and sequential exercise bouts following cryotherapy on concentric and eccentric strength in the quadriceps. J Athl Train. 1993;28:320 –323. 12 Rubley MD, Denegar CR, Buckley WE, Newell KM. Cryotherapy, sensation, and isometric-force variability. J Athl Train. 2003;38:113–119. 13 Kimura IF, Gulick DT, Thompson GT. The effect of cryotherapy on eccentric plantar flexion peak torque and endurance. J Athl Train. 1997;32:124 –126. 14 Merrick MA, Jutte LS, Smith ME. Cold modalities with different thermodynamic properties produce different surface and intramuscular temperatures. J Athl Train. 2003;38:28 –33. 15 Kanlayanaphotporn R, Janwantanakul P. Comparison of skin surface temperature during the application of various cryotherapy modalities. Arch Phys Med Rehabil. 2005;86:1411–1415. 16 Janwantanakul P. Different rate of cooling time and magnitude of cooling temperature during ice bag treatment with and with damp towel wrap. Phys Ther Sport. 2004;5:156 –161. 17 Kennet J, Hardaker N, Hobbs S, Selfe J. Cooling efficiency of 4 common cryotherapeutic agents. J Athl Train. 2007;42: 343–348. 18 Zemke JE, Andersen JC, Guion WK, et al. Intramuscular temperature responses in the human leg to two forms of cryotherapy: ice massage and ice bag. J Orthop Sports Phys Ther. 1998;27:301–307. 19 Warren TA, McCarty EC, Richardson AL, et al. Intra-articular knee temperature changes: ice versus cryotherapy device. Am J Sports Med. 2004;32:441– 445. 20 Myrer JW, Measom G, Fellingham GW. Temperature changes in the human leg during and after two methods of cryotherapy. J Athl Train. 1998;33:25–29. 21 Belitsky RB, Odam SJ, Hubley-Kozey C. Evaluation of the effectiveness of wet ice, dry ice, and cryogen packs in reducing skin temperature. Phys Ther. 1987;67: 1080 –1084.
Number 4
22 Jutte LS, Merrick MA, Ingersoll CD, Edwards JE. The relationship between intramuscular temperature, skin temperature, and adipose thickness during cryotherapy and rewarming. Arch Phys Med Rehabil. 2001;82:845– 850. 23 Myrer JW, Myrer KA, Measom GJ, et al. Muscle temperature is affected by overlying adipose when cryotherapy is administered. J Athl Train. 2001;36:32–36. 24 Otte JW, Merrick MA, Ingersoll CD, Cordova ML. Subcutaneous adipose tissue thickness alters cooling time during cryotherapy. Arch Phys Med Rehabil. 2002;83: 1501–1505. 25 Herrera E, Camargo DM, Delgado DC, Salvini TF. Reliability of superficial peroneal, sural, and medial plantar nerve conduction studies: analysis of statistical methods. J Clin Neurophysiol. 2009;26:295–380. 26 Greathouse DG, Currier DP, Joseph BS, et al. Electrophysiologic responses of human sural nerve to temperature. Phys Ther. 1989;69:914 –922. 27 Dioszeghy P, Stalberg E. Changes in motor and sensory nerve conduction parameters with temperature in normal and diseased nerve. Electroencephalogr Clin Neurophysiol. 1992;85:229 –235. 28 Halar EM, DeLisa JA, Brozovich FV. Nerve conduction velocity: relationship of skin subcutaneous and intramuscular temperatures. Arch Phys Med Rehabil. 1980;61:199–203. 29 Halar EM, Delisa JA, Brozovich FV. Peroneal nerve conduction velocity: the importance of temperature correction. Arch Phys Med Rehabil. 1981;62:439 – 443. 30 Halar EM, DeLisa JA, Soine TL. Nerve conduction studies in upper extremities: skin temperature corrections. Arch Phys Med Rehabil. 1983;64:412– 416. 31 McMeeken J, Lewis MM, Cocks S. Effects of cooling with simulated ice on skin temperature and nerve conduction velocity. Aust J Physiother. 1984;30:111–114. 32 Pagano M. Principles of Biostatistics. Belmont, CA; Duxbury Press; 2000. 33 Malone TR, Engelhardt DL, Kirkpatrick JS, Bassett FH. Nerve injury in athletes caused by cryotherapy. J Athl Train. 1992;27: 235–237. 34 Oh SJ. Clinical Electromyography: Nerve Conduction Studies. 3rd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2002. 35 Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet. 1986;8:307–310. 36 Siegel S, Castellan NJ Jr. Nonparametric Statistics for the Behavioral Sciences. 2nd ed. New York, NY: McGraw-Hill Inc; 1988. 37 Vickers AJ, Altman DG. Statistics notes: analysing controlled trials with baseline and follow up measurements. BMJ. 2001; 323:1123–1124. 38 Tikuisis P, Gonzalez RR, Pandolf KB. Thermoregulatory model for immersion of humans in cold water. J Appl Physiol. 1988; 64:719 –727.
April 2010
Cold Modalities and Nerve Conduction 39 Luzzati V, Mateu L, Marquez G, Borgo M. Structural and electrophysiological effects of local anesthetics and of low temperature on myelinated nerves: implication of the lipid chains in nerve excitability. J Mol Biol. 1999;286:1389 –1402. 40 Kiernan MC, Cikurel K, Bostock H. Effects of temperature on the excitability properties of human motor axons. Brain. 2001; 124:816 – 825.
April 2010
41 Bolton CF, Sawa GM, Carter K. The effects of temperature on human compound action potentials. J Neurol Neurosurg Psychiatry. 1981;44:407– 413. 42 Saeki Y. Effect of local application of cold or heat for relief of pricking pain. Nurs Health Sci. 2002;4:97–105.
43 Drez D, Faust DC, Evans JP. Cryotherapy and nerve palsy. Am J Sports Med. 1981; 9:256 –257.
Volume 90
Number 4
Physical Therapy f
591
Case Report
Development of a Therapeutic Exercise Program for Patients With Osteoarthritis of the Hip Linda Fernandes, Kjersti Storheim, Lars Nordsletten, May Arna Risberg L. Fernandes, PT, MSc, is a PhD student, Norwegian Research Center for Active Rehabilitation (NAR), Orthopedic Centre, Oslo University Hospital, Ullevaal and Hjelp24NIMI, Oslo, Norway. Mailing address: FoU, Ortopedisk Senter, Bevegelsesdivisjonen Bygg 73, 2, Ullevål Sykehus Kirkeveien 166, 0407 Oslo, Norway. Address all correspondence to Ms Fernandes at: linda.fernandes@medisin. uio.no. K. Storheim, PT, PhD, is Senior Researcher, Norwegian Research Center for Active Rehabilitation (NAR), Orthopedic Centre, Oslo University Hospital, Ullevaal and Hjelp24NIMI. L. Nordsletten, MD, PhD, is Professor, Orthopedic Center, Oslo University Hospital, Ullevaal. M.A. Risberg, PT, PhD, is Associate Professor, Norwegian Research Center for Active Rehabilitation (NAR), Orthopedic Centre, Oslo University Hospital, Ullevaal and Hjelp24NIMI, and Norwegian School of Sport Sciences, Oslo, Norway. [Fernandes L, Storheim K, Nordsletten L, Risberg MA. Development of a therapeutic exercise program for patients with osteoarthritis of the hip. Phys Ther. 2010;90:592– 601.]
Background and Purpose. No detailed exercise programs specifically for patients with hip osteoarthritis (OA) have been described in the literature. This lack of data creates a gap between the recommendation that people with OA should exercise and the type and dose of exercises that they should perform. The purpose of this case report is to describe and demonstrate the use of a therapeutic exercise program for a patient with hip OA. Case Description. A 58-year-old woman with hip OA completed a 12-week therapeutic exercise program (TEP) with a 6-month follow-up. The patient reported hip pain, joint stiffness, and limited physical function, and she had decreased hip range of motion (ROM) at baseline. Outcomes. The patient performed 19 sessions during the TEP, with a mean of 19.5 exercises per session. She increased the resistance in 3 of 5 strength (forcegenerating capacity) training exercises and achieved the highest degree of difficulty in all functional exercises. During the TEP and follow-up, the patient reported improvements in pain, joint stiffness, and physical function. Performance improved on the following physical tests: isokinetic peak torque strength (60°/s) in hip extension (40%), hip flexion (27%), knee extension (17%), and knee flexion (42%); hip ROM extension (8°); and 6-minute walk distance (83 m).
Discussion. The patient experienced less pain and improved physical function and physical test outcomes after intervention and at the 6-month follow-up. The main challenges when prescribing an exercise program for a patient with hip OA are monitoring the exercises to provide improvements without provoking persistent pain and motivating the patient to achieve long-term adherence to exercising. Randomized clinical trials are needed to evaluate the efficacy of this TEP in patients with hip OA.
© 2010 American Physical Therapy Association
Post a Rapid Response or find The Bottom Line: www.ptjournal.org 592
f
Physical Therapy
Volume 90
Number 4
April 2010
Therapeutic Exercise Program for Patients With Osteoarthritis of the Hip
N
onoperative management of osteoarthritis (OA) has been suggested to prevent or delay the impact of disability, and interventions such as patient education and exercise have been recommended as the first choice of treatment.1,2 A review of the guidelines and recommendations for the management of hip and knee OA showed that aerobic and strengthening exercise were the most frequently reported interventions.3 The evidence showing the effect of exercises, so far, has been based on studies of patients with knee OA. A recent meta-analysis, in which the authors were able to extract hip joint–specific data from 9 randomized clinical trials (RCTs) that evaluated the effect of exercise on both knee and hip OA, showed a significant effect size for exercise (0.43).4 Strengthening exercise was the most common type of exercise in the 9 RCTs included in the metaanalysis, and the authors suggested that strengthening exercise might be the most effective type of exercise.4 However, to our knowledge, no specific exercise programs for patients with hip OA have been described in the literature. To be able to recommend exercise therapy for patients with hip OA, we need more knowledge and detailed descriptions of the exercises included in such therapy programs, why these specific exercises are included, and the dose for each exercise. All RCTs examining the effect of exercise therapy programs should include specific descriptions of the different exercises and the exercise dose. The purpose of this case report, therefore, is to describe the development of and demonstrate the use of a therapeutic exercise program (TEP) for patients with hip OA. More specific aims of the TEP are to reduce pain and to improve strength (force-generating capacity), flexibility, and physical function.
April 2010
Table 1. Patient Characteristics at Baseline Variable
Data
Age (y)
58
Sex
Female 2
a
Body mass index (kg/m )
24.5
Minimum joint space (mm)
Target hip: 2.6 Contralateral hip: 2.3
Harris Hip Score (0–100 points)a
79
Medication
None
Comorbidities
None
0⫽extreme pain and limited function, 100⫽no pain or functional limits.
Patient History and Review of Systems The patient was included in a large RCT evaluating the effect of exercise in addition to patient education for patients with hip OA. The patient had attended a group-based patient education program.5 At inclusion, the patient was examined clinically by a physical therapist (L.F.) and an orthopedic surgeon (L.N.). The radiograph was examined by the orthopedic surgeon. Written informed consent was obtained from the patient. The patient was a 58-year-old woman who worked full-time as an information consultant and who was seeking health care because of hip pain. She had radiographically verified hip OA bilaterally6 (Tab. 1) and reported unilateral hip pain located over the right gluteal area, groin, and inner thigh (target hip). Thus, the painful hip was selected as the target hip. She reported no low back pain, knee pain, or any comorbidities, and she did not take any medications. The first hip pain episode was 9 years before inclusion in the RCT, and she reported having intermittent pain since then. The pain increased after walking on hard surfaces, but walking on paths or hiking in the woods did not provoke pain. The patient reported sensations of morning stiffness and stiffness after inactivity. At
inclusion, the patient scored 79 points on the Harris Hip Score (HHS)7 (Tab. 1). At our institution, an HHS of ⬍60 points has been used as the cutoff criterion for surgery. The clinical impression of the patient was consistent with symptoms typical of patients with hip OA, although the symptoms were not severe enough that surgery was considered.
Examination The questionnaires were completed and physical tests were performed at baseline, after intervention, and at follow-up 6 months after the intervention. Administrations and tests were carried out by a physical therapist (L.F.). The questionnaires that were administered were: (1) the disease-specific Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC)–VA3.1,8 (2) the Physical Activity Scale for the Elderly (PASE),9 and (3) the Medical Out-
Available With This Article at ptjournal.apta.org • eAppendix: Therapeutic Exercise Program for Patients With Osteoarthritis of the Hip • Audio Abstracts Podcast This article was published ahead of print on February 25, 2010, at ptjournal.apta.org.
Volume 90
Number 4
Physical Therapy f
593
Therapeutic Exercise Program for Patients With Osteoarthritis of the Hip Table 2. Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) Scores (0 –100 mm)a at Baseline, After Intervention, and at 6-Month Follow-up Dimension Pain
Item Mean score How much pain do you have?
Stiffness
Physical function
22.8
3.6
10.8
26
4
17
22
4
15
At night while in bed
24
3
5
Sitting or lying
15
3
9
Standing upright
27
4
8
42.5
9.0
7.5
After awakening in the morning
41
10
11
After sitting, lying, or resting later in the day
44
8
4
17.5
5.6
8.1
Descending stairs
10
4
7
Ascending stairs
28
5
6
Rising from a sitting position
32
9
8
3
5
4
Bending over to floor
34
4
22
Walking on flat surface
16
5
5
Getting in and out of car
16
13
9
Going shopping
15
3
6
Putting on socks or stockings
31
4
12
Rising from bed
23
6
9
Taking off socks/stockings
27
6
18
Lying in bed
10
2
3
Getting in and out of bath
17
10
13
Sitting
10
3
4
Standing
Getting on and off toilet
a
6-Month Follow-up
Going up or down stairs
Mean score What degree of difficulty do you have?
After Intervention
Walking on flat surface
Mean score How severe is your stiffness?
Baseline
6
3
4
Heavy domestic duties
12
11
6
Light domestic duties
7
3
2
0⫽no pain/stiffness/difficulty, 100⫽extreme pain/stiffness/difficulty.
comes Study’s 36-Item Short-Form Health Survey (SF-36), version 2.10 All questionnaires are considered reliable and valid measures.8,11–13 The physical tests were: (1) a test of isokinetic concentric peak torque muscle strength measured with a dynamometer* at 60°/s for hip and knee extension and flexion14; (2) the SixMinute Walk Test (6MWT)15; (3) a
submaximal cycle ergometer† test16; and (4) range of motion (ROM) in hip flexion and extension, abduction and adduction, and medial (internal) and lateral (external) rotation16 –18 measured with a 1-degree–increment plastic goniometer.‡ After the 6MWT, the patient scored the maximum pain intensity experienced during
* Technogym SpA, Via Perticari, 20 Gambettola, Italy.
†
594
f
Physical Therapy
Volume 90
‡
Monark Exercise AB, 780 50 Vansbro, Sweden. Medema, Box 1169, 171 23 Solna, Sweden.
Number 4
the 6MWT on a 100-mm visual analog scale (VAS). Clinical Impression The patient scored 22.8 mm on the WOMAC pain scale and 57.5 points on the SF-36 bodily pain scale at baseline (Tabs. 2 and 3). Selfreported physical function on the WOMAC and SF-36 were 17.5 mm and 85 points, respectively. On the physical tests, the patient showed
April 2010
Therapeutic Exercise Program for Patients With Osteoarthritis of the Hip Table 3. Harris Hip Score (HHS), Physical Activity Scale for the Elderly (PASE), and 36-Item Short-Form Health Survey (SF-36) Scores at Baseline, After Intervention, and at 6-Month Follow-up Measure
Baseline
HHS (0–100 points)b PASE (0–315 points)
79
c
SF-36 (0–100 points)b
111.59
c
91
140.35
Reference Valuesa –
90.07
–
95
100
85.663
Role limitations–physical
100
100
100
77.6
Role limitations–emotional
100
100
100
84.3
100
80
73.8
100
57.5
100
100
86.0
Mental health
90
95
90
79.5
Vitality
75
75
68.75
62.0
General health perceptions
95
95
75
74.7
Social functioning
b
96
6-Month Follow-up
85
Physical functioning
Bodily pain
a
After Intervention
Missing data (–). 0⫽extreme pain and limited function, 100⫽no pain or functional limits. 0⫽not active, 315⫽extremely active.
similar muscle strength in both the target joint and the contralateral joint but had less ROM in hip extension, abduction, and lateral rotation in the target joint than in the contralateral joint and less hip extension and abduction in the target joint compared with normative data.19 She walked 665 m during the 6MWT and scored 22 mm on the VAS, and she was classified as having a “high” predicted aerobic capacity20 based on the cycle test (Tab. 4). In summary, the baseline data showed that the patient had hip OA with mild pain21 and an acceptable symptom state.22 She had experienced intermittent hip pain, with the pain distribution commonly seen in patients with hip OA.23 Pain increased while walking on flat surfaces, standing, and walking up or down stairs, and she experienced mild pain during the 6MWT. She had limited ROM during hip extension, abduction, and lateral rotation, and she reported having difficulty putting on socks, bending to the floor, and rising from a sitting position. She agreed to participate in a 12-week exercise program.
April 2010
Therapeutic Exercise Program Warm-up and Walking Instructions The exercise program started with 5 to 10 minutes of warm-up walking on a treadmill or cycling on a stationary cycle (exercises 1A and 1B in the eAppendix, available at ptjournal. apta.org). As shown in biomechanical studies, patients with hip OA appear to alter their walking pattern, probably because of pain and altered joint loading.24,25 The patient was instructed to walk symmetrically (ie, to maintain an equal cadence during walking and to extend the hip during the push-off phase of gait.) The intensity of the warm-up was set to 12 to 13 on the Borg Rating of Perceived Exertion Scale.26 Strength Training Two case-control studies of patients with hip OA have shown less muscle mass and muscle strength in the pelvis and thigh muscles compared with control participants or the control limb.27,28 Despite the existence of a few studies on the importance of muscle strength for patients with
hip OA,29 strength training has been considered a key factor in maintaining physical independence and is recommended to be performed twice weekly, both in rehabilitation and in public health studies.30 –32 Progression procedures for the strengthening exercises were aimed at increasing the resistance. Strengthening exercises for both hip and core muscles were included based on previous studies of hip muscle activation during the performance of core exercises.33 The patient performed hip extension of the gluteal muscle in a standing position, leaning halfway forward on stabilization pads (exercise 2C in the eAppendix). Crunches were performed lying supine on a mat with hip and knees partially flexed (exercise 2E in the eAppendix). Bridging also was performed on a mat, starting with 2-legged support (exercise 2Fa in the eAppendix) and advancing to 1-legged support, with the other leg extended and lifted about 20 cm above the floor (exercise 2Fb in the eAppendix). Side-lying hip abduction was performed on a mat on the floor (exercise 2Ga in the eAppen-
Volume 90
Number 4
Physical Therapy f
595
Therapeutic Exercise Program for Patients With Osteoarthritis of the Hip Table 4. Muscle Strength Tests, Six-Minute Walk Test (6MWT), Submaximal Ergometer Cycle Test, and Hip Range of Motion (ROM) of the Target Hip and Contralateral Hip at Baseline, After Intervention, and at 6-Month Follow-upa Baseline Target Hip
Test Isokinetic strength at 60°/s (N䡠m)
Hip extension Hip flexion Knee extension Knee flexion
Target Hip
Contralateral Hip
83
143
122
139
132
68
76
67
89
87
100 1.43b
97 1.39b
48
49
c
Cycle test (mL/kg⫻min) Flexion
6-Month Follow-up Target Hip
70
106 1.51b
109 1.56b
65
665.4
Visual analog scale (0–100 mm)
Contralateral Hip
102
6MWT (m)
Hip ROM (°)
After Intervention
Contralateral Hip
60
117 1.67b
114 1.63b
68
720.0
Reference Values
1.9419,b
71 528.064
748.2
22
0
11
38
40
36
37–4165,d 127–15566
143
138
139
136
–
–
Extension
1
9
9
9
–
–
16–35
Abduction
28
34
31
29
–
–
35–50
Adduction
25
25
26
25
–
–
24–37
Medial rotation
50
48
59
48
–
–
34–71
Lateral rotation
26
42
32
48
–
–
25–56
a
Missing data (–). Newton-meters per kilogram. 0⫽no pain, 100⫽extreme pain. d High oxygen uptake capacity. b c
dix). Progression in the hip abduction exercise was included in the side-lying plank exercise (exercises 2Gb and 2Gc in the eAppendix). Leg extension and leg curl exercises were performed to strengthen the quadriceps and hamstring muscles, respectively (exercises 2A and 2B in the eAppendix). Heel-raise exercise to strengthen the gastrocnemius muscles was performed in a standing position with weight on the shoulders (exercise 2D in the eAppendix). Functional Exercises The rationale for including functional exercises in this TEP was based on hip OA studies in which patients experienced difficulties in performing activities of daily living, such as rising from a sitting position, standing, and ascending and descending stairs.29,34 The functional exercises for this TEP were chosen to imitate those movements required in daily activities. The procedures to ensure progression in the functional 596
f
Physical Therapy
Volume 90
exercises were aimed at increasing the degree of difficulty by reducing the base of support, adding dynamic movements, or increasing the range though which a movement was performed.35 The functional exercises for this TEP included squats performed initially from a standing position and progressing to standing on a balance pad and further to squats with weight on the shoulders (exercises 3Aa– c in the eAppendix). The other functional exercises were: single-leg stance on a balance pad, with progression to single-leg squat (exercises 3Ba and 3Bb in the eAppendix); forward and sideways lunges (exercise 3C and 3D in the eAppendix); and step-up and step-down onto a stool at 2 different heights and advancing to a balance pad on the stool (exercise 3E in the eAppendix). All exercises emphasized the need to perform the exercise accurately by controlling the movement. An accurate perfor-
Number 4
mance of the exercises was defined as the patient’s performing the exercises through the full available ROM. Control of movement for dynamic double-limb–support exercises was defined as keeping the knees over the balls of the feet, referred to as the “athletic position.”36 For single-limb– support exercises, control of movement was defined as keeping the knees over balls of the feet and maintaining hip alignment (ie, not dropping or rotating the pelvis) while performing the exercises. Flexibility Exercises and Stretching Decreased ROM in the hip joint is common in patients with hip OA.37–39 The ROM exercises for this TEP involved both relaxed, repetitive movements and static stretching. The relaxed, repetitive movements were performed lying on a mat on the floor with one leg suspended in a sling fixed to the ceiling. The patient repeatedly moved the leg in April 2010
Therapeutic Exercise Program for Patients With Osteoarthritis of the Hip hip flexion and extension and in hip abduction and adduction for 2 minutes in each direction (exercises 4A and 4B in the eAppendix). The static stretching followed the stretching exercises that were included in the patient education program reported by Klassbo et al,5 which emphasized maintenance of ROM needed for activities of daily living, rather than stretching of specific muscles (exercises 5A–D in the eAppendix). The patient was asked to hold the static stretch for 30 seconds in each direction. Dose No study has described the optimal dose of exercise in patients with hip OA. The total amount of physical activity can be described in terms of intensity, volume, frequency, and duration.32,40 A systematic review on progressive resistance training in elderly people showed that the intensity of the training was the strongest factor affecting strength and functional outcome.41 Similarly, a study comparing high- versus low-resistance strength training in patients with knee OA showed consistently larger effect sizes for the high-resistance training group.42 The latest update of physical activity recommendations are to perform 8 to 10 strengthening exercises of the major muscle groups twice or more per week.32,35 For untrained and recreationally trained individuals, an intensity of 60% to 80% of 1 repetition maximum (1RM), a volume of 4 sets, and a frequency of 2 to 3 times per week have been recommended,43,44 with at least 6 weeks of progressive training to achieve muscular hypertrophy.45 Therefore, the TEP was set as 3 sets of 8 repetitions for strengthening exercises, corresponding to 70% to 80% of 1RM,46 and 3 sets of 10 repetitions for functional exercises, 2 to 3 sessions per week for 12 weeks. Single-leg exercises were performed April 2010
with both the target leg and the contralateral leg. In total, the TEP comprised 26 different exercises of the lower limbs and trunk (eAppendix). The patient was supervised by a physical therapist who modified the TEP during the sessions according to the patient’s capacity by choosing exercises from the 26 suggested exercises (eAppendix). Performance of 10 strengthening and functional exercises per session and exercising twice a week for 12 weeks was considered acceptable adherence.44 Exercise Regulation to Pain Level The physical therapist introduced a pain scale as a tool to help the patient modify the TEP to her pain.47 The therapist informed the patient that the exercise could provoke some pain, especially during the initial exercise period, but this pain should not concern the patient. The therapist also emphasized that the exercises should not exceed the limit for “acceptable pain”22 and that the pain level should decrease to the same level as prior to the exercise session within 24 hours after exercising. Pain after exercise in patients with knee OA has been reported as transient pain (ie, even if the pain increased immediately after an exercise session, the pain decreased to an even lower level later in the day following exercise).48 The therapist determined when to adjust the exercise intensity upward or downward. The intensity was increased in the strengthening exercises by increasing the resistance when the patient could tolerate more than 8 repetitions.46,49 The intensity was increased in the functional exercises by increasing the degree of difficulty of the exercise when the patient was able to perform 10 repetitions with a controlled movement. If the patient reported that an exercise was more painful than “acceptable pain,” the intensity was decreased in the strengthening
exercises by reducing the resistance and in the functional exercises by decreasing the degree of difficulty or excluding the exercise. If pain persisted, the exercise intensity was decreased further until the pain level became acceptable. Because the side-lying plank (exercises 2B and 2C in the eAppendix) and squat with weights (exercise 3Ac in the eAppendix) exercises may stress the neck, shoulders, and back, these exercises were adjusted if the patient experienced any neck, shoulder, or back pain.
Outcomes The patient attended the TEP for 19 sessions during the 12-week period. She performed between 13 and 20 different exercises, with a mean of 19.5 exercises per session. The patient chose to warm up on a stationary cycle for all 19 sessions. She increased the resistance during the hip extension strength exercise, the heel-raise, and the squat exercises with weights, and she progressed on the bridging and hip abduction exercises during the training period (Tab. 5). She tried to increase resistance of the leg extension (exercise 2A in the eAppendix) at session 17 but was unsuccessful because of hip pain (Tab. 5). The patient improved her scores on the WOMAC (Tab. 2), HHS, and SF-36 physical functioning scales (Tab. 3) during the follow-up period. After intervention, the patient reported no pain (100 points) on the SF-36 body pain scale, and she reported mild pain (80 points) at the 6-month follow-up. Compared with the baseline score, the PASE showed a higher activity level score after the intervention and a lower activity level score at the 6-month follow-up (Tab. 3). Isokinetic peak torque strength increased in hip extension (36%), hip flexion (27%), knee extension (17%), and knee flexion (42%) at the 6month follow-up (Tab. 4). The 6MWT
Volume 90
Number 4
Physical Therapy f
597
Therapeutic Exercise Program for Patients With Osteoarthritis of the Hip Table 5. Training Diary of the 19 Sessions Performed by the Patient During the Exercise Interventiona Stage of Program
Exercise
1A
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
Walk Cycle
1B
x
x
x
x
x
x
x
x
x
x
x
Strength training, 3 sets ⫻ 8 repetitions
Leg extension
2A
30
30
30
30
30
30
30
30
30
30
30
Functional exercises, 3 sets ⫻ 10 repetitions
Flexibility
x
x
x
x
x
30
30
30
30
30
17 18 19 –
–
–
x
x
x
35
30
30
Leg curl
2B
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
40
Hip extension
2C
40
40
40
40
45
50
50
50
50
–
55
55
55
55
55
55
60
60
60
Heel-raise
2D
30
30
30
30
30
30
30
30
30
30
35
35
35
35
35
40
35
35
35
Crunches
2E
x
x
x
x
x
x
x
x
x
–
x
–
x
x
x
x
x
x
x
Bridging
2Fa
x
x
x
x
x
–
–
–
–
–
–
–
–
–
–
–
–
–
–
2Fb
–
–
–
–
–
x
x
x
x
–
x
–
x
x
x
x
x
x
x
Squats
Single-leg stance/ squat
b
Session
Warm-up
Hip abduction
a
Exercise No.b
2Ga
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
2Gb
x
x
x
x
x
–
–
–
–
–
–
–
–
–
–
–
–
–
–
2Gc
–
–
–
–
–
x
x
x
x
x
x
–
x
x
x
x
x
x
x
3Aa
x
x
x
x
x
x
x
–
–
–
–
–
–
–
–
–
–
–
–
3Ab
–
–
–
–
–
–
–
x
x
x
x
x
x
x
x
x
x
x
x
3Ac
20
20
20
20
25
25
25
25
25
25
27.5
27.5
27.5
27.5
27.5
27.5
30
30
30
3Ba
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
3Bb
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
Lunge
3C
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
Sideways lunge
3D
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
Step-up/stepdown
3E
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
Flexion/extension
4A
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
Abduction/ adduction
4B
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
x
Extension
5A
x
x
x
x
x
x
x
x
x
x
x
–
x
x
x
x
x
x
x
Abduction
5B
x
x
x
x
x
x
x
x
x
x
x
–
x
x
x
x
x
x
x
Lateral (external) rotation
5C
x
x
x
x
x
x
x
x
x
x
x
–
x
x
x
x
x
x
x
Medial (internal) rotation
5D
x
x
x
x
x
x
x
x
x
x
x
–
x
x
x
x
x
x
x
Exercise performed (x). Exercise not performed (–). Exercise performed and values represent the resistance (in kilograms) used in the exercise. See eAppendix (available at ptjournal.apta.org) for descriptions of exercises in the therapeutic exercise program.
distance walked increased from baseline (665 m) to the 6-month follow-up (748 m), along with a decrease in pain on the VAS from 22 to 11 mm. The predicted maximal aerobic capacity varied from 38 to 40 to 36 mL/kg⫻min from baseline to after the intervention to the follow-up test (Tab. 4). From baseline to after the intervention, ROM increased by 8 degrees in hip extension, by 9 degrees in medial ro-
598
f
Physical Therapy
Volume 90
tation, and by 6 degrees in lateral rotation in the target hip (Tab. 4).
Discussion The purpose of this case report is to describe and demonstrate the use of a TEP designed specifically for patients with hip OA. The patient completed the TEP with no complications. We thought it important to set the dose individually with a grad-
Number 4
ual progression of the resistance and degree of difficulty while keeping the pain level within an acceptable range. The TEP included different types of exercises aimed at reducing pain, strengthening the muscles, increasing flexibility, and improving the patient’s physical function.
April 2010
Therapeutic Exercise Program for Patients With Osteoarthritis of the Hip Strength Training and Functional Exercises Strength training was included in the TEP for several reasons. First, case-control studies have shown low muscle strength and muscle hypotrophy in patients with hip OA.27,28 Second, strength training has been recommended as one treatment modality for patients with hip or knee OA.3,4 Third, RCTs have shown benefits from strength training on physical function and physical independence in young adults and older adults.50 It is impossible to separate strengthening and functional exercises completely because of the overlap between exercises, as functional exercises also include strengthening components (eg, the squat exercise). Similarly, in a patient with low initial functional status, functional exercises such as squats or stepping up and down can be exhausting and thus could be classified as strength training. Therefore, some of the functional exercises could be classified as strengthening exercises. During the follow-up period, the patient’s isokinetic peak torque strength values in hip extension and flexion and knee extension and flexion increased by 17% to 42%, indicating improvements in strength.51,52 The patient’s scores on the WOMAC pain, stiffness, and physical function scales, the HHS, and the SF-36 physical functioning and bodily pain scales improved from baseline to after the intervention. Improvements after the intervention exceeded the minimal perceptible clinical improvements,7,53,54 indicating a potential difference from baseline. The aims of the TEP to reduce pain and to improve strength, flexibility, and physical function appear to have been achieved by this patient. The distance walked in the 6MWT increased, and the patient reported reduced pain during the follow-up period, indicating a trend toward better walking capacity.55 Compared with April 2010
the value at baseline, the predicted aerobic capacity calculated from the cycle test was higher after the intervention but was lower after the 6-month follow-up. This finding may be explained by the reduced activity reported in the PASE during the same period. Flexibility Exercises The rationale for including flexibility exercises in the TEP was that some studies have shown reduced hip ROM and the sensation of stiffness in patients with hip OA.8,37,38 In addition, hip extension and lateral rotation have been reported to be associated with high levels of disability.39 It is important to maintain sufficient hip ROM to manage activities of daily living, which was one purpose of the static stretching exercises in the TEP. The patient increased right hip extension by 8 degrees from baseline to after the intervention, indicating a real increase. The other ROM differences from baseline to after the intervention did not exceed the measurement errors and might not be regarded as real changes.56 The small changes in ROM might indicate that, although the TEP was unable to change ROM, the program helped maintain hip ROM. In contrast, the WOMAC stiffness score decreased from 42.5 mm at baseline to 7.5 mm during the follow-up period, a change that exceeds the minimum clinically important difference57 and thereby indicates a real change. Individually Adjusted Exercise Program There are no reviews comparing a standard regimen with an individually adjusted exercise programs for patients with OA. However, individually designed and supervised exercise programs for patients with low back pain were found to be superior to an unsupervised exercise program.58 Benefits of individualized and supervised exercise programs may be attributed to individually de-
signed exercises and individually set doses. It was considered important that the physical therapist considered the resistance and degree of difficulty of the exercises and the patient’s pain level when setting the dose of the TEP. Pain provoked by exercise has been shown to reduce adherence to the exercise program,59 and we believed it was important to obtain thorough information to keep the pain level within an acceptable range. The patient had one episode of hip pain when trying to increase the load in the leg extension exercise in the 17th session (Tab. 5), so the physical therapist reduced the resistance in the next session. Otherwise, progression in the resistance was seen in hip extension, heel-raise, and squats with weights, and progression in the degree of difficulty was seen in bridging, hip abduction, and squats. Motivation for Exercising The patient attended an education program5 before starting the TEP. The purpose of the education program is to empower patients to manage pain relief themselves and to improve or maintain physical function. This program, in itself, might have motivated the patient to adopt a more active lifestyle. Motivation is a key factor for long-term adherence to exercise and has been considered crucial to maintaining the benefits of exercise.60 It has been shown to be important for the patient to understand why exercise would be beneficial to ensure adherence to an exercise program.61,62 Adherence was defined as performing 10 strengthening and functional exercises per session and exercising twice weekly during the study period. The patient performed a mean of 19.5 exercises per session and a total of 19 sessions during the exercise period. She fulfilled the criterion for adherence for the number of exercises per session, but she did not adhere fully to the number of sessions per week. We
Volume 90
Number 4
Physical Therapy f
599
Therapeutic Exercise Program for Patients With Osteoarthritis of the Hip believed an individually adjusted exercise program with supervision for 12 weeks should give the patient sufficient confidence in to monitor and adjust the exercises to her pain level, thus encouraging the her to continue to exercise over the long term.
Conclusion The main challenge associated with a TEP for patients with hip OA is balancing the progression in such a manner that it does not provoke persistent pain while improving muscular strength and physical function. We recommend that the physical therapist should provide thorough information about the benefits of exercise and how to adjust exercise intensity according to pain level. The patient described in this case report achieved fairly good exercise adherence and had no complications. Her pain level decreased and her muscular strength, walking distance, and physical function increased after the intervention and were maintained at the 6-month follow-up. In summary, the patient showed reduced pain and improved physical function over the follow-up period. Randomized clinical trials are needed to evaluate the efficacy of this TEP for patients with hip OA. All authors provided concept/idea/project design and writing. Ms Fernandes and Dr Nordsletten provided data collection, project management, and the patient. Ms Fernandes, Dr Storheim, and Dr Nordsletten provided data analysis. Dr Storheim, Dr Nordsletten, and Dr Risberg provided fund procurement and consultation (including review of manuscript before submission). Dr Nordsletten and Dr Risberg provided facilities/ equipment. Dr Risberg provided institutional liaisons. The authors acknowledge Sigmund Ruud, senior physical therapist specializing in musculoskeletal diseases and manual therapy, Ullern Clinic; Karin Rydevik, PT, MSc, Hjelp24NIMI; and Emilie Jul-Larsen Aas, PT, BSc, Oslo University Hospital, for feedback on the practical applications of the therapeutic exercise program. They also thank Mr Ruud for guiding the patient through the
600
f
Physical Therapy
Volume 90
therapeutic exercise program. The authors acknowledge Ingrid Eitzen, PT, MSc, Oslo University Hospital, for reading and commenting on the manuscript. The study was carried out according to the Helsinki Declaration and was approved by the regional medical research ethics committee. This case report was supported by grants from the EXTRA funds from the Norwegian Foundation for Health and Rehabilitation, through the Norwegian Rheumatism Association, and by the South Eastern Norway Regional Health Authority. This article was received March 11, 2009, and was accepted October 25, 2009. DOI: 10.2522/ptj.20090083
References 1 Buckwalter JA, Stanish WD, Rosier RN, et al. The increasing need for nonoperative treatment of patients with osteoarthritis. Clin Orthop Rel Res. 2001;(385): 36 – 45. 2 Pollard TCB, Gwilym SE, Carr AJ. The assessment of early osteoarthritis. J Bone Joint Surg Br. 2008;90:411– 421. 3 Zhang W, Moskowitz RW, Nuki G, et al. OARSI recommendations for the management of hip and knee osteoarthritis, part I: critical appraisal of existing treatment guidelines and systematic review of current research evidence. Osteoarthritis Cartilage. 2007;15:981–1000. 4 Hernandez-Molina G, Reichenbach S, Zhang B, et al. Effect of therapeutic exercise for hip osteoarthritis pain: results of a meta-analysis. Arthritis Rheum. 2008;29; 59:1221–1228. 5 Klassbo M, Larsson G, Harms-Ringdahl K. Promising outcome of a hip school for patients with hip dysfunction. Arthritis Rheum. 2003;49:321–327. 6 Danielsson LG. Incidence of osteoarthritis of the hip (coxarthrosis). Clin Orthop Rel Res. 1966;45:67–72. 7 Hoeksma HL, Van Den Ende CH, Ronday HK, et al. Comparison of the responsiveness of the Harris Hip Score with generic measures for hip function in osteoarthritis of the hip. Ann Rheum Dis. 2003;62:935– 938. 8 Bellamy N, Buchanan WW, Goldsmith CH, et al. Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcomes to antirheumatic drug therapy in patients with osteoarthritis of the hip or knee. J Rheumatol. 1988;15:1833–1840. 9 Washburn RA, Zhu W, McAuley E, et al. The physical activity scale for individuals with physical disabilities: development and evaluation. Arch Phys Med Rehabil. 2002;83:193–200. 10 Ware JE Jr. SF-36 health survey update. Spine. 2000;25:3130 –3139.
Number 4
11 Schuit AJ, Schouten EG, Westerterp KR, Saris WH. Validity of the Physical Activity Scale for the Elderly (PASE): according to energy expenditure assessed by the doubly labeled water method. J Clin Epidemiol. 1997;50:541–546. 12 Loland NW. Reliability of the Physical Activity Scale for the Elderly (PASE). Eur J Sport Sci. 2002;2:1–12. 13 McHorney CA, Ware JE Jr, Raczek AE. The MOS 36-Item Short-Form Health Survey (SF-36), II: psychometric and clinical tests of validity in measuring physical and mental health constructs. Med Care. 1993;31: 247–263. 14 Keating JL, Matyas TA. The influence of subject and test design on dynamometric measurements of extremity muscles. Phys Ther. 1996;76:866 – 889. 15 Finch E, Brooks D, Stratford PW, Mayo NE. Physical Rehabilitation Outcome Measures. 2nd ed. Toronto, Ontario, Canada: Canadian Physiotherapy Association; 2002. 16 Norkin C, White DJ. The hip. In: Biblis M, Rhyner S, Seitz A, Forgione LJ, eds. Measurement of Joint Motion. A Guide to Goniometry. 3rd ed. Philadelphia, PA: FA Davis Co; 2003:183–220. 17 Aalto TJ, Airaksinen O, Harkonen TM, Arokoski JP. Effect of passive stretch on reproducibility of hip range of motion measurements. Arch Phys Med Rehabil. 2005; 86:549 –557. 18 Bierma-Zeinstra SM, Bohnen AM, Ramlal R, et al. Comparison between two devices for measuring hip joint motions. Clin Rehabil. 1998;12:497–505. 19 Baron R. Normative data for muscle strength in relation to age, knee angle and velocity. Wiener Medizinische Wochenschrift. 1995;145:600 – 606. 20 Astrand I. Aerobic work capacity in men and women with special reference to age. Acta Physiol Scand Suppl 1960;49:1–92. 21 Kapstad H, Hanestad B, Langeland N, et al. Cutpoints for mild, moderate and severe pain in patients with osteoarthritis of the hip or knee ready for joint replacement surgery. BMC Musculoskeletal Disord. 2008;9:55. 22 Tubach F, Ravaud P, Baron G, et al. Evaluation of clinically relevant states in patient reported outcomes in knee and hip osteoarthritis: the patient acceptable symptom state. Ann Rheum Dis. 2005;64:34 –37. 23 Khan AM, McLoughlin E, Giannakas K, et al. Hip osteoarthritis: where is the pain? Ann R Coll Surg Engl. 2004;86:119 –121. 24 Nakamura H. Gait analysis in coxarthrosis. Kurume Med J. 1999;46:1–7. 25 Aminian K, Trevisan C, Najafi B, et al. Evaluation of an ambulatory system for gait analysis in hip osteoarthritis and after total hip replacement. Gait Posture. 2004;20: 102–107. 26 Borg GAV. Psychophysical bases of perceived exertion. Med Sci Sports Exerc. 1982;14:377–381.
April 2010
Therapeutic Exercise Program for Patients With Osteoarthritis of the Hip 27 Arokoski MH, Arokoski JP, Haara M, et al. Hip muscle strength and muscle cross sectional area in men with and without hip osteoarthritis. J Rheumatol. 2002;29: 2187–2195. 28 Rasch A, Anders H, Dalen N, Berg HE. Reduced muscle radiological density, cross-sectional area, and strength of major hip and knee muscles in 22 patients with hip osteoarthritis. Acta Orthop. 2007;78: 505–510. 29 van Dijk GM, Dekker J, Veenhof C, Van Den Ende CH; for the Carpa Study Group. Course of functional status and pain in osteoarthritis of the hip or knee: a systematic review of the literature. Arthritis Rheum. 2006;55:779 –785. 30 Suetta C, Magnusson SP, Beyer N, Kjaer M. Effect of strength training on muscle function in elderly hospitalized patients. Scand J Med Sci Sports. 2007;17:464 – 472. 31 Bischoff HA, Roos EM. Effectiveness and safety of strengthening, aerobic, and coordination exercises for patients with osteoarthritis. Curr Opin Rheumatol. 2003;15: 141–144. 32 Haskell WL, Lee IM, Pate RR, et al. Physical activity and public health: updated recommendation for adults from the American College of Sports Medicine and the American Heart Association. Med Sci Sports Exerc. 2007;39:1423–1434. 33 Arokoski JP, Kankaanpaa M, Valta T, et al. Back and hip extensor muscle function during therapeutic exercises. Arch Phys Med Rehabil. 1999;80:842– 850. 34 Majewski M, Bischoff-Ferrari HA, Gruneberg C, et al. Improvements in balance after total hip replacement. J Bone Joint Surg Br. 2005;87:1337–1343. 35 Chodzko-Zajko WJ, Proctor DN, Fiatarone Singh MA, et al; for the American College of Sports Medicine. American College of Sports Medicine position stand: exercise and physical activity for older adults. Med Sci Sports Exerc. 2009;41:1510 –1530. 36 Myer GD, Ford KR, Hewett TE. Rationale and Clinical Techniques for Anterior Cruciate Ligament Injury Prevention Among Female Athletes. J Athl Train. 2004;39: 352–364. 37 Arokoski MH, Haara M, Helminen HJ, Arokoski JP. Physical function in men with and without hip osteoarthritis. Arch Phys Med Rehabil. 2004;85:574 –581. 38 Birrell F, Croft P, Cooper C, et al. Predicting radiographic hip osteoarthritis from range of movement. Rheumatology (Oxford). 2001;40:506 –512. 39 Steultjens MP, Dekker J, van Baar ME, et al. Range of joint motion and disability in patients with osteoarthritis of the knee or hip. Rheumatology (Oxford). 2000;39: 955–961. 40 Wernbom M, Augustsson J, Thomee´ R. The influence of frequency, intensity, volume and mode of strength training on whole muscle cross-sectional area in humans. Sports Med. 2007;37:225–264.
April 2010
41 Latham NK, Bennett DA, Stretton CM, Anderson CS. Systematic review of progressive resistance strength training in older adults. J Gerontol A Biol Sci Med Sci. 2004;59:48 – 61. 42 Jan MH, Lin JJ, Liau JJ, et al. Investigation of Clinical effects of high- and lowresistance training for patients with knee osteoarthritis: a randomized controlled trial. Phys Ther. 2008;88:427– 436. 43 Rhea M. A meta-analysis to determine the dose response for strength development. Med Sci Sports Exerc. 2003;35:456 – 464. 44 Peterson M. Maximizing strength development in athletes: a meta-analysis to determine the dose-response relationship. J Strength Cond Res. 2004;18:377–382. 45 Kraemer WJ, Adams K, Cafarelli E, et al; for the American College of Sports Medicine. American College of Sports Medicine position stand: progression models in resistance training for healthy adults. Med Sci Sports Exerc. 2002;34:364 –380. 46 Braith RW, Graves JE, Leggett SH, Pollock ML. Effect of training on the relationship between maximal and submaximal strength. Med Sci Sports Exerc. 1993;25:132–138. 47 Thomee´ R. A comprehensive treatment approach for patellofemoral pain syndrome in young women. Phys Ther. 1997;77: 1690 –1703. 48 Focht B, Ewing V, Gauvin L, Rejeski W. The unique and transient impact of acute exercise on pain perception in older, overweight, or obese adults with knee osteoarthritis. Ann Behav Med. 2002;24:201–210. 49 McDermott AY, Mernitz H. Exercise and older patients: prescribing guidelines. Am Fam Physician. 2006;74:437– 444. 50 American College of Sports Medicine position stand: exercise and physical activity for older adults. Med Sci Sports Exerc. 1998;30:992–1008. 51 Schaubert KL, Bohannon RW. Reliability and validity of three strength measures obtained from community-dwelling elderly persons. J Strength Cond Res. 2005;19: 717–720. 52 Bohannon RW, Andrews AW. Standards for judgments of unilateral impairments in muscle strength. Percept Mot Skills. 1999; 89(3 pt 1):878 – 880. 53 Ehrich EW, Davies GM, Watson DJ, et al. Minimal perceptible clinical improvement with the Western Ontario and McMaster Universities Osteoarthritis Index questionnaire and global assessments in patients with osteoarthritis. J Rheumatol. 2000; 27:2635–2641. 54 Angst F, Aeschlimann A, Stucki G. Smallest detectable and minimal clinically important differences of rehabilitation intervention with their implications for required sample sizes using WOMAC and SF-36 quality of life measurement instruments in patients with osteoarthritis of the lower extremities. Arthritis Rheum. 2001;45: 384 –391.
55 Redelmeier DA, Bayoumi AM, Goldstein RS, Guyatt GH. Interpreting small differences in functional status: the Six-Minute Walk Test in chronic lung disease patients. Am J Respir Crit Care Med. 1997;155: 1278 –1282. 56 Holm I, Bolstad B, Lutken T, et al. Reliability of goniometric measurements and visual estimates of hip ROM in patients with osteoarthrosis. Physiother Res Int. 2000; 5:241–248. 57 Bellamy N, Carette S, Ford PM, et al. Osteoarthritis antirheumatic drug trials, III: setting the delta for clinical trials—results of a consensus development (Delphi) exercise. J Rheumatol. 1992;19:451– 457. 58 Hayden J, van Tulder M, Tomlinson G. Systematic review: strategies for using exercise therapy to improve outcomes in chronic low back pain. Ann Intern Med. 2005;142:776 –785. 59 Linton S, Hellsing A, Bergstrom G. Exercise for workers with musculoskeletal pain: does enhancing compliance decrease pain? J Occup Health. 1996;6:177–189. 60 Zhang W, Doherty M, Arden N, et al; for the EULAR Standing Committee for International Clinical Studies Including Therapeutics (ESCISIT). EULAR evidence-based recommendations for the management of hip osteoarthritis: report of a task force of the EULAR Standing Committee for International Clinical Studies Including Therapeutics (ESCISIT). Ann Rheum Dis. 2005; 64:669 – 681. 61 Lively MW. Sports medicine approach to low back pain. South Med J. 2002;95: 642– 646. 62 Liddle SD, Baxter GD, Gracey JH. Exercise and chronic low back pain: what works? Pain. 2004;107:176 –190. 63 Loge JH, Kaasa S. Short-Form 36 (SF-36) health survey: normative data from the general Norwegian population. Scand J Public Health. 1998;26:250 –258. 64 Enright P, Sherrill D. Reference equations for the Six-Minute Walk in healthy adults. Am J Respir Crit Care Med. 1998;158: 1384 –1387. 65 Cink RE, Thomas TR. Validity of the Astrand-Ryhming nomogram for predicting maximal oxygen intake. Br J Sports Med. 1981;15:182–185. 66 Svenningsen S, Terjesen T, Auflem M, Berg V. Hip motion related to age and sex. Acta Orthop. 1989;60:97–100.
Volume 90
Number 4
Physical Therapy f
601
Perspective Conceptual Model for Physical Therapist Management of Chronic Fatigue Syndrome/ Myalgic Encephalomyelitis Todd E. Davenport, Staci R. Stevens, Mark J. VanNess, Christopher R. Snell, Tamara Little T.E. Davenport, PT, DPT, OCS, is Assistant Professor, Department of Physical Therapy, Thomas J. Long School of Pharmacy and Health Sciences, University of the Pacific, 3601 Pacific Ave, Stockton, CA 95211 (USA). Address all correspondence to Dr Davenport at:
[email protected]. S.R. Stevens, MA, is Executive Director, Pacific Fatigue Laboratory, Department of Sport Sciences, University of the Pacific. M.J. VanNess, PhD, is Associate Professor, Department of Sport Sciences, University of the Pacific. C.R. Snell, PhD, is Professor and Chair, Department of Sport Sciences, University of the Pacific.
Fatigue is one of the most common reasons why people consult health care providers. Chronic fatigue syndrome/myalgic encephalomyelitis (CFS/ME) is one cause of clinically debilitating fatigue. The underdiagnosis of CFS/ME, along with the spectrum of symptoms that represent multiple reasons for entry into physical therapy settings, places physical therapists in a unique position to identify this health condition and direct its appropriate management. The diagnosis and clinical correlates of CFS/ME are becoming better understood, although the optimal clinical management of this condition remains controversial. The 4 aims of this perspective article are: (1) to summarize the diagnosis of CFS/ME with the goal of promoting the optimal recognition of this condition by physical therapists; (2) to discuss aerobic system and cognitive deficits that may lead to the clinical presentation of CFS/ME; (3) to review the evidence for graded exercise with the goal of addressing limitations in body structures and functions, activity, and participation in people with CFS/ME; and (4) to present a conceptual model for the clinical management of CFS/ME by physical therapists.
T. Little, PT, EdD, DMT, FAAOMPT, is Associate Professor, Department of Physical Therapy, Thomas J. Long School of Pharmacy and Health Sciences, University of the Pacific. [Davenport TE, Stevens SR, VanNess MJ, et al. Conceptual model for physical therapist management of chronic fatigue syndrome/myalgic encephalomyelitis. Phys Ther. 2010; 90:602– 614.] © 2010 American Physical Therapy Association
Post a Rapid Response or find The Bottom Line: www.ptjournal.org 602
f
Physical Therapy
Volume 90
Number 4
April 2010
Management of Chronic Fatigue Syndrome/Myalgic Encephalomyelitis
F
atigue is a symptom common to many illnesses, such as cancer, depression, autoimmune diseases, hormonal disorders, and infections, and it is associated with poorer health outcomes and a high degree of recidivism in the general population.1,2 Most health conditions that cause fatigue, such as fatigue secondary to deconditioning, cancer, and neuromuscular disorders, have etiologies that are attributable to specific pathologies that may respond favorably to various forms of intervention, such as physical therapist management. However, some people may demonstrate fatigue related to causes that remain unclear.
Recent studies have begun to improve the collective understanding of chronic fatigue syndrome/myalgic encephalomyelitis (CFS/ME) as one cause of clinically debilitating fatigue. Chronic fatigue syndrome/myalgic encephalomyelitis affects 1 to 4 million adults in the United States,3 with women accounting for up to 75% of cases.4 Prevalence in the United States ranges from 230 to 420 per 100,000 adults,5 with the mean age at onset ranging from 29 to 35 years.4 Up to 85% of CFS/ME cases may be undiagnosed,3 and the actual prevalence of CFS/ME may be 6 to 10 times higher than presently understood,6 suggesting that a vast majority of CFS/ME cases are untreated and unaccounted for by epidemiological studies.7 The large number of cases that may remain undiagnosed places a premium on the recognition and management of CFS/ME and CFS/ME-like conditions by physical therapists. The purpose of this analysis was 4-fold. First, we summarize the diagnosis of CFS/ME with the goal of promoting the optimal recognition of this condition by physical therapists. Second, we discuss aerobic system and cognitive deficits that may lead to the clinical presentation of CFS/ April 2010
ME. Third, we review the evidence for graded exercise with the goal of addressing limitations in body structures and functions, activity, and participation in people with CFS/ME. We conclude by presenting a conceptual model, based on current scientific evidence, for the clinical management of CFS/ME by physical therapists.
to be most representative of the clinical features of the health condition by many clinicians and researchers,14 although the term myalgic encephalomyelitis continues to find common usage in the clinical community secondary to popular support from people with CFS/ME.15 Therefore, both terms were adopted for the purposes of this analysis.
Characteristic Clinical Findings of CFS/ME That Guide Diagnosis
Various attempts to characterize CFS/ME were made in the late 20th and early 21st centuries.13 The most current and common case definition was created when the Centers for Disease Control and Prevention convened an international working group in 1994. To meet the case definition criteria for CFS/ME, an individual must report persistent or relapsing, debilitating fatigue for which a preexisting illness or psychiatric comorbidity cannot be found as an explanation. According to the 1994 case definition of Fukuda and colleagues,16 to meet the criteria for CFS/ME, an individual must have persistent or relapsing fatigue for greater than 6 months (Fig. 1). The fatigue of CFS/ME may be characterized by either gradual or sudden onsets, and it may be progressive or relapsing and remitting during the course of the condition.
Chronic fatigue syndrome/myalgic encephalomyelitis is not a new health condition. The neurologists Beard8 and Goetz9 were among the first to characterize a health condition that they called neurasthenia in the latter half of the 19th century; neurasthenia was described as a combination of fatigue, anxiety, headache, impotence, and neuralgia. In the early 20th century, Gilliam10 documented an outbreak in Los Angeles, California, of a health condition that resembled poliomyelitis and that he called atypical poliomyelitis. Various outbreaks of health conditions resembling CFS/ME were recorded in the United States and elsewhere throughout the 20th century.11 Each of these outbreaks was described by its own region-specific and sometimes pejorative terminology. The term epidemic myalgic encephalomyelitis was coined as a result of a Royal Society of Medicine symposium in 1978.12 This development was notable because it represented the medical community’s first acknowledgement that CFS/ME was a distinct disease process rather than a behavioral disorder. Chronic fatigue syndrome/myalgic encephalomyelitis reached the popular consciousness in the United States after an outbreak in the Reno-Lake Tahoe, Nevada, region in the 1980s. Research involving this outbreak resulted in the name chronic fatigue syndrome.13 This term is considered
In addition to the specific criteria for fatigue, CFS/ME is characterized by a broad spectrum of nonspecific physical examination findings. Therefore, to meet the case definition criteria for the diagnosis of CFS/ME, an individual concurrently must exhibit
Available With This Article at ptjournal.apta.org • Audio Abstracts Podcast This article was published ahead of print on February 25, 2010, at ptjournal.apta.org.
Volume 90
Number 4
Physical Therapy f
603
Management of Chronic Fatigue Syndrome/Myalgic Encephalomyelitis
Figure 1. Conceptual model for clinical management of people with chronic fatigue syndrome/myalgic encephalomyelitis (CFS/ME). (A) The first presentation of symptoms in people with CFS/ME hypothetically occurs when the intensity of physical activity (solid line) exceeds the tolerance for physical activity (bars). (B) In response to symptoms, people with CFS/ME adapt their behaviors by either activity and symptom avoidance (dashed line) or activity and symptom fluctuation (dotted line). (C) The initiation of a pacing self-management program is hypothesized to stabilize activity levels within the tolerance for physical activity, such as that indicated by heart rate biofeedback below the anaerobic threshold. (D) After symptoms and function are stabilized, progression of anaerobic exercise to graded aerobic exercise may be used to increase tolerance for physical activity.
at least 4 additional symptoms, such as postexertion malaise (PEM) for at least 24 hours after exercise, impaired memory or concentration, nonrefreshing sleep, muscle pain, pain in multiple joints without signs of inflammation, headaches of a new type or severity, sore throat, and tender cervical or axillary lymph nodes. Secondary symptoms of joint pain and headaches may facilitate entry into physical therapy settings for people with CFS/ME. Perhaps the most prominent feature of CFS/ME is PEM, which is usually defined as a general feeling of discomfort or unease after even minimal physical activity.16 Indeed, among the various health conditions that are associated with fatigue, increased PEM that can alter daily activities for up to 2 weeks17 appears to be unique to CFS/ME.18,19
604
f
Physical Therapy
Volume 90
Other diagnostic systems that have been described in the literature are the Oxford criteria20 and the Canadian Consensus Document.21 The Canadian Consensus Document case definition for CFS/ME21 (Tab. 1) may be useful because it displays significantly greater power than the criteria established by Fukuda and colleagues16 to differentiate people with CFS/ME from people with fatigue related to psychiatric health conditions.22 Across all definitions, it is important to emphasize that the fatigue of CFS/ME must be unexplained by another somatic or psychiatric health condition, making CFS/ME a diagnosis of exclusion. Given the prominence of pain and fatigue in the symptomatology of CFS/ME,16,23 physical therapists may be among the first health care providers to recognize this health condition and direct appropriate management.
Number 4
Aerobic System and Cognitive Changes Contributing to the Clinical Presentation of CFS/ME A growing body of research confirms the presence of oxidative impairments in CFS/ME. Our research group24 and others25 have identified clinical evidence of oxidative metabolic impairments during graded exercise tests. VanNess and colleagues24 identified a range of oxidative impairments during graded exercise testing in participants meeting the case definition criteria of Fukuda and colleagues16 for CFS/ME. Although the maximum oxygen con˙ O2max) for participants sumption (V in that study24 ranged from 36% to ˙ O2max predicted for 76% of the V matched sedentary participants, in only half of the participants were
April 2010
Management of Chronic Fatigue Syndrome/Myalgic Encephalomyelitis Table 1. Case Definition Criteria for Identification of Chronic Fatigue Syndrome/Myalgic Encephalopathy Case Definition of Fukuda et al16 Requires the Presence of Profound Fatigue, Defined as Follows: • Symptom duration of ⱖ6 mo • Causes a substantial reduction in occupational, personal, social, or educational activities • Unrelieved with rest • Other medical or psychiatric conditions excluded
Also Requires 4 or More of the Following Features: • Postexertion malaise lasting ⬎24 h • Impaired memory or concentration • Nonrefreshing sleep • Muscle pain • Pain in multiple joints without signs of inflammation • Headaches of new type or severity • Sore throat • Tender cervical or axillary lymph nodes
American Medical Association guidelines for moderate to severe aerobic impairments met. This finding suggests that a spectrum of aerobic energy system impairments may be responsible for the reduced tolerance of physical activity observed in people with CFS/ME despite their common medical diagnosis. Evidence suggests that disruption of the aerobic energy system may be associated with a combination of genetic, cellular, and systemic metabolic deficits.26 –28 Investigators also have found evidence at the cellular level of mitochondrial dysfunction and impaired energy metabolism,29 oxidative damage to mitochondrial deoxyribonucleic acid,30 and poor recovery of adenosine triphosphate after exercise,31 which may be responsible for the observed deficits in repeated graded exercise test performances. These cellular and systemic impairments in the aerobic energy April 2010
Oxford Criteria,20 Requiring the Presence of Severe Disabling Fatigue as the Main Symptom and Further Defined as Follows: • Symptom duration of ⱖ6 mo • Affects both physical and mental functions • Symptoms are present ⱖ50% of the time • Definite onset and not lifelong • Other medical or psychiatric conditions excluded
Case Definition of Canadian Consensus Document21
Also Requires at Least 1 Symptom From 2 of the Following Categories:
Requires the Presence of the Following Features: • Symptom duration of ⱖ6 mo for adults and ⱖ3 mo for children • New onset of unexplained, persistent, or recurrent physical or mental fatigue that substantially reduces activity levels • Postexertion malaise • Other medical or psychiatric conditions excluded
system lead to a reduced functional capacity that limits an individual’s ability to sustain and repeat functional activities. Aerobic system impairments appear to be related to maladaptive sympathetic autonomic responses,32 perhaps in response to a triggering event, such as an injury or illness, in people with an apparent genetic predisposition.33 Over time, these maladaptive responses are suspected of causing dysregulation of the normal hypothalamus-pituitary axis34 and sympathetic autonomic responses32 and an overall reduction in tolerance for physical effort.35 Autonomic dysregulation is thought to be responsible for the orthostatic intolerance36 and abnormal heart rate (HR) responses to exercise37 exhibited by some people with CFS/ME. In addition to aerobic system impairments, people with CFS/ME exhibit centrally mediated disturbances in attention, perception, and affect.35
• Autonomic manifestations (eg, orthostatic intolerance, light-headedness, extreme pallor, nausea, irritable bowel syndrome, urinary frequency, bladder dysfunction, palpitations with or without cardiac arrhythmias, exertional dyspnea) • Neuroendocrine manifestations (eg, loss of thermostatic stability, intolerance of extremes of heat and cold, marked weight change, loss of adaptability, worsening of symptoms with stress) • Immune manifestations (eg, tender lymph nodes; recurrent sore throat; recurrent flulike symptoms; general malaise; new sensitivities to food, medications, or chemicals)
On average, people with CFS/ME rate their effort during physical tasks significantly higher than do people who are healthy38; this factor may lead to an overall decrease in maximal exertion. Paradoxically, however, anecdotal observations from various sources35,39 indicate that behavioral responses to symptom exacerbations in people with CFS/ME range from the maintenance of a sedentary lifestyle to abrupt increases in activity during periods of symptom remission that serve to exacerbate symptoms (Fig. 1). Patient-reported mental fatigue19 and maladaptive overactivity40 are cognitive and psychological correlates of morbidity in people with CFS/ME. Some authors have suggested that comorbid kinesiophobia41 or depressive symptoms42 also may be responsible for these disablements in CFS/ME; however, these findings are inconsistent,41 and functional deficits remain even after these variables are controlled for.43
Volume 90
Number 4
Physical Therapy f
605
Management of Chronic Fatigue Syndrome/Myalgic Encephalomyelitis Arguably, the cognitive effects of CFS/ME may serve as the greatest barrier to optimal research, diagnosis, and effective management of this health condition. Early attempts to characterize CFS/ME involved the assumption that it was primarily a psychiatric disorder.9 Our experience suggests that some clinicians still seem to view CFS/ME through this lens; this belief could be a source of stigma for people with CFS/ME.
Clinical Outcomes Associated With Graded Exercise in People With CFS/ME Graded aerobic exercise as an intervention for people with CFS/ME has been the focus of several studies, even though, as a general rule, deconditioning may not play a role in disablements secondary to CFS/ME.44 To assess the clinical effects of exercise in people with CFS/ME, we conducted a literature search with the search terms “exercise” AND “chronic fatigue syndrome” OR “myalgic encephalomyelitis” in the Cochrane Database of Systematic Reviews and the EMBASE, ERIC, MEDLINE, PEDro, Ovid Healthstar, Ovid Global Health, and PSYCHINFO databases. The initial search yielded 694 distinct citations. The abstracts of these citations were scanned for references to treatment responses in people with chronic fatigue, yielding 94 citations. These citations were then analyzed for observational, comparison, or randomized studies that reported on the effects of an exercise intervention in adults with CFS/ME according to existing diagnostic criteria, yielding 9 citations for consideration in this analysis. These studies are summarized in Table 2. Body structure and function deficits (eg, fatigue, muscle strength [forcegenerating capacity], cognitive processing, maximal and submaximal exercise test variables) and personal
606
f
Physical Therapy
Volume 90
factors (eg, mood) in people with CFS/ME appeared to improve consistently with graded aerobic exercise (Tab. 2). Fulcher and White45 found a significantly lower mean HR during treadmill testing in participants in a graded exercise group than in participants in a flexibility exercise group, as well as significantly lower ratings of perceived exertion (RPEs) in participants who received graded exercise than in participants who received flexibility exercise. Physiologic improvements were maintained at the 3-month follow-up in the graded exercise group. Moss-Morris et al46 reported significant improvements in physical, mental, and total fatigue in the exercise group compared with the control group. Pardaens and colleagues47 identified significant improvements in duration, peak power (ability to perform work over time), and peak respiratory exchange ratio, as well as a significant improvement in isokinetic hamstring muscle strength, in response to a graded exercise intervention combined with cognitive behavioral therapy. Wearden and colleagues48 found significantly greater functional work capacity in people who received graded exercise than in people in comparison groups, but fatigue ratings did not differ among the groups. Wallman et al42 noted significant improvements in resting HR, resting blood pressure, power, peak oxygen consumption, respiratory exchange ratio, and net blood lactate production in participants who received graded exercise compared with participants who received flexibility and relaxation exercises. Significant improvements also were documented in depression, anxiety, physical fatigue, and mental fatigue in participants who received graded exercise compared with participants who received flexibility and relaxation exercises. Modified Stroop word color identification test performance was significantly better in participants who received graded exercise. Despite the consistent im-
Number 4
provements in body structure and function deficits in response to graded exercise across studies, Pardaens and colleagues47 reported that these improvements were weakly correlated with or were not significantly correlated with functional improvements. Although there seemed to be a qualitative trend toward improvement, observations of qualitative and quantative improvements at the levels of activity and participation were more mixed (Tab. 2). Fulcher and White45 and Pardaens et al47 found that scores on subscales of the Medical Outcomes Study 36-Item Health Survey Questionnaire (SF-36) were significantly higher in participants who received graded exercise than in participants who received a comparison intervention. Powell and colleagues49 identified significantly higher SF-36 physical functioning subscale scores in the intervention groups than in the control group. At the 2-year followup50, all groups maintained significant improvements, and no significant differences persisted among the groups. Moss-Morris et al46 documented no significant differences in SF-36 physical functioning subscale scores among the groups, and Wallman et al42 did not identify significant changes in activity. It is notable that many studies conducted to examine the effectiveness of graded exercise had high dropout rates, suggesting the presence of a subpopulation for whom graded exercise interventions may be ineffective at the levels of body structure and function, activity, and participation. Additional research seems to be necessary to clarify the effects of different exercise volumes on activity and participation in people with CFS/ME. A major challenge of research and clinical practice related to exercise prescription for people with CFS/ME is the phenomenon that increased self-reported activity levels during intervention may be a direct result of the exercise program at the expense April 2010
Management of Chronic Fatigue Syndrome/Myalgic Encephalomyelitis Table 2. Studies of Comparative Effectiveness Involving Graded Aerobic Exercise in People With Chronic Fatigue Syndrome/Myalgic Encephalomyelitisa Intervention Program
Participants
Pardaens et al47,b
Study
Evaluated the effect of a 6-mo program of cognitive behavioral intervention (consisting of group discussions, relaxation and breathing exercises, and psychiatric and medical consultations, as needed) combined with graded exercise on exercise capacity measures. Intervention was provided in 4-h treatment sessions twice per week for the first month and once per week for the next 5 mo.
116 participants meeting the criteria proposed by Fukuda et al16 48% of the sample had an additional diagnosis of fibromyalgia
• • • • •
SF-36 Symptom Checklist-90 Causal Attribution List Self-Efficacy Scale Graded exercise testing on stationary bicycle with analysis of expired gases • Isokinetic dynamometry of the right and left quadriceps and hamstring muscles
Outcome Measures
• Significant improvements in all SF-36 subscale scores except emotional limitations • Significant improvement in Causal Attribution List nonphysical subscale score • Significant improvements in duration, peak power, and peak respiratory exchange ratio with graded exercise testing • Significant improvement in isokinetic hamstring muscle strength • V˙O2max was not correlated with or was weakly correlated with changes in quality-of-life measures
Main Study Findings
Fulcher and White45
Tested the effect of a 12-wk graded aerobic exercise program with a home exercise component provided in weekly sessions on global impression of change in status. The home exercises were to be completed at least 5 d/wk, with initial sessions of 5–15 min at an intensity of 40% of V˙O2max. The daily prescription was increased in consultation with the subject to maxima of 30 min/d and 60% of V˙O2max. Control intervention consisted of relaxation and flexibility exercises.
66 participants meeting the Oxford criteria20 (49 women)
• Individual global impression of change • Graded treadmill walking test with analysis of expired gases • Perceived exertion on exercise testing • Isometric quadriceps muscle testing • SF-36 • Visual analog scale for physical, mental, and total fatigue
• Significantly more participants in the exercise group (51%) than in the control group (27%) rated themselves as “much better” or “very much better” • Significantly lower mean heart rate during treadmill testing in the graded exercise group than in the flexibility exercise group • Significantly lower ratings of perceived exertion in the graded exercise group than in the flexibility exercise group • SF-36 total, physical functioning, and general health subscale scores were significantly higher in the graded exercise group than in the flexibility exercise group • Physiological improvements were maintained at 3-mo follow-up in the graded exercise group • Qualitatively, larger proportions of participants in the graded exercise group than in the flexibility exercise group rated themselves at 1-y follow-up as improved, were working or studying at least part time, and considered themselves regularly active
Moss-Morris et al46
Investigated the effectiveness and mechanisms of change attributed to graded exercise. Intervention group participated in 12 wk of graded walking, starting at 40%– 50% of V˙O2max, 4–5 times/wk. Specific intensity was set at a level unlikely to exacerbate symptoms, as determined in collaboration with the subject. Exercise duration was increased 3–5 min/wk, with final goals of 30 min and 70% of V˙O2max. Control group received standardized medical care and advice.
61 participants meeting the criteria proposed by Fukuda et al16
• Individual global impression of change • SF-36 • 14-item fatigue questionnaire • Graded treadmill exercise testing • Perceived exertion on exercise testing • Illness Perceptions Questionnaire–Revised • Illness Management Questionnaire
• Significantly more participants in the graded exercise group (48%) than in the control group (21%) rated themselves as “much better” or “very much better” at the end of 12 wk and at the 6-mo follow-up • Significant improvements in physical, mental, and total fatigue in the exercise group compared with the control group • No significant differences between groups in SF-36 physical functioning subscale scores (Continued)
April 2010
Volume 90
Number 4
Physical Therapy f
607
Management of Chronic Fatigue Syndrome/Myalgic Encephalomyelitis Table 2. Continued Study
Intervention Program
Participants
Outcome Measures
Main Study Findings
al49
Documented the effect of a psychoeducational intervention to encourage graded exercise at 1-y follow-up. Intervention groups received medical assessments and then evidence-based explanations of symptoms to encourage graded activity. Interventions included minimum intervention (2 in-person sessions), telephone intervention (minimum intervention ⫹ seven 30-min telephone contacts), and maximum intervention (minimum intervention ⫹ seven 1-h in-person contacts). Control group received standardized medical care, consisting of medical assessment, advice, and an information booklet that explained symptoms to encourage graded activity.
148 participants meeting the Oxford criteria20
• SF-36 • Hospital Anxiety and Depression (HAD) Scale • 4-item sleep problem questionnaire • Individual global impression of change
• Significantly more participants in the intervention groups (84%) than in the control group (12%) rated themselves as “much better” or “very much better” • Significantly higher SF-36 physical functioning subscale scores in the intervention groups than in the control group • Significantly lower fatigue scores in the intervention groups than in the control group
Powell et al50
Measured the effect of a psychoeducational intervention to encourage graded exercise at 2-y follow-up. Intervention groups received medical assessments and then evidence-based explanations of symptoms to encourage graded activity. Interventions included minimum intervention (2 in-person sessions), telephone intervention (minimum intervention ⫹ seven 30-min telephone contacts), and maximum intervention (minimum intervention ⫹ seven 1-h in-person contacts). Control group received standardized medical care, consisting of medical assessment, advice, and an information booklet that explained symptoms to encourage graded activity.
114 participants meeting the Oxford criteria and participating in the original trial of Powell et al49
• SF-36 • HAD Scale • 4-item sleep problem questionnaire • Individual global impression of change
• All groups maintained significant improvements compared with baseline measurements, and no significant differences in physical functioning and fatigue scores persisted among the groups • Significant improvements in physical functioning and fatigue scores over time in control group participants who were allowed to cross over into treatment groups after the conclusion of the first study period
Wallman et al42
Tested the effectiveness of a 12-wk program of graded exercise with pacing on physiological, psychological, and cognitive function. Intervention group participated in 5–15 min of exercise for large muscles; intensity was based on the mean heart rate during exercise testing. Participants were instructed to exercise every other day, unless they had a symptom relapse. In that case, subsequent exercise sessions were reduced to a length assessed by the participants to be manageable. Control group received relaxation and flexibility exercises.
61 participants meeting the criteria proposed by Fukuda et al16
• Submaximal bicycle exercise testing with analysis of expired gases and periodic blood sampling • Perceived exertion on exercise testing • Older Adult Exercise Status Inventory • 14-item fatigue rating scale • HAD Scale • Modified Stroop word color identification test
• Significant improvements in resting heart rate, resting blood pressure, power, V˙O2max, respiratory exchange ratio, and net blood lactate production in participants who received graded exercise compared with participants who received flexibility and relaxation exercises • Significant improvements in depression, anxiety, physical fatigue, and mental fatigue in participants who received graded exercise compared with participants who received flexibility and relaxation exercises
Powell et
(Continued)
608
f
Physical Therapy
Volume 90
Number 4
April 2010
Management of Chronic Fatigue Syndrome/Myalgic Encephalomyelitis Table 2. Continued Study
Intervention Program
Participants
Outcome Measures
Main Study Findings
Wallman et al (continued)
Wearden et al48
• Activity levels among groups were not significantly different • Significant improvements in modified Stroop word color identification test performance in participants who received graded exercise compared with participants who received flexibility and relaxation exercises Assessed the effect of a 26wk regimen of fluoxetine and graded exercise on fatigue, general health status, physical work capacity, and mood. Participants were randomized among 4 groups: fluoxetine ⫹ graded exercise, fluoxetine placebo ⫹ graded exercise, fluoxetine ⫹ graded exercise placebo, and fluoxetine placebo ⫹ graded exercise placebo. The fluoxetine dose was 20 mg. Participants who received fluoxetine placebo received an inert substance. Participants who received graded exercise were encouraged to exercise 3 times/wk for 20 min at an intensity at or exceeding 75% of V˙O2max. Participants who received graded exercise placebo were instructed to exercise when they felt capable but did not receive specific instructions.
n⫽136 meeting Oxford criteria20
• Graded exercise testing with analysis of expired gases • 14-item fatigue rating scale • Medical Outcomes Survey • HAD Scale
• Significant improvement in mood in participants who received fluoxetine at 12 wk • Functional work capacity was significantly increased at week 12 and week 26 in the exercise groups • Neither intervention was significantly associated with reduced fatigue ratings, although a qualitative trend toward improvement in the exercise groups was noted • Dropouts were significantly more frequent in the exercise groups than in the nonexercise groups
a All studies were controlled studies, unless otherwise noted. One observational study61 was excluded because the description of the exercise intervention was incomplete. Two controlled studies51,52 were excluded because the criteria for inclusion were unclear. SF-36⫽Medical Outcomes Study 36-Item Health Survey Questionnaire, V˙O2max⫽maximum oxygen consumption. b Observational study.
of actual increases in daily activities. For example, Black and colleagues51 studied the effectiveness of advice to increase physical activity on daily physical activity, fatigue, mood, and pain in participants who were diagnosed with CFS/ME by a physician and in matched sedentary control participants. Participants were requested to increase their physical activity during the first 2 weeks of the program in an incremental manner. Intervention included a 6-week walking program April 2010
consisting of an increase in daily walking time to a total of 30% more than pretreatment levels. The authors documented a significant increase in physical activity in participants with CFS/ ME by use of an accelerometer, leading to their conclusion that the intervention was a feasible and effective way to facilitate increased physical activity in this population. However, selfreported mood became significantly worse in participants with CFS/ME than in control participants over time.
Ratings of fatigue and pain, which were already significantly different between groups throughout the study, increased over time in participants with CFS/ME. A secondary analysis of the data revealed that participants with CFS/ME could actually maintain the increase in physical activity for the first 4 to 10 days of the study, exercising for a mean of 23 minutes.52 In contrast, during the final 3 weeks of the study, physical activity time fell to approximately 8 minutes per day. The
Volume 90
Number 4
Physical Therapy f
609
Management of Chronic Fatigue Syndrome/Myalgic Encephalomyelitis
Figure 2. Hypothetical relationship among duration of functional activity, metabolic energy pathways, and functional impairments in people with chronic fatigue syndrome/ myalgic encephalomyelitis (CFS/ME). The creatine phosphate-adenosine triphosphate immediate (teal line) and anaerobic short-term (red line) energy systems predominate during the first 2 minutes of activity (dashed black line). Activities longer than 2 minutes in duration are characterized by a rapid decrease in the contributions of these shortterm energy systems and an exponential increase in the contribution of the aerobic long-term energy system (blue line). It has been hypothesized that activities longer than 2 minutes in duration aggravate symptoms and functional deficits in people with CFS/ME because they have aerobic system impairments.
decrease in physical activity appeared to be coincident with the timing of the worsening mood and increased symptoms. It is possible that a similar trend has been observed in studies documenting significant increases in functioning in association with graded exercise in people with CFS/ME, but the use of quantitative measurements to document daily physical activity and secondary analyses of existing studies are uncommon.
Model for Clinical Management of CFS/ME The goal for the clinical management of CFS/ME should include an individ610
f
Physical Therapy
Volume 90
ual’s return to an optimal level of function. Disablement attributable to CFS/ME results from maladaptive physiologic responses that lead to overloaded body systems (Fig. 1). This assertion is consistent with the predictions of physical stress theory,53 which suggests that changes in relative stress in tissues result in predictable tissue adaptations ranging from decreased stress tolerance to maintenance to increased stress tolerance. Therefore, consistent with physical stress theory53 and other models for the clinical management of CFS/ME,35,39 optimal functioning for people with CFS/ME first de-
Number 4
pends on the achievement of a level of tissue stress that corresponds to symptom and functional maintenance. This objective will require the implementation of strategies that focus on pacing self-management to promote energy conservation and rest. After maintenance is achieved, physical stress must be provided in a manner that results in improved stress tolerance. As mentioned before, earlier models encouraged the use of graded exercise to achieve increased tissue stress tolerance.35,39 However, our experience has been that an approach beginning with therapeutic activities and exercises that avoid excessive use of the impaired aerobic system also is promising because it may mitigate the subsequent functional impairments associated with PEM. This approach involves shortduration exercises that are completed at intensities below an estimated anaerobic threshold (AT). After anaerobic activities are initiated without symptom exacerbation, progression to graded aerobic exercise can be undertaken. Finally, the chronic and episodic nature of this health condition4 requires physical therapists to ensure that people with CFS/ME demonstrate adequate development of the cognitive and physical skills necessary to facilitate longterm symptom self-management and optimal functioning. Pacing Self-Management: Criteria and Strategies Strategies to promote energy conservation appear to be important in people with CFS/ME. To date, recommendations for pacing self-management in people with CFS/ME have been made on the basis of symptom acuity and irritability.35,39 Although these criteria seem to be intuitive, they may fail to account for the rapid changes in function that are characteristic of CFS/ ME. An impaired perception of effort in people with CFS/ME may interfere with the optimal maintenance of symptom-free activity levels if pacing April 2010
Management of Chronic Fatigue Syndrome/Myalgic Encephalomyelitis self-management criteria that are based solely on symptomatology are used. Also, the ability to predict when activities might cause a flare-up of symptoms and functional deficits would be most helpful. The metabolic impairments observed in people with CFS/ ME suggest the need to limit the intensity of activity to avoid excessive use of the aerobic energy system (Fig. 2). Therefore, we suggest that AT may be a critical objective limit for ensuring that physical activity is maintained at an appropriate level to facilitate optimal pacing self-management. The use of criteria based on HR biofeedback for pacing self-management may be helpful in people with CFS/ ME.54 We assert that HR biofeedback should be used to ensure that activities are conducted at intensities below the AT. The gold standard for establishing the HR at the AT is a graded exercise test with periodic blood sampling. However, even in clinical settings in which graded exercise and blood sampling procedures are prohibitive in terms of cost, time, or expertise, the need to index activity pacing recommendations to quantitative data remains. Therefore, the HR at the AT may be ˙ O2max estimated indirectly from V measurements obtained during submaximal exercise testing; alternatively, the HR corresponding to RPEs of 13 to 15 during submaximal exercise testing may be used. Clinicians should bear in mind that although submaximal exercise testing demonstrates good discriminative validity for people with CFS/ME versus people without CFS/ME,55 such testing also demonstrates high within-subject variability.56 Frequent reassessment will be necessary to maintain confidence in the AT estimates obtained from submaximal testing. In untrained people, the AT has been approxi˙ O2max.57 Thus, we mated at 55% V suggest that clinicians may approximate the HR at the AT by calculating April 2010
˙ O2max as a 55% of the HR at the V starting point, although specific establishment of the HR at the AT in this population requires additional research. We suggest that people’s RPEs during submaximal exercise testing also may be used to approximate the threshold for pacing self-management, in which ratings below 13 to 15 represent activity at an intensity below the AT.57 Although exercise testing is an important clinical tool for guiding the clinical management of CFS/ ME, we emphasize that it should be conducted with extra care secondary to the risk for orthostatic impairments and the high likelihood of symptom exacerbation. Extra care includes adequate individual education regarding expected outcomes, with the goal of achieving appropriate consent for testing, as well as close symptom and physiologic monitoring to determine the need for test cessation or modification. In line with earlier models for the clinical management of CFS/ME,35,39 the importance of energy conservation in reducing symptom exacerbation and promoting optimal functioning in people with CFS/ME indicates that pacing self-management strategies should be provided in the context of a comprehensive psychoeducational program. The 5 A’s construct, originally proposed by the Canadian Task Force on Preventive Health Care,58 has been applied to individual education as part of clinical management programs for other chronic health conditions.59 The components of the 5 A’s construct58— assess, advise, agree, assist, and arrange—also may be applied to people with CFS/ME. This construct contains a learning needs assessment, which includes determining the presence of behaviors that may exacerbate symptoms as well as the individual’s preferred behavior change goals, methods, and constraints. An
individual’s level of knowledge related to CFS/ME and overall health literacy also should be ascertained. Advice for people with CFS/ME includes specific and personalized behavior change counseling based on the learning needs assessment and centered on maintaining physical exertion below the HR at the AT. People should be encouraged to wear an HR monitor for this purpose. Given the variability inherent in the use of many tests for establishing the HR at the AT, we suggest that a 10% margin below the estimated HR should be used as the critical threshold for pacing self-management to account for variability and ensure that an individual’s exertion remains below the AT. Therefore, the monitor’s alarm should be set to sound if the HR exceeds 10% below the HR at the AT. Frequent breaks including diaphragmatic breathing, alternate positions, and adaptive equipment should be prescribed to assist people in maintaining the target level of physical exertion. Further identification of activities that exacerbate symptoms and, as a result, necessitate attention may be aided by the use of activity logs, in which an individual with CFS/ME records the time of day, type and duration of activity, and symptoms for systematic analysis by the individual and the physical therapist. People with CFS/ME and their attending physical therapists should agree on the goals and methods of psychoeducational interventions for behaviors. The behavior change plan should be a collaborative effort between the individual and the attending physical therapist. People with CFS/ME may require assistance in implementing the behavior change plan; physical therapists may provide such assistance during both clinic visits and distance consultations. Finally, arrangement of follow-up contacts allows for additional assistance and alteration of the behavior change plan and reinforces the importance of behavior change to the individual.
Volume 90
Number 4
Physical Therapy f
611
Management of Chronic Fatigue Syndrome/Myalgic Encephalomyelitis Physical therapists are in a unique position to follow up with clinical brief counseling strategies because of their relatively frequent contact with people (compared with that of other health care providers). Exercise Interventions Exercise interventions for people with CFS/ME must be carefully customized to reflect the unique needs of each individual. The existing literature mentions 2 critical issues in prescribing physical activity for people with CFS/ME. First, clear communication between the individual and the physical therapist about the effects of the exercise program is critical to avoid the perception that physical activity has been increased because of increased physical capacity instead of the self-fulfilling prophecy associated with starting an exercise program. Second, aerobic system impairments associated with CFS/ME result in functional impairments that may not be amenable to training in people with CFS/ME compared with people who are sedentary. We assert that exercise interventions for people with CFS/ME require a combination of compensation and rehabilitation approaches to physical training in which training begins with activities that provide stress to the unimpaired anaerobic energy system before the impaired aerobic energy system is stressed. Therefore, we advocate a training approach in which initial therapeutic activities are short duration, low intensity, and directed toward specific contributing impairments in body structures and functions. Because oxidative phosphorylation serves as the primary metabolic pathway in activities lasting longer than 2 minutes (Fig. 2), aerobic system impairments in people with CS/ME would seem to limit activities longer than 2 minutes because of the risk of developing symptoms and functional deficits associated with PEM. Therefore, we recommend thera612
f
Physical Therapy
Volume 90
peutic activities that last less than 2 minutes and are conducted at an intensity consistent with an HR that is 10% below the HR at the AT or RPEs below 13 to 15. Previous studies demonstrated that reducing exercise time and intensity is effective in reducing symptoms of PEM in people with CFS/ME.60 These recommendations regarding duration and intensity are flexible; clinicians should be guided by the individual’s immediate and latent responses to therapeutic activities to determine appropriate exercise volume. We recommend that activities initially consist of stretching and activerange-of-motion (AROM) exercises to improve region-specific strength and flexibility, because deficits in strength and flexibility may be the source of increased energy expenditure through suboptimal movement mechanics. The specific exercises incorporated into the flexibility and AROM program depend on the clinician’s thorough examination and evaluation of potentially contributing pathomechanics. After participating in a stretching and AROM program that does not reproduce symptoms of PEM, people may advance to strength training in which the focus is on shortduration, low-intensity strengthening with maintenance of adequate rest intervals. Clinicians should use caution during the creation and progression of the resistance training program because the safety and effectiveness of these interventions in people with CFS/ME require additional research. Finally, people with CFS/ME may advance to short-duration, low-intensity interval training. As starting criteria, the duration of the intervals should not exceed 2 minutes, and the intensity should not exceed an HR that is 10% below the HR at the AT. Progression of interval training should involve increasing the number and intensity of intervals while maintaining a training range that prevents excessive use of the impaired aerobic sys-
Number 4
tem in people with CFS/ME. Interval training should involve functional retraining whenever possible, according to the physical therapist’s evaluation of the individual’s disablement. When short-duration interval training can be completed successfully, clinicians should consider initiating shortduration aerobic interval training, which can be advanced in an incremental manner according to people’s symptoms, as described elsewhere.39 Despite the importance of exercise to address physical conditioning in some people with CFS/ME, the healthrelated quality of life of people with CFS/ME is only weakly correlated with exercise capacity measurements. This fact underlines the importance of multimodal treatment, including individual education and pacing selfmanagement, to address the activity and participation limitations in people with CFS/ME.
Summary The prevalence of CFS/ME emphasizes the importance of the recognition and management of CFS/ME and CFS/ME-like conditions by physical therapists. The prominent features of CFS/ME include aerobic system impairment and centrally mediated disturbances in attention, perception, and affect. Pacing self-management criteria based on HR biofeedback may be helpful in ensuring that activities are conducted at an intensity below the AT in people with CFS/ME. The mode and intensity of exercise-based interventions for people with CFS/ME must be carefully customized to reflect the unique needs of each individual. Future studies should continue to clarify the roles of pacing self-management and exercise in the context of a comprehensive clinical management program for people with CFS/ME. All authors provided concept/idea/project design. Dr Davenport, Dr Snell, and Dr Little provided writing. Dr VanNess and Dr Snell provided data collection and analysis, fund
April 2010
Management of Chronic Fatigue Syndrome/Myalgic Encephalomyelitis procurement, and facilities/equipment. Dr Davenport and Dr Snell provided project management and institutional liaisons. Dr Davenport provided clerical support. Dr Davenport, Ms Stevens, Dr Snell, and Dr Little provided consultation (including review of manuscript before submission). This article was received February 16, 2009, and was accepted November 15, 2009. DOI: 10.2522/ptj.20090047
References 1 van der Windt DA, Dunn KM, SpiesDorgelo MN, et al. Impact of physical symptoms on perceived health in the community. J Psychosom Res. 2008;64:265– 274. 2 Nijrolder I, van der Horst H, van der Windt D. Prognosis of fatigue: a systematic review. J Psychosom Res. 2008;64:335–349. 3 Centers for Disease Control and Prevention. Chronic fatigue syndrome. Available at: http://www.cdc.gov/cfs/. Accessed December 12, 2008. 4 Cairns R, Hotopf M. A systematic review describing the prognosis of chronic fatigue syndrome. Occup Med (Lond). 2005; 55:20 –31. 5 Jason LA, Richman JA, Rademaker AW, et al. A community-based study of chronic fatigue syndrome. Arch Intern Med. 1999; 159:2129 –2137. 6 Reeves WC, Jones JF, Maloney E, et al. Prevalence of chronic fatigue syndrome in metropolitan, urban, and rural Georgia. Popul Health Metr. 2007;5:5. 7 Prins JB, van der Meer JW, Bleijenberg G. Chronic fatigue syndrome. Lancet. 2006; 367:346 –355. 8 Beard G. Neurasthenia, or nervous exhaustion. The Boston Medical and Surgical Journal. 1869;3:217–221. 9 Goetz CG. Poor Beard!! Charcot’s internationalization of neurasthenia, the “American disease.” Neurology. 2001;57: 510 –514. 10 Gilliam AG. Epidemiological Study on an Epidemic, Diagnosed as Poliomyelitis, Occurring Among the Personnel of Los Angeles County General Hospital During the Summer of 1934. Washington, DC: US Government Printing Office; 1934. 11 Ramsay AM. Postviral Fatigue Syndrome: The Saga of Royal Free Disease. New York, NY: Gower Medical Publishing; 1986. 12 Epidemic myalgic encephalomyelitis. Br Med J. 1978;1:1436 –1437. 13 Holmes GP, Kaplan JE, Gantz NM, et al. Chronic fatigue syndrome: a working case definition. Ann Intern Med. 1988;108: 387–389. 14 Evengard B, Schacterle RS, Komaroff AL. Chronic fatigue syndrome: new insights and old ignorance. J Intern Med. 1999; 246:455– 469.
April 2010
15 Jason LA, Eisele H, Taylor RR. Assessing attitudes toward new names for chronic fatigue syndrome. Eval Health Prof. 2001; 24:424 – 435. 16 Fukuda K, Straus SE, Hickie I, et al. The chronic fatigue syndrome: a comprehensive approach to its definition and study. International Chronic Fatigue Syndrome Study Group. Ann Intern Med. 1994;121: 953–959. 17 Sisto SA, Tapp WN, LaManca JJ, et al. Physical activity before and after exercise in women with chronic fatigue syndrome. QJM. 1998;91:465– 473. 18 Cook DB, Nagelkirk PR, Peckerman A, et al. Exercise and cognitive performance in chronic fatigue syndrome. Med Sci Sports Exerc. 2005;37:1460 –1467. 19 Sorensen B, Streib JE, Strand M, et al. Complement activation in a model of chronic fatigue syndrome. J Allergy Clin Immunol. 2003;112:397– 403. 20 Sharpe MC, Archard LC, Banatvala JE, et al. A report— chronic fatigue syndrome: guidelines for research. J R Soc Med. 1991; 84:118 –121. 21 Carruthers BM, Jain AK, DeMeirleir KL, et al. Myalgic encephalomyelitis/chronic fatigue syndrome: clinical working case definition, diagnostic and treatment protocols (a consensus document). Journal of Chronic Fatigue Syndrome. 2003;11:7–115. 22 Jason LA, Torres-Harding SR, Jurgens A, Helgerson J. Comparing the Fukuda et al. criteria and the Canadian case definition for chronic fatigue syndrome. Journal of Chronic Fatigue Syndrome. 2004;12:37–52. 23 Meeus M, Nijs J, Meirleir KD. Chronic musculoskeletal pain in individuals with the chronic fatigue syndrome: a systematic review. Eur J Pain. 2007;11:377–386. 24 VanNess JM, Snell CR, Strayer DR, et al. Subclassifying chronic fatigue syndrome through exercise testing. Med Sci Sports Exerc. 2003;35:908 –913. 25 De Becker P, Roeykens J, Reynders M, et al. Exercise capacity in chronic fatigue syndrome. Arch Intern Med. 2000;160: 3270 –3277. 26 Bains W. Treating chronic fatigue states as a disease of the regulation of energy metabolism. Med Hypotheses. 2008;71:481– 488. 27 Sorensen B, Jones JF, Vernon SD, Rajeevan MS. Transcriptional control of complement activation in an exercise model of chronic fatigue syndrome. Mol Med. 2009; 15:34 – 42. 28 Whistler T, Jones JF, Unger ER, Vernon SD. Exercise responsive genes measured in peripheral blood of women with chronic fatigue syndrome and matched control subjects. BMC Physiol. 2005;5:5. 29 Whistler T, Taylor R, Craddock RC, et al. Gene expression correlates of unexplained fatigue. Pharmacogenomics. 2006; 7:395– 405. 30 Fulle S, Mecocci P, Fano G, et al. Specific oxidative alterations in vastus lateralis muscle of individuals with the diagnosis of chronic fatigue syndrome. Free Radic Biol Med. 2000;29:1252–1259.
31 Wong R, Lopaschuk G, Zhu G, et al. Skeletal muscle metabolism in the chronic fatigue syndrome: in vivo assessment by 31P nuclear magnetic resonance spectroscopy. Chest. 1992;102:1716 –1722. 32 Cordero DL, Sisto SA, Tapp WN, et al. Decreased vagal power during treadmill walking in individuals with chronic fatigue syndrome. Clin Auton Res. 1996;6: 329 –333. 33 Fang H, Xie Q, Boneva R, et al. Gene expression profile exploration of a large dataset on chronic fatigue syndrome. Pharmacogenomics. 2006;7:429 – 440. 34 Van Den Eede F, Moorkens G, Van Houdenhove B, et al. Hypothalamic-pituitary-adrenal axis function in chronic fatigue syndrome. Neuropsychobiology. 2007;55:112–120. 35 Van Houdenhove B, Verheyen L, Pardaens K, et al. Rehabilitation of decreased motor performance in individuals with chronic fatigue syndrome: should we treat low effort capacity or reduced effort tolerance? Clin Rehabil. 2007;21:1121–1142. 36 Naschitz JE, Rosner I, Rozenbaum M, et al. The head-up tilt test with haemodynamic instability score in diagnosing chronic fatigue syndrome. QJM. 2003;96:133–142. 37 Broderick G, Craddock RC, Whistler T, et al. Identifying illness parameters in fatiguing syndromes using classical projection methods. Pharmacogenomics. 2006; 7:407– 419. 38 Wallman KE, Sacco P. Sense of effort during a fatiguing exercise protocol in chronic fatigue syndrome. Res Sports Med. 2007; 15:47–59. 39 Nijs J, Paul L, Wallman K. Chronic fatigue syndrome: an approach combining selfmanagement with graded exercise to avoid exacerbations. J Rehabil Med. 2008; 40:241–247. 40 Van Houdenhove B, Bruyninckx K, Luyten P. In search of a new balance: can high “action-proneness” in individuals with chronic fatigue syndrome be changed by a multidisciplinary group treatment? J Psychosom Res. 2006;60:623– 625. 41 Nijs J, De Meirleir K, Duquet W. Kinesiophobia in chronic fatigue syndrome: assessment and associations with disability. Arch Phys Med Rehabil. 2004;85:1586 – 1592. 42 Wallman KE, Morton AR, Goodman C, et al. Randomised controlled trial of graded exercise in chronic fatigue syndrome. Med J Aust. 2004;180:444 – 448. 43 Blackwood SK, MacHale SM, Power MJ, et al. Effects of exercise on cognitive and motor function in chronic fatigue syndrome and depression. J Neurol Neurosurg Psychiatry. 1998;65:541–546. 44 Bazelmans E, Bleijenberg G, Van Der Meer JW, Folgering H. Is physical deconditioning a perpetuating factor in chronic fatigue syndrome? A controlled study on maximal exercise performance and relations with fatigue, impairment and physical activity. Psychol Med. 2001;31:107–114. 45 Fulcher KY, White PD. Randomised controlled trial of graded exercise in individuals with the chronic fatigue syndrome. BMJ. 1997;314:1647–1652.
Volume 90
Number 4
Physical Therapy f
613
Management of Chronic Fatigue Syndrome/Myalgic Encephalomyelitis 46 Moss-Morris R, Sharon C, Tobin R, Baldi JC. A randomized controlled graded exercise trial for chronic fatigue syndrome: outcomes and mechanisms of change. J Health Psychol. 2005;10:245–259. 47 Pardaens K, Haagdorens L, Van Wambeke P, et al. How relevant are exercise capacity measures for evaluating treatment effects in chronic fatigue syndrome? Results from a prospective, multidisciplinary outcome study. Clin Rehabil. 2006;20:56 – 66. 48 Wearden AJ, Morriss RK, Mullis R, et al. Randomised, double-blind, placebo-controlled treatment trial of fluoxetine and graded exercise for chronic fatigue syndrome. Br J Psychiatry. 1998;172:485– 490. 49 Powell P, Bentall RP, Nye FJ, Edwards RH. Randomised controlled trial of individual education to encourage graded exercise in chronic fatigue syndrome. BMJ. 2001;322: 387–390. 50 Powell P, Bentall RP, Nye FJ, Edwards RH. Individual education to encourage graded exercise in chronic fatigue syndrome: 2-year follow-up of randomised controlled trial. Br J Psychiatry. 2004;184:142–146.
614
f
Physical Therapy
Volume 90
51 Black CD, O’Connor PJ, McCully KK. Increased daily physical activity and fatigue symptoms in chronic fatigue syndrome. Dyn Med. 2005;4:3. 52 Black CD, McCully KK. Time course of exercise induced alterations in daily activity in chronic fatigue syndrome. Dyn Med. 2005;4:10. 53 Mueller MJ, Maluf KS. Tissue adaptation to physical stress: a proposed “physical stress theory” to guide physical therapist practice, education, and research. Phys Ther. 2002;82:383– 403. 54 Nijs J, Almond F, De Becker P, et al. Can exercise limits prevent post-exertional malaise in chronic fatigue syndrome? An uncontrolled clinical trial. Clin Rehabil. 2008;22:426 – 435. 55 Wallman KE, Morton AR, Goodman C, Grove R. Physiological responses during a submaximal cycle test in chronic fatigue syndrome. Med Sci Sports Exerc. 2004;36: 1682–1688. 56 Nijs J, Demol S, Wallman K. Can submaximal exercise variables predict peak exercise performance in women with chronic fatigue syndrome? Arch Med Res. 2007;38: 350 –353.
Number 4
57 Demello JJ, Cureton KJ, Boineau RE, Singh MM. Ratings of perceived exertion at the lactate threshold in trained and untrained men and women. Med Sci Sports Exerc. 1987;19:354 –362. 58 Whitlock EP, Orleans CT, Pender N, Allan J. Evaluating primary care behavioral counseling interventions: an evidence-based approach. Am J Prev Med. 2002;22:267–284. 59 Davenport TE, Kulig K, Matharu Y, Blanco CE. The EdUReP model for nonsurgical management of tendinopathy. Phys Ther. 2005;85:1093–1103. 60 Clapp LL, Richardson MT, Smith JF, et al. Acute effects of thirty minutes of lightintensity, intermittent exercise on individuals with chronic fatigue syndrome. Phys Ther. 1999;79:749 –756. 61 Sadlier M, Evans JR, Phillips C, Broad A. A preliminary study into the effectiveness of multi-convergent therapy in the treatment of heterogeneous individuals with chronic fatigue syndrome. Journal of Chronic Fatigue Syndrome. 2000;7:93–101.
April 2010
CARE V Conference Series
Qualitative Research Ethics: Enhancing Evidence-Based Practice in Physical Therapy Anne Townsend, Susan M. Cox, Linda C. Li
Background. Increasing challenges to health care systems and the prominence of patient-centered care and evidence-based practice have fostered the application of qualitative approaches in health care settings, prompting discussions of associated ethical issues in a range of disciplines. Objectives. The purposes of this work were to identify and describe the application and value of qualitative health research for physical therapy and to identify ethical considerations in a qualitative research study.
Design. This was a qualitative interview study with telephone follow-ups. Methods. Forty-six participants were interviewed about their early experiences with rheumatoid arthritis. They also were asked what motivated them to volunteer for the study. To inform the discussion of ethics in qualitative health research, this study drew on the in-depth interviews, took a descriptive approach to the data, and applied the traditional ethical principles of autonomy, justice, and beneficence to the study process.
Results. Ethical issues emerged in this qualitative health research study that were both similar to and different from those that exist in a positivist paradigm (eg, clinical research). With flexibility and latitude, the traditional principle approach can be applied usefully to qualitative health research.
Conclusions. These findings build on previous research and discussion in physical therapy and other disciplines that urge a flexible approach to qualitative research ethics and recognize that ethics are embedded in an unfolding research process involving the role of the subjective researcher and an active participant. We suggest reflexivity as a way to recognize ethical moments throughout qualitative research and to help build methodological and ethical rigor in research relevant to physical therapist practice.
A. Townsend, PhD, is Research Associate, The W. Maurice Young Centre for Applied Ethics, University of British Columbia, 235-6356 Agricultural Rd, Klinck Building, Vancouver, British Columbia, Canada V6T 1Z2, and Affiliate Researcher, Arthritis Research Centre of Canada, Vancouver, British Columbia, Canada. Address all correspondence to Dr Townsend at:
[email protected]. S.M. Cox, PhD, is Assistant Professor, The W. Maurice Young Centre for Applied Ethics, University of British Columbia. L.C. Li, PT, PhD, is Assistant Professor and Harold Robinson/ Arthritis Society Chair in Arthritic Diseases, Department of Physical Therapy, University of British Columbia, and Research Scientist, Arthritis Research Centre of Canada, Vancouver, British Columbia, Canada. [Townsend A, Cox SM, Li LC. Qualitative research ethics: enhancing evidence-based practice in physical therapy. Phys Ther. 2010;90:615– 628.] © 2010 American Physical Therapy Association
Post a Rapid Response or find The Bottom Line: www.ptjournal.org April 2010
Volume 90
Number 4
Physical Therapy f
615
Qualitative Research Ethics
I
ncreasing challenges to health care systems and the prominence of patient-centered care and evidencebased practice (EBP) have fostered the application of qualitative approaches in health and illness concepts,1–12 prompting discussions of associated ethical issues in a range of disciplines (eg, medicine,1 ethics,13–15 social science,16 health care17,18). Although there is no unified definition or agreedupon way of doing qualitative research, it has been noted that the qualitative approach has salience for physical therapy, given its efforts toward EBP,19 a patient-centered approach, and the call to focus on the ethics of care.20 According to Jensen, “Qualitative methods provide researchers with the tools to examine social settings and human behavior. The methods are well suited to studying the complex, multidimensional environments present in physical therapy practice and education.”21(p492) This article contributes to the discussion on the utility and value of qualitative health research and associated ethical concerns. Drawing on a qualitative interview study, we apply the traditional ethical principles of autonomy, justice, and beneficence22 to the qualitative health research process and outline: (1) the development of medical research and governance; (2) the application and value of qualitative health research; (3) research strategies in a qualitative health research study; and (4) emerging ethical consider-
Available With This Article at ptjournal.apta.org • Discussion Podcast: Participants to be determined. • Audio Abstracts Podcast This article was published ahead of print on March 4, 2010, at ptjournal.apta.org.
616
f
Physical Therapy
Volume 90
ations in the qualitative research process. In our discussion, we suggest the process of reflexivity as a way to foster ethical and methodological rigor in applied qualitative health research and, ultimately, to offer enhanced care. The concept of reflexivity has been used in a variety of ways in a range of disciplines; for the purposes of this article, it is a researcher self-awareness that frames actions and interactions during the research process. Being reflexive engenders attending to participant priorities, respecting participant experiences in the context of their daily lives, and building relationships based on mutual respect and shared information in the health research process.
Background: Medical Research and Governance Health research is inherently a moral enterprise, characterized by asymmetrical relationships of trust and power, underpinned by ethical tensions between means (eg, potential for risks posed to volunteers in the research process) and ends (the quest for knowledge for the greater good).23 Guiding moral principles in research ethics consider the overall potential benefits accrued against the possible harms to volunteers, and formalized structures of research governance exist to ensure that research is conducted in an ethical manner and that researchers act with integrity.24 The advancement of medical knowledge and associated research scandals involving abuses of power by health care professionals and gross harm to research participants25 spawned the development of systematic ethical guidance in medical research.26 The protective measures introduced rested on basic and shared moral principles. Research volunteers were to be fully informed of the nature of the research, decide freely about participation, and be assured that their participation would not affect their health care and treat-
Number 4
ment. Regulatory safeguards and guidelines were declared in the Nuremberg code (1949),27 the World Medical Association Declaration of Helsinki (1964; last revised in 2008)28 and the Belmont Report (1979).29 Increased federal accountability for human subjects research is further illustrated in updated ethics guidelines, such as the Council for International Organizations of Medical Sciences⫺World Health Organization’s International Ethical Guidelines for Biomedical Research Involving Human Subjects (1993; updated 2002),30 the Canadian Tri-Council Policy Statement (1998; amendments in 2000, 2002, and 2005),31 and the Statement of Ethical Practice for the British Sociological Association (2002; appendix updated 2004).32 In the United States, the Office for Human Research Protections details research regulation, offers education (is currently running a campaign to inform the general public about research participation), and recently updated its document on compliance oversight.33 Historically, guidelines for human subject protection are anchored in the biomedical model,1 with clinical trials acting as the benchmark of research governance.34 Broadly, the biomedical model rests in the positivist paradigm, which typically (although not always) generates quantitative data. The positivist approach is associated with a hypothetico-deductive model of science—a systematic process in which observable “facts” are collected, variables are scrutinized, hypotheses are tested, reliability and validity are measured, and statistical generalizations are made (Fig. 1). The researcher objectively observes the data and collects the results. In contrast, qualitative research is associated with the interpretivist paradigm, an inductive approach to studying naturally occurring phenomena and understanding multiple April 2010
Qualitative Research Ethics
Figure 1. Model illustrating examples of positivist and interpretivist approaches to research. Based on: Shepard KF, Jensen GM, School BJ, et al. Alternative approaches to research in physical therapy: positivism and phenomenology. Phys Ther. 1993:73:88 –97.
realities; it applies different criteria to validity and reliability, and an assessment is made of how transferable the findings are to different settings (Fig. 1). Qualitative researchers attempt to interpret the meaning people attach to their experiences and investigate the complexity, context, and process of “lived experience.” Generally, ethnography, although its definition is debated, is a favored interpretive approach that typically involves participant observation with interviews. Often researchers are unable to immerse themselves in the lives of those beApril 2010
ing studied, but conduct a series of interviews (often termed “ethnographic”) in an attempt to gain an in-depth understanding of individual experience. The researcher coconstructs the data generated, acting as the research instrument. Samples can range from single cases to largescale studies, depending on the research aim.35 Quantitative and qualitative approaches typically ask different questions and gain different types of evidence, all of which are needed to fully inform an evidence base.20 It should be noted that positivist and interpretivist approaches
are not always tied to particular methods and that researchers often are involved in mixed-methods initiatives and make decisions about research design based on pragmatic factors rather than philosophical preferences.36 Although qualitative research is increasingly addressed in research compliance guidelines for the protection of human subjects,31 the principle-based approach to conducting ethical research is commonly cited as the most appropriate framework for ensuring human sub-
Volume 90
Number 4
Physical Therapy f
617
Qualitative Research Ethics ject protection: “Respect for autonomy, beneficence and nonmaleficence has become a key component of any discussion of the researcher-researched relationship, which, it is argued, will ensure that the end objective in qualitative research does nor override the rights, health, well-being and care of research participants.”37(pp1151–1152) However, evidence indicates that medical research ethics committees encounter difficulties when assessing qualitative research, which subsequently has hampered research of sensitive topics.1,38
The Application and Value of Qualitative Health Research The increasing relevance of qualitative research to health care practice has been documented.3,35 Typically designed to reveal a range of experiences and identify commonalities and differences between groups or individuals, methodological approaches include grounded theory, phenomenology, and narrative. Investigation reveals how outcomes are achieved and situations unfold and highlights the interactions and minutiae of daily life. Qualitative approaches have increased our understanding of medication use (eg, revealing patient ambivalence and the shortcomings of the compliance model9) and highlighted contradictions and tensions in practicing selfmanagement not identified by a variable-based approach.8 Qualitative studies have offered explanations for unexpected or contradictory findings revealed by quantitative studies and shortcomings of a rational-choice decision-making model in seeking care and have identified unintended consequences of service developments.35 In physical therapy, grounded theory studies20,39,40 have revealed the complex ways in which health care practitioner experts make decisions. Such
618
f
Physical Therapy
Volume 90
studies contribute knowledge to an evidence base that cannot be generated by a quantitative approach.41 Overall, qualitative research can offer useful stand-alone projects, evidence alongside quantitative methods, or a stage of a mixed methods project and can contribute to a body of work for synthesis.42 Here we describe an in-depth interview study and then focus on ethical issues that arose throughout the study process. We use the traditional principles of autonomy, justice, and beneficence/nonmaleficence as an organizing framework22,34 and comment on their utility in qualitative research.
Research Strategies in a Qualitative Health Research Study: The Early Rheumatoid Arthritis HelpSeeking Experience (ERAHSE) Project Our research aim was to gain an indepth understanding of the meanings people applied to early symptoms of rheumatoid arthritis (RA) and their impact and the actions individuals took in the context of their daily lives. We wanted to investigate the “lived experience” of illness.4 We draw on interview extracts and detailed field notes (from both the pilot study and the main study) to explore and illustrate the links between method and ethics. Because this is a discussion article that draws on data, we use basic description,43 remaining close to the data, and do not make highly conceptual or abstract statements. As Sandelowski noted, qualitative description is a valuable method, but often neglected as a stand-alone approach to presenting qualitative data.43 We compare the quantitative and qualitative paradigms (Fig. 1) and identify the ways in which traditional ethical principles can be applied (Fig. 2),
Number 4
focusing on the qualitative research context.44 Recruitment We recruited individuals who had been diagnosed with RA in the 12 months prior to their recruitment and who were English speaking and lived in British Columbia, Canada. In our pilot study,45 recruitment packages were mailed to 163 family physicians and 4 rheumatologists for forwarding to patients who fulfilled the inclusion criteria. Out of a target of 10 participants, we recruited 8 participants (4 recruited through family physicians and 4 recruited through rheumatologists). Follow-up contact was attempted with the health care professionals, but no further participants were recruited. This paucity of participants reflects the low incidence of new cases of RA (33 per 1,000 in North America), but also reveals problems, common to both qualitative and quantitative research,46,47 with recruiting participants through health care professionals.48 –50 In most cases, we were unable to speak to family physicians to discuss recruitment but spoke to assistants, several of whom reported physicians did not take part in this type of research. The pilot recruitment strategy was considered adequate for the pilot phase (we gained in-depth data that warranted further investigation) but insufficient for the main study, in which we wanted to gain a range of experiences based on diverse social and illness factors in a range of settings. We also wanted to identify similarities and differences among individuals. For the main study, we recruited participants through patient organization Web sites, newsletters, and information leaflets at local arthritis centers, as well as 4 rheumatologists’ offices. We gained 38 participants (37 female and 1 male). Although we originally had sought 36 participants, we extended recruitment (and April 2010
Qualitative Research Ethics
Figure 2. Model illustrating theoretical approaches in research and associated ethical issues.
adapted information leaflets) in the hope of gaining more male participants; however, this attempt was unsuccessful, and recruitment ended for practical reasons. Although in the main study we provided more opportunity to “selfselect” via patient organizations and arthritis centers, we encountered recruitment problems. Despite requiring an equal number of male and female participants to gain insight into gendered experiences, recruiting only 1 male participant limited the findings. Also, the majority of participants who contacted us did so April 2010
via the recruitment leaflets, not their health care professionals. Participants indicated (in their initial contact and during the interview) that a major reason for taking part in the study was the hope that they would gain treatment or care benefits or that they could share their story to assist others. This participant selfselection may have biased the sample. It also was possible that those volunteers with few treatment or illness problems saw no reason to participate and that those who were among the most disadvantaged, with multiple conditions, may not have had access to the recruitment infor-
mation or could have had other priorities. This possibility does not dilute the significance of the findings, but it does highlight that caution and clarity are needed when making data-driven claims, for instance, about transferability of the findings to other settings. Interviews The guide was organized around 3 broad, overlapping areas: (1) early symptoms, including impact and illness actions; (2) interactions with health care professionals and gaining a diagnosis; and (3) post-diagnosis experiences. We also asked individ-
Volume 90
Number 4
Physical Therapy f
619
Qualitative Research Ethics Box 1. Autonomy Quote 1: I heard about this study from my doctor. He told me about it when I went to see him. He showed me the information and said I should ring you. He said he suggested me because I was more alert than the others. (Iris) Quote 2: My doctor told me about the study. He said I should be able to give you all sorts of information; he knows I do my own research, and thought I would be good in the study. (Ian) Quote 3: My doctor told me about it [the study]. Well, we have been trying all these things, and nothing works. He thought it might help. He thought I might learn something, so I got in touch with you. Anything would help. (Lynn)
uals their motivations for participating in the research. Informed consent was obtained, and all participants agreed to a follow-up telephone interview for elaboration and clarification and for the interviews to be audio-recorded. Detailed field notes were taken. The interviews were transcribed verbatim, and identifying information was removed from the transcripts. Pseudonyms, chosen by the participants, have been used for all interview data. The structure, content, and form of the guide were designed to elicit open and detailed responses from the study participants, giving them opportunities to discuss their priorities. To avoid, as far as possible, researcher bias and medical model preconceived ideas about treatments and decision making, consumer collaborators (volunteers on the research team with a diagnosis of arthritis) contributed to the interview guide, offering their perspective on what topics should be included, the type of language or phrasing to be used, and the order of the questions asked. Thus, the language was anchored in the everyday world rather than in a medical, therapeutic, or theoretical paradigm. The interview was designed to help build rapport (mutual trust and emotional affinity) and aid validity (gain an in-depth account as close as possible to participants’ experiences). Practical issues included arranging the interview for the convenience of the participant in terms of venue, timing, and comfort (eg, stretch breaks were offered).
620
f
Physical Therapy
Volume 90
Knowledge Translation We define knowledge translation as an exchange of knowledge and perspective among an interdisciplinary team (including practitioners and consumers), as well as strategies for dissemination to a range of stakeholders and decision-makers in applied settings. We also include the transfer of information between the research team and participants. Communication strategies were built into the research design to facilitate team discussions and negotiate different perspectives and their application in answering the research question. Patient groups and health practitioners contributed key perspectives to the interview guide and dissemination process, which was designed to offer a range of relevant data formats and outlets encouraging best use of the collective findings (peer-reviewed publications; specialized and plain language/information documents, conferences, roundtables, workshops, education initiatives, and updates and results regularly published on arthritis Web sites). Dissemination activities continue to be aimed at different stakeholders to build a bank of knowledge leading to actionable research/ action. Regular (quarterly) progress reports were sent out to all participants and the research team, updating them on all dissemination activities. The analysis to date has informed an interactive educational initiative called the Animated Self-serve Webbased Research Tool (ANSWER),
Number 4
which is near completion, to enhance the patient experience, underpinned by a shared decision-making approach and recognition of the role of partnership in health care relationships. This initiative operationalizes the concepts of integrated knowledge translation and end-of-project knowledge translation.51 The findings also have led to a longitudinal help-seeking study in another health care setting (the United States) and a Canada-wide survey to suggest hypotheses about help-seeking to add to the knowledge base.
Emerging Ethical Considerations in the Qualitative Research Process We apply the concepts of autonomy, justice, and beneficence/nonmaleficence to discuss ethical concerns that emerged in our study. We use this principle-based approach because it typically is used as a framework in assessing human protection in biomedical research. However, it increasingly is regarded as unsuitable in assessing ethical issues that arise in a qualitative paradigm1 (Fig. 2). Some claim it is insufficient and neglects the context of research,37 whereas others defend its utility if the principles are used flexibly and context is addressed.52 Autonomy Autonomy has been defined as the capacity to think, decide, and act on the basis of a freely made decision.52 Beauchamp and Childress22 identified 2 fundamental components of April 2010
Qualitative Research Ethics autonomy: (1) liberty (independence from controlling influences) and (2) agency (capacity for intentional action). Hewitt stated, “In the context of research, informed consent is an explicit agreement by participants to participate in the research after receiving and comprehending information regarding the nature of the research. Such consent is given without threat or inducement and requires that participants have the mental capacity to give consent and voluntariness”37(p1152) From a qualitative perspective, informed consent typically is considered an ongoing process,15 of which recruitment is a part. In our pilot study, participants were recruited through the offices of their health care professionals, provoking potential ethical concerns regarding decision making and research participation. Our study was designed to ensure, as far as possible, that health care professionals had no personal contact with patients and that participation remained confidential. Communications between the researcher and health care professionals were limited to mailing recruitment documents and associated general followups and inquiries. Recruitment documents were to be mailed to relevant patients from health care professionals’ offices. However, in the pilot study, several participants reported interpersonal contact with their physician regarding participation (Box 1, quotes 1, 2, and 3). It has been noted that potential issues of inducement and consent emerge in the health care setting if a study participant is recruited through his or her physician.49 Issues of vulnerability and power may influence decisions to participate, and being recruited by an individual’s health care professional may pose obstacles to the process of free and informed consent.
Typically, participants in our pilot study described their physicians as pivotal resources and, in some cases, as allies as they faced early-stage RA and accessed a range of medical services and information at a particularly vulnerable time in their illness trajectory. This relationship has a potentially coercive influence on study participation. Study information from a trusted health care professional (particularly when gained in the consultation) may put implicit pressure on the patient to participate. It would be paternalistic to assume that the patient feels obliged or coerced into volunteering, but safeguards are needed to ensure the decision-making process to take part in health research is fully informed and unencumbered.15 Also, if physicians select patients beyond the inclusion criteria, they may deny access to others by declining to pass on study information, perhaps for paternalistic reasons to “protect” patients from an assumed burden of participating or for deeming that the research and patient are “not well matched.” Any interpersonal communication between patient and health professional regarding participation raises potential ethical concerns. Although patients may not necessarily be perceived as forming a vulnerable group, in our study, we found that individuals moved through vulnerable moments in their illness trajectory, and this vulnerability may have implications for how to recruit patients for research while offering them every opportunity to volunteer without being coercive in subtle ways. Throughout this process a reflexive self-awareness when interacting with potential patient participants may assist those involved in recruitment to navigate a path between being overly paternalistic and practicing a subtle or nuanced coercion. Respect for autonomy encompasses an acknowledgment of agency and
April 2010
respect for the participants’ priorities, experiences, and motivations. An interview situation may leave the participant vulnerable to an implicit agenda of the researcher, who attempts to balance listening to detailed accounts, against the aims of the research and practical considerations such as time. There is the potential to “mute” the participant by categorizing some aspects of the conversation as “going off topic” and keeping to an agenda of topics to be covered. Considering many of the participants expressed they had taken part in the study to share their experiences, with several commenting “I wanted to tell my story” or “I have waited a long time for this” and with others coming to the interview with logs of their experiences, this is of particular significance. Built into the topic guide were opportunities for participants to ask questions and give feedback about the interview, as well as introduce or elaborate on their own priorities. When applying an ethics lens to the interview, the strategic process of “active listening” becomes suffused with “ethical moments.” Negotiating the requirements of the research aim and the autonomy of the participant in the reality of the interview situation brings to the surface the “at odds” relationship of researcher and researched, which often is muted by descriptions of rapport building. This process illustrates the ongoing ethical moments that researchers face and the need for a reflexive approach; that is, the researcher reflects on the details of the research process and her or his role and interactions with participants and takes actions accordingly. Participants are vulnerable to being misrepresented. Because the researcher co-constructs the interview with the participant and analyzes and interprets the talk, the potential to misrepresent the individual and associated groups is always present.
Volume 90
Number 4
Physical Therapy f
621
Qualitative Research Ethics Box 2. Justice Quote 1: He [family physician] told me there was a 1-year waiting list, so I said, “Well, put me on that list, and I want to see somebody about it.” I just didn’t like the idea . . . to think my fingers could be so swollen, I can’t bend them for a whole year. And then he gave me Celebrex,a so I went home. That’s when I e-mailed you. I was so frustrated that I went on the Internet again. It upset me . . . . I was thinking . . . rheumatoid arthritis because of inflammation . . . even though my doctor did not [think that]. I read [that] on the Internet, anyway, and I e-mailed you. I wanted to talk to somebody. I wanted somebody to dump it on. (Nicolette) Quote 2: As soon as I read it [study information leaflet], I thought there is somebody out there that’s listening and going to be proactive and maybe help anybody else that’s going through this. That is my only concern. People have to get into a rheumatologist early; you can’t wait 8 months to see a rheumatologist. I thought this can help somebody down the road. (Maple) Quote 3: Because I was hoping that . . . I know I’m not the only one . . . and that if we all have similar experiences, maybe things will change. I want to prevent somebody else going through this, and if you learn about it, you can do something. (Lee) Quote 4: I love research. I just love it and I think it’s so important, and it’s also a way of giving back some of the good things I have been given, right? And I always know . . . often it’s hard to find people to do your research, right? And I just think . . . I am always researching, right? And for me, it’s [learning about the results] the natural outcome. If you ever publish anything, I would love to read anything you find out. (Flossie) a
G.D. Searle & Co, Div of Pfizer, 235 E 42nd St, New York, NY 10017-5755.
In our study, interdisciplinary discussions highlighted different aspects of the transcripts as significant, and this was a constant reminder of the need to be “ethically vigilant,” staying as true as possible to the reported experiences of the participant by, for example, drawing on the preinterview communications and field notes to contextualize the interviews when interpreting the data. This process extends beyond being transparent when making data-driven claims and offering participants opportunities to comment on the emerging findings; it is underpinned by a reflexivity that involves a continual self-awareness of the researchers’ personal responses to the data. Justice Hewitt stated, “Concepts of justice are explained in terms of what is deserved by each individual, and to what each individual is entitled, without partiality and with the aim of delivering equitable treatment”37(p1153) Attending to this principle means offering participants a fair and equitable distribution of burdens and benefits. Although other forms of justice, such as procedural justice, may be applicable here, it is beyond the 622
f
Physical Therapy
Volume 90
scope of this article to explore the range of ways justice may be conceptualized and utilized in discussions of ethical aspects of the qualitative paradigm. Many of the volunteers who made contact conveyed how they hoped to benefit in some way from study participation. Some hoped for information to help them manage their symptoms or for advice about a medical system that they found difficult to navigate (Box 2, quote 1). We offered a resource sheet to all participants who contacted us, with details of Web sites and patient- and arthritis-related organizations that offered advice and information. Some participants reported receiving the resource sheet was a benefit of taking part in the study. Some participants described the opportunity to share their experiences as a benefit and stated that taking part in the study helped them face and manage a new diagnosis of a debilitating, painful chronic illness. In such circumstances, being denied access to research can be perceived as a justice issue, for example, when a gatekeeper (eg, a health care pro-
Number 4
fessional) offers study information to some patients for reasons beyond the formal inclusion criteria, denying others the opportunity for potential benefits. Patients may interpret being denied access to research as being denied a voice, which may be particularly significant for those who feel they are not being heard and that their experiences were “invalidated” in the medical consultation. The context in which individuals volunteer to participate in health research has implications for the ways in which researchers communicate nonaccess to potential participants who may not fulfill the study criteria and who may see subsequent nonparticipation as being denied a form of care. The most disadvantaged individuals may not have access to research participation for a number of reasons; they may be preoccupied with negotiating multiple conditions (social, medical, and personal).53 Some of our participants volunteered but did not take part due to a reported inability to negotiate their illness, daily life activities, and participation. Others may not have access to a gatekeeping professional (who may or April 2010
Qualitative Research Ethics may not provide them with study information); they may live in remote communities or may not have access to recruitment materials from the Internet or patient organizations. Giving participants the opportunity to express their experiences is a justice issue because the interview contributes to our understanding of human experience. Particular meanings, processes, and complexities might remain hidden or misunderstood if research does not generate in-depth knowledge of a range of experiences under investigation. Reflexive construction of an interview schedule or topic guide means researchers being aware of the potential to impose their perspectives in subtle ways on the interview, through their use of language and choice of content and providing “spaces” for participants to prioritize their concerns (in keeping with the research’s broad aims). Analysis and interpretation of interviews also are justice issues if a qualitative study designed to explore the complex messiness of the lived experience of illness is explained in terms of variables and cause and effect more suitable to the positivist paradigm. It also is important to apply theory to qualitative accounts in order to respect the experiences of individual accounts, while recognizing the broader context in which personal experiences take place, and to assess the feasibility of transferring the findings to other settings. Careful researcher attention to the complexity, multidimensionality, and nuances of context and to the relevance of theory work to guard against a distortion of findings, which may stigmatize groups, while neglecting fundamental social problems. Dissemination of the qualitative findings also can be framed as a justice concern if participants give their time and share their experiences (which may be burdensome), with April 2010
the understanding that they are contributing to a knowledge base to improve practice (which they may perceive as a benefit). Participants in our study hoped to contribute to medical knowledge and improve patient experience of illness, treatment, and care; they conveyed helping others as a benefit of taking part in the research, even if they felt they personally would not benefit directly from research participation (Box 2, quotes 2 and 3). Despite increasing applicability and visibility, qualitative research still is neglected in some prominent medical journals and can be seen as anecdotal, or criticized on the basis of quantitative measures of reliability, validity, and generalizability.54 Efforts need to be made to publish qualitative research widely and make it accessible to different communities and disciplines. It is a waste of funding resources if findings are not considered useful, worthy of publication, or disseminated to suitable stakeholders. Most participants in our study reported an interest in knowing the results of the project. Some identified how they perceived learning about the research outcomes as a benefit of taking part (Box 2, quote 4). The accounts make explicit the responsibility the researcher has to participants55 to attempt to disseminate the research findings and improve awareness of the results among all relevant stakeholders, including participants themselves. A reflexive self-awareness about the details and implications of the research process invites researchers to be conscious of the justice issues from study inception to dissemination.
Beneficence/ Nonmaleficence The principles of beneficence and nonmaleficence involve an obligation to provide benefits for the patient and to balance such benefits
against risks22 and require that the researcher should do the patient no harm and should prevent harm and remove existing sources of harm.56 This concern highlights a potential conflict for health researchers who fulfill multiple roles, such as nurse research coordinators who advocate on behalf of the researcher (eg, principal investigator) and the patient, as well as the potential for misunderstanding when the patient places trust in the health researcher. Such situations raise ethical issues and have been identified as warranting a sort of “trust wariness” on behalf of the participant.15 Risks include overburdening individuals due to physician estimations of patient suitability beyond the inclusion criteria when not engaging fully with risks (or problems in assessing risks) that are associated with qualitative research rather than those in clinical trials. When obtaining consent for an interview study, it is never possible to accurately estimate risk, in terms of emotional upset, but we can anticipate its potential and take appropriate measures to prevent harm. In our study, respect for the priorities of the participant and the aims of the project were negotiated throughout the interview. We attempted to recognize when it was inappropriate to probe further, despite the interview guide, as the risk to a person’s emotional well-being may outweigh the hope of gaining rich data. For example, seemingly straightforward questions (from the perspective of the interviewer) could elicit an emotional response (Box 3, quotes 1 and 2). Ongoing decisions about how the interview unfolds need to be made in vivo. Some participants talked in emotional terms when describing what prompted their first appointment, as it symbolized a loss in their lives and the start of what was, in some cases, a debilitating illness and
Volume 90
Number 4
Physical Therapy f
623
Qualitative Research Ethics Box 3. Beneficence/Nonmaleficence Quote 1: Researcher: When you went to the doctor, what made you go then? Jean: Yes, I was having pains in my hands and in my fingers. These joint points . . . I did have a lot of stress at work. I wasn’t feeling well. I did, I do have depression. I had one child [who] had problems of depression, too . . . [who] was seeing a couple of counselors and taking some anti-depressants medication . . . [who] did go into a hospital at one time. All of this, of course, was stressful. [My child] committed suicide . . . and from that time, I was diagnosed with depression because I was going to consider suicide myself. I made a bit of an attempt. It was not a good one. Quote 2: Researcher: You said your children were very helpful; how important is that? Sarah: Huge. Um [pause] in the beginning [voice faltering, sounds upset], before we really knew what was going on, I was overwhelmed. I couldn’t hold a knife to cut cheese to make sandwiches, and in those days the kids were much younger, one would have a meat sandwich, and one would have chicken, one would have ham, one would have roast beef, one would have mustard, one would have mayonnaise, one would have, you know. . . . . I had it all organized. It was part of my morning ritual, and it was important to me to be able to do that.
an uncertain future. In another example, a seemingly innocuous question caused emotional upset for the participant, who was close to tears as she responded to a probe to gain further detail about family life and her children (Box 3, quote 2). This participant was able to continue in paid employment, but a routine task for her symbolized being a mother, and she conveyed suffering and loss beyond her functional debility. Participants may offer personal and sensitive information spontaneously, with no apparent upset, or they may become distressed unexpectedly (Box 3, quotes 1 and 2). The observation here is that aspects of risk are subtle and ongoing and emotional risk in an interview situation cannot be anticipated in advance, but the potential for it can be, and researchers need to attempt “ethical listening”—a blend of flexibility, structure, sensitivity, and pragmatism. The resource sheet we provided also included counseling service details (a free service offered locally). There also is a risk of confusing the research interview with the therapeutic interview, so the participant may have an inaccurate estimation of risks and benefits. One participant noted that she was so frustrated with 624
f
Physical Therapy
Volume 90
the health care system that she wanted to “dump” on somebody (Box 2, quote 1). Another participant reported severe illness, debilitating symptoms, acute reactions to medication, and depression and anticipated that she would find it helpful “just to talk to someone.” Although participants may gain benefits from the interview, the potential for psychological distress— during or after the interview— should not be underestimated. Also, the line between building rapport in a research interview and offering a therapeutic interview encounter needs to be navigated with care. One participant related how, in a previous interview in a hospital setting, she had shared sensitive details and found it difficult to adjust to the sudden termination of the interview and being back out “on the street.” Another participant described discussing help-seeking and had later spent some time reflecting on her previous actions and wondering whether she could have consulted her family physician earlier and prevented symptom deterioration. Participants also reported more practical burdens and inconvenience. Some conveyed negotiating participation with their medication regimens, appointments, daily life, and
Number 4
symptoms (fatigue, pain, and discomfort). Others described the impact of the interview. One participant noted how she had been surprised by feeling physically tired on returning to work after her lunchtime interview and had not anticipated any negative impact, despite her employer advising of this possibility. It is difficult to assess these potential risks prior to an interview study, but this difficulty underscores the emerging ethics in qualitative research and the need for a reflexive approach, anticipating ethical moments and how to manage them throughout the process, as well as building measures into the study design that will minimize harm.
Discussion We identified the value of qualitative health research57 and ethical issues that emerged in our in-depth interview study, applying the principlebased approach of autonomy, justice, and beneficence/nonmaleficence, which traditionally is associated with human subject protection in medical research, although increasingly criticized for its suitability to qualitative research. Our findings mirrored ethical concerns that arise in quantitative research (eg, issues of coercion and recruitment through health care professionals). We also April 2010
Qualitative Research Ethics identified ethical considerations in the research process (Fig. 2). There are fundamental differences between the positivist and interpretivist paradigms (Fig. 1) that are core to the different protections required for participants. The interpersonal researcherresearched relationship, which involves ongoing interactions, building mutual trust and respect, the active role of the participant and researcher in co-constructing the data generation, and the qualitative interpretation of the findings, is in direct contrast to the ideal of the passive research participant and the “arms-length” scientific researcher who observes the phenomenon and collects the results. The unfolding nature of qualitative research and the intersubjectivity of researcher and researched shifts the notion of informed consent as “a moment in time” and the lynchpin of human protection to a process embedded with ethical moments58 inextricably linked to research design and methodology inviting reflexivity throughout.59 Our findings are limited to issues that arose in our interview study. We did not design the project to analyze ethical aspects of research, but have descriptively drawn on field notes and interview extracts to illustrate ethical issues that emerged. We use the traditional principle-based approach as an exploratory exercise; we do not claim this framework is the most effective way of considering human subject protection. Applying the traditional approach to our qualitative study extends our knowledge of the utility and value of this approach in context and is a strength of this article. In line with our study, recruitment through gate-keeping has been seen as a practical obstacle to research in the primary care setting.48,50 Some authors have suggested that physicians may not be suitably familiar with qualitative research and its potential contribution to EBP,60,61 and April 2010
other authors have identified how this lack of familiarity is paralleled in quantitative research.62 Other authors broach ethical aspects and ask whether family physicians should be encouraged to recruit patients by offering both reimbursements and payments46 and whether clinician recruitment of patients can be assumed to be coercive or whether this assumption is another form of paternalism.49 Delaney suggested that research should be prioritized and made “directly relevant” to the training and development of the nonacademic clinician.50 Because ethical issues can emerge throughout the research process and are not confined to particular stages, such as gaining informed consent, a reflexive approach characterized by a thoughtful and ongoing self-awareness, attention to detail, and sensitivity to the individual’s role in the research enterprise (eg, health care professional as research worker/recruiter) highlights relationship dynamics and enhances ethical oversight. Education and training that delineate the ethical differences between the quantitative and qualitative approaches seem key if health care professionals continue to play a role in research, including recruitment. Gate-keeping can diminish autonomy in both the positivist and interpretivist paradigms. Our findings support other patient-participant reports of being recruited or influenced by health care professionals to participate in research such as in clinical trials.15 Other authors have called for an investigation into barriers to recruitment of groups perceived as vulnerable because of misguided ethical guidelines.63 Such barriers raise questions that also were prompted in our study: When does an individual’s illness status induce vulnerability, and what is the impact on decisions to participate and giving informed consent?38 Gatekeeping also can threaten the valid-
ity of the findings; if the research fails to generate useful data (eg, due to a compromised sample),45 the subsequent wasted resources and production of research of limited validity are ethical issues.49 As in our study, other types of recruitment may be more appropriate to offer individuals the opportunity to exercise self-selection. This opportunity to exercise self-selection seems more in keeping with patient-centered care and collaborative research. However, sampling problems still arose in our study, illustrating that particular groups may be under-represented in research and their experiences neglected in EBP. In Canada, effective strategies need to be put in place to ensure groups and individuals have access to participation. These strategies may involve building trust and reaching remote communities and disadvantaged groups. Reflecting our findings, the interview has been described as a moral endeavor,64 and 3 types of potential ethical problems in qualitative interviews have been identified: (1) the design itself, (2) the research relationship, and (3) the process involved in interpreting qualitative data, and particularly the role of the researcher in co-constructing the findings (results).65 Cox also noted the importance of ethical listening and being sensitive to when, and when not, to probe.66 Interviewer awareness of insensitive probing is indicative of a reflexive approach; sometimes the offer of “good data” may need to be relinquished in favor of “good ethics.” The potential for psychological harm when sharing sensitive information in in-depth interviews and the impact of taking part have been considered,67 as well as the potential to distort findings.68 Our findings illustrated how interviews are an effective way of gaining experience of suffering69 and how ethics emerge in interpersonal rela-
Volume 90
Number 4
Physical Therapy f
625
Qualitative Research Ethics tionships in the interview situation.70 The researcher has a moral obligation to ensure that the research is worthwhile when people share time and suffering,71 and because there is scope to interpret and co-construct the realities that are then part of the knowledge base, acting reflexively is crucial to ethical practice.37 Our experiences revealed that it was difficult to know when ethical issues would arise, or how nuanced (constructing the interview guide) or unique (psychological harms) they would be. Ethical issues emerged in the construction of the interview guide and the practical circumstances of the interview. Skillful researchers try to avoid imposing their own structures and assumptions upon interviewees’ view of the world.3 Careful recording of field notes and analytic reflection are required. Field notes serve as the written account of what the researcher sees, hears, experiences, and thinks and supplement the interview data. The interview is an ethical process, from the creation of the interview guide to the analysis of the accounts. This approach to ethics in research is informed by a process model of research ethics, whereby considerations of autonomy and respect for the perspectives and experiences of consumers are central from the early stages of study design.58 Regarding issues of knowledge translation, our interdisciplinary team faced challenges in discussions that highlighted our disciplinary assumptions and diverse perspectives. To facilitate mutual respect and shared decision making, measures were built into the research process to ensure inclusion of all perspectives; ongoing team communications were crucial in order to clarify meaning, recognize subtle findings, and communicate complexity effectively.
626
f
Physical Therapy
Volume 90
There also are ethical implications for knowledge dissemination and the development, application, and visibility of qualitative methods in health research. Although there has been seminal qualitative health research for more than 50 years (eg, about the patient-practitioner relationship72 and help-seeking73), disagreement exists about both the value and acceptance of the utility of the qualitative paradigm. Although previously ignored or viewed as a weak source of evidence, Sandelowski43 noted that qualitative research is now considered essential to the EBP goal of improving health care. Atkin54 described how patient accounts are still seen as “meaningless” and not regarded as evidence, but rather as anecdotal. Thus, despite the apparent shift beyond the oppositional quantitative and qualitative paradigms, opposition still characterizes many evaluative discussions in an interdisciplinary research context. Atkin noted that an emphasis on positivist-based priorities and quality appraisal may discourage the more reflexive engagement that characterizes qualitative methods and asked: “Is there a danger that by pressing for the inclusion of qualitative accounts in more applied research, often dominated by quantitative methods, we risk producing analysis that is descriptive, uncontextualised and little more than a token gesture, paying lip service to individual experience?”54 This question illustrates the inextricable link between ethics and methodology: if lip service is paid to investigating the “lived experience” of illness in all its messiness, there is a danger that we will not do justice to participant accounts.
Conclusion We have identified: (1) that the traditional principle-based approach can be usefully applied to qualitative research ethics, if we use the concepts flexibly and prioritize context
Number 4
and process, and (2) that the realities of doing “everyday ethics” highlight ethical moments and the need for a reflexive stance at every level and stage of research. Reflexivity involves the researcher always being aware of the role of self in the process, a sensitivity to the ethical concerns that may arise, and taking measures to prevent harm (eg, anticipating that participants may experience psychological distress). An ongoing awareness of the ethical moments that may arise throughout the research process and one’s role in the process increases ethical rigor at all stages of research, such as when designing the study (offering counseling services or a resource sheet with useful and valid Web sites) and during the interview (how to listen sensitively and when and when not to probe). Ethical, rigorous, and useful qualitative research involves being accountable to research participants and echoes the ethical concerns in physical therapy. An ethos of patientcentered care and issues of empathy, autonomy, and respect are highlighted in our account of qualitative research from the study’s inception to the dissemination of results. An ethical lens in qualitative health research brings particular challenges and insights to methodological concerns about recruitment procedures, the interview or other data-generation processes, and knowledge translation, and, ultimately, the usefulness of the findings to bring benefits to patients/ consumers (and health care professionals) and to enhance health care. Dr Townsend conceptualized the article and wrote all drafts of the manuscript, conducted the majority of the interviews, led the analysis, and was the principal investigator during the pilot phase of the project. Dr Cox consulted on the ethical issues emerging from the project and data analysis and assisted with the writing of the manuscript by reading the manuscript, making comments, and clarifying content. Dr Li was the principal investigator of the ERAHSE Project. As the
April 2010
Qualitative Research Ethics senior author, she provided guidance on drafts and contributed to the preparation of the manuscript. The authors thank the participants, who shared their time and experiences, and the members of the ethics review board who reviewed the study protocol. Ethical approval was obtained from the University of British Columbia Behavioral Research Ethics Board and the Vancouver Health Research Institute. This article was developed from a paper given at the CARE V International Conference; April 23–25, 2008; Oslo, Norway. This research was funded by the Canadian Institutes of Health Research (CIHR). Dr Cox was supported through a career award from the Michael Smith Foundation for Health Research during the early part of this research. Dr Li was supported by a CIHR New Investigator Award and an American College of Rheumatology Research and Education Foundation Health Professional New Investigator Award. This article was received December 4, 2008, and was accepted December 28, 2009. DOI: 10.2522/ptj.20080388
References 1 Richards H, Swartz L. Ethics of qualitative research: are there special issues for health services research? Fam Pract. 2002;19: 135–139. 2 Alderson P. On Doing Qualitative Research Linked to Ethical Healthcare. London, United Kingdom: The Wellcome Trust; 1999. 3 Britten N. Qualitative interviews in qualitative research. In: Pope C, Mays N, eds. Qualitative Research in Healthcare. London, United Kingdom: BMJ Publishing Group; 1996:28 –35. 4 Bury M. Chronic illness as biographical disruption. Sociol Health Illn.1982;4:167– 182. 5 Ong B, Coady D. Qualitative Research: Its Relevance and Use in Musculoskeletal Medicine. Chesterfield, Derbyshire, United Kingdom: Arthritis Research Campaign; 2006. 6 Williams G. The genesis of chronic illness: narrative re-construction. Sociol Health Illn. 1984;6:175–200. 7 Donovan J, Blake D. Qualitative study of interpretation of reassurance among patients attending rheumatology clinics: “Just a touch of arthritis doctor?” BMJ. 2000; 320:541– 4. 8 Townsend A, Wyke S, Hunt K. Selfmanaging and managing self: practical and moral dilemmas in accounts of living with chronic illness. Chronic Illn. 2006;2:185– 194.
April 2010
9 Townsend A, Hunt K, Wyke S. Managing multiple morbidity in mid-life: a qualitative study of attitudes to drug use. BMJ. 2003; 327:837. 10 Townsend A, Wyke S, Hunt K. Frequent consulting and multiple morbidity: a qualitative comparison of “high” and “low” consulters of GPs. Fam Pract. 2008;25: 168 –175. 11 Corrigan M, Cupples ME, Smith SM, et al. The contribution of qualitative research in designing a complex intervention for secondary prevention of coronary heart disease in two different healthcare systems. BMC Health Serv Res. 2006;6:90. 12 Jack S. Utility of qualitative research findings in evidence-based public health practice. Public Health Nurs. 2006;23:277– 283. 13 Cox S, Townsend A, Preto N, et al. Ethical challenges and evolving practices in research on health research. Health Law Rev. 2009;17:33–39. 14 Owen M, Emerson C, Kolpack P, et al. Informing governance through evidencebased research on REBs: challenges and opportunities. Health Law Rev. 2009;17: 40 – 46. 15 McDonald M, Townsend A, Cox S, et al. Trust in health research relationships: accounts of human subjects. J Empir Res Human Res Ethics. 2008;3:35– 47. 16 Hammersly M. Taking Sides in Social Research. London, United Kingdom: Routledge; 2000. 17 Shaw I. Ethics in qualitative research and evaluation. J Soc Work. 2003;3:9 –29. 18 Townsend A. A Canadian researcher’s perspective: the human subject experience. Paper presented at: CARE V International Conference; April 23–25, 2008; Oslo, Norway. 19 Henley LD, Frank DM. Reporting ethical protections in physical therapy research. Phys Ther. 2006;86:499 –509. 20 Purtillo RB. Thirty-First Mary McMillan Lecture: Time to harvest, a time to sow: ethics for a shifting landscape. Phys Ther. 2000;80:1112–1119. 21 Jensen GM. Qualitative methods in physical therapy research: a form of disciplined inquiry. Phys Ther. 1989;69:492–500. 22 Beauchamp T, Childress J. Principles of Biomedical Ethics. 5th ed.Oxford, United Kingdom: Oxford University Press; 2001. 23 Orb A, Eisenhauer L, Wynaden D. Ethics in qualitative research. J Nurs Scholarsh. 2001;33:93–96. 24 Emanuel E, Wendler D, Grady C. What makes clinical research ethical? J Am Med Assoc. 2000;283:2701–2711. 25 Reverby S, ed. Tuskagee’s Truths: Rethinking the Tuskagee Syphilis Study. Studies in Social Medicine Series. Chapel Hill, NC: The University of North Carolina Press; 2000. 26 Riessman C, Mattingly C. Introduction: toward a context-based ethics for social research in health. Health (London). 2005; 9:427– 429.
27 Trials of War Criminals Before the Nuremberg Military Tribunals Under Control Council Law. Washington, DC: Government Printing Office; 1949;181– 182. 28 World Medical Association. Declaration of Helsinki: ethical principles for medical research involving human subjects. Available at: http://www.wma.net/en/30publi cations/10policies/b3/index.html. Accessed October 2009. 29 US Department of Health Education and Welfare. The Belmont Report: Ethical Principles and Guidelines for the Protection of Human Subjects of Research. Washington, DC: The National Commission for the Protection of Human Subjects of Biomedical and Behavioural Research; 1979. 30 World Health Organization. International Ethical Guidelines for Biomedical Research Involving Human Subjects. Geneva, Switzerland: Council for International Organizations of Medical Sciences (CIOMS) in collaboration with the World Health Organization; 2002. 31 Canadian Institutes of Health Research. Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans. Ottawa, Ontario, Canada: Natural Sciences and Engineering Research Council of Canada, Social Sciences and Humanities Research Council of Canada; 1998 (with 2000, 2002 and 2005 amendments). 32 Statement of Ethical Practice for the British Sociological Association; Belmont, Durham, United Kingdom: British Sociological Association; March 2002, appendix updated May 2004. 33 US Department of Health and Human Services, Office for Human Research Protections. Available at: http://www.hhs.gov/ ohrp/compliance. Accessed October 2009. 34 Gillon R. Medical ethics: four principles plus attention to scope. BMJ. 1994;309: 184 –188. 35 Barbour R. The role of qualitative research in broadening the “evidence base” for clinical practice. J Eval Clin Pract. 2000;6: 155–163. 36 Shepard KF, Jensen GM, Scholl BJ, et al. Alternative approaches to research in physical therapy: positivism and phenomenology. Phys Ther. 1993;73:88 –97. 37 Hewitt J. Ethical components of researcher researched relationships in qualitative interviewing. Qual Health Res. 2007;17:1149 –1159. 38 Moyle W. Unstructured interviews: challenges when participants have a major depressive illness. J Adv Nurs. 2002;39:266 – 273. 39 Edwards I, Jones M, Carr J, et al. Author response to invited commentary on “clinical reasoning strategies in physical therapy.” Phys Ther. 2004;84:334 –335. 40 Resnick L, Jensen GM. Using clinical outcomes to explore the theory of expert practice in physical therapy. Phys Ther. 2003;83:1090 –1106. 41 Edwards I, Jones M, Carr J, et al. Clinical reasoning strategies in physical therapy. Phys Ther. 2004;84:312–330.
Volume 90
Number 4
Physical Therapy f
627
Qualitative Research Ethics 42 Mayan MJ. Essentials of Qualitative Inquiry. Walnut Creek, CA: Left Coast Press Inc; 2009:137. 43 Sandelowski M. Whatever happened to qualitative description? Res Nurs Health. 2003;23:334 –340. 44 Swisher LL. A retrospective analysis of ethics knowledge in physical therapy (1970 – 2000). Phys Ther. 2002;82:692–706. 45 Townsend A, Cox SM, Adam P, Li LL. Recruitment in qualitative health research: methodological and ethical issues arising from a study on experiences of rheumatoid arthritis. Paper presented at: Annual Canadian Arthritis Network Scientific Conference Proceedings; November 30 –December 2, 2006; Winnipeg, Manitoba, Canada. 46 Draper H, Wilson S, Flanagan S, Ives J. Offering payments, reimbursements and incentives to patients and family doctors to encourage participation in research. Fam Pract. 2009;26:231–238. 47 Graffy J, Grant J, Boase S, et al. UK research staff perspectives on improving recruitment and retention to primary care research: nominal group exercise. Fam Pract. 2009;26:48 –55. 48 Salmon P, Peters, S, Rogers A, et al. Peering through the barriers in GPs’ explanations for declining to participate in research: the role of professional autonomy and the economy of time. Fam Pract. 2007;24:269 –275. 49 Wilson S, Draper H, Ives J. Ethical issues regarding recuitment to research studies within the primary care consultation. Fam Pract. 2008;25:456 – 461. 50 Delaney B. Engaging practitioners in research: time to change the values of practice rather than the way research is carried out? Fam Pract. 2007;24:207–208. 51 Li LC, Adam P, Townsend AF, et al. Improving healthcare consumer effectiveness: an animated, self-serve, Web-based research tool (ANSWER) for people with early rheumatoid arthritis. BMC Med Inform Decis Mak. 2009;9:40.
628
f
Physical Therapy
Volume 90
52 Gillon R. Ethics needs principles: four can encompass the rest and respect for autonomy should be “first among equals.” J Med Ethics. 2003;29:307–312. 53 Parry O, Bancroft A, Snick W, Amos A. Nobody home: issues of respondent recruitment in areas of social deprivation. Crit Public Health. 2001;11:305–311. 54 Atkin K. “Meaningless patient stories”? The politics of doing qualitative research in multi-disciplinary contexts. Plenary Address presented at: Annual Qualitative Health Research Conference; October 4 – 6, 2009; Vancouver, British Columbia, Canada. 55 Townsend A. Policy, practice, and funded qualitative health research: interview accounts and interviewer accountability. Paper presented at: Annual Congress of Qualitative Inquiry Conference; May 2–5, 2007; Urbana-Champaign, IL. 56 Tuxhill C. Ethical aspects of critical care. In: Millar B,Bernard P, eds. Critical Care Nursing. London, United Kingdom: Balliere Tindall; 1994:250 –272. 57 Alderson P. Qualitative Research: A Vital Resource for Ethical Healthcare. London, United Kingdom: The Wellcome Trust; 1999:13. 58 Guillemin M, Gillam L. Ethics, reflexivity and “ethically important moments” in research. Qual Inquiry. 2004;10:261–280. 59 Shaw IF. Ethics in qualitative research and evaluation. J Soc Work. 2003;3:9 –29. 60 Malturud K. Qualitative research: standards, challenges and guidelines. Lancet. 2001;358:483– 488. 61 Turgeon J, Cote L. Qualitative research in family medicine: an inevitable development. Can Fam Physician. 2000;46: 2171–2172. 62 Herber OR, Schnepp W, Rieger MA. Recruitment rates and reasons for community physicians’ non-participation in an interdisciplinary intervention study on leg ulceration. BMC Med Res Methodol. 2009; 9:61.
Number 4
63 Hewison J, Haines A. Confidentiality and consent in medical research: overcoming barriers to recruitment in health research. BMJ. 2006;333:300 –302. 64 Kvale S. Interviews: An Introduction to Qualitative Research Interviewing. London, United Kingdom: Sage; 1996. 65 Ramos M. Some ethical implications of qualitative research. Res Nurs Health. 1989;12:57– 63. 66 Cox SM. Context, communication and paradox: on learning not to ask “overly sensitive” questions. In: Hallowell N, Lawten J, Gregory S, eds. Reflections on Research: The Realities of Doing Research in the Social Sciences. Berkshire, United Kingdom: Open University Press; 2005:chap 2. 67 Morse J. Are there risks in qualitative research? Qual Health Res. 2001;11:3– 4. 68 Cushing A. Historical and epistemological perspectives on research and nursing. J Adv Nurs. 1994;20:406 – 411. 69 Clarke J, Febbraro A, Hatzpantelis M, et al. Poetry and prose: telling the stories of formerly homeless mentally ill people. Qual Inquiry. 2005;11:913–932. 70 Richards H, Elmslie C. The “doctor” or the “girl from the university”? Considering the influence of professional roles on qualitative interviewing. Fam Pract. 2000;17:71– 75. 71 Thorne S, Derbyshire P. Land mines in the field: a modest proposal for improving the craft of qualitative research. Qual Health Res. 2005;15:1105–1113. 72 Balint M. The Doctor, His Patient, and the Illness. New York, NY: International Universities Press; 1957. 73 Koos E. The Health of Regionville: What the People Thought and Did About It. 2nd ed. New York, NY: Hafner Publishing Co; 1954.
April 2010
CARE V Conference Series Continuing Professional Development Is Associated With Increasing Physical Therapists’ Roles in Arthritis Management in Canada and the Netherlands Linda C. Li, Emalie J. Hurkmans, Eric C. Sayre, Thea P. M. Vliet Vlieland
Background and Objective. This study explored the relationships among the roles assumed by physical therapists in arthritis care and their previous participation in arthritis courses for continuing professional development (CPD).
Design. A cross-sectional mail survey was conducted. Method. A total of 600 Canadian physical therapists and 461 Dutch physical therapists practicing in orthopedics were randomly selected to participate in a mail survey. The questionnaire covered areas related to their clinical practice, previous participation in arthritis-related CPD courses, and roles in the management of osteoarthritis (OA) and rheumatoid arthritis (RA). Poisson regression was used to explore the associations between physical therapists’ participation in arthritis-related CPD courses and the number of roles they assumed in OA and RA care, after adjusting for personal characteristics, arthritis caseload, and country of practice.
Results. The survey response rates were 47.7% in Canada and 50.5% in the Netherlands. A total of 424 participants (Canada⫽224, the Netherlands⫽200) had treated patients with OA in the previous month, and 259 participants (Canada⫽68, Netherlands⫽191) had treated patients with RA in the previous month. The most common roles reported by participants were providing traditional physical therapy interventions and providing postsurgical care. Arthritis-related CPD courses significantly increased (ie, multiplied) the expected number of roles assumed by physical therapists by a factor of 1.32 (95% confidence interval⫽1.11, 1.56) in OA management and 1.69 (95% confidence interval⫽1.34, 2.13) in RA management.
Limitations. Physical therapists’ roles in arthritis management were obtained through self-reporting, which might differ from their actual clinical practice.
Conclusions. This exploratory analysis highlights the association between participation in arthritis-related CPD courses and the roles assumed by physical therapists in OA and RA management. Further research is needed to understand the effects of CPD activities on other areas of physical therapist practice and on patients’ outcomes.
L.C. Li, PT, PhD, is Assistant Professor and Harold Robinson/ Arthritis Society Chair in Arthritic Diseases, Department of Physical Therapy, University of British Columbia, Vancouver, British Columbia, Canada, and Research Scientist, Arthritis Research Centre of Canada, 895 W 10th Ave, Room 324, Vancouver, British Columbia, Canada V5Z 1L7. Address all correspondence to Dr Li at: lli@ arthritisresearch.ca. E.J. Hurkmans, PT, MSc, is a PhD candidate in the Department of Rheumatology, Leiden University Medical Center, Leiden, the Netherlands. E.C. Sayre, PhD, is Postdoctoral Fellow, School of Population & Public Health and Department of Medicine, University of British Columbia, and Arthritis Research Centre of Canada. T.P.M. Vliet Vlieland, PT, MD, PhD, is Associate Professor, Department of Orthopaedics and Department of Rheumatology, Leiden University Medical Center. [Li LC, Hurkmans EJ, Sayre EC, Vliet Vlieland TPM. Continuing professional development is associated with increasing physical therapists’ roles in arthritis management in Canada and the Netherlands. Phys Ther. 2010;90:629 – 642.] © 2010 American Physical Therapy Association
Post a Rapid Response or find The Bottom Line: www.ptjournal.org April 2010
Volume 90
Number 4
Physical Therapy f
629
Continuing Professional Development and Physical Therapists’ Roles in Arthritis Management
A
rthritis is the most common cause of severe, chronic pain and disability.1,2 From the patient’s perspective, the most important goals of arthritis care are to control pain,3–5 to limit functional debility, and to maintain a normal life.6 In 2005, 21.4 million Americans aged 65 years and over reported having arthritis or chronic joint symptoms, and this number is expected to double by 2030.7 In Canada, the estimated prevalence of arthritis in individuals aged 15 years or older was 16% in 2000,2 and it was projected to increase by almost 1% every 5 years.2,8 A similar trend has been observed in Europe.9 –11 However, despite the increasing demands for arthritis care, there is a shortage of arthritis specialists in some countries. For example, the Canadian Rheumatology Association recommended 1 rheumatologist per 70,000 Canadians,12 meaning that there should be about 480 rheumatologists in Canada. However, there are only 353 rheumatologists, with most of them practicing in urban regions.13 To address the gap in the demand and supply in arthritis care, new health services models have been developed that involve expanding the roles of rehabilitation professionals.14,15 Physical therapists are in a unique position to expand their roles in arthritis care because of their knowledge of the musculoskeletal system and their skills in assessing and managing orthopedic conditions. For example, physical therapists in the United Kingdom may receive on-the-
Available With This Article at ptjournal.apta.org • Audio Abstracts Podcast This article was published ahead of print on March 4, 2010, at ptjournal.apta.org.
630
f
Physical Therapy
Volume 90
job training from medical specialists so they can work as consultants to triage primary care referrals to rheumatology and orthopedic clinics.16 –18 In Canada, physical therapists may work as primary therapists who function as case managers and multiskilled health care providers (ie, the therapist may provide physical therapy and occupational therapy interventions).19 –22 Some physical therapists have begun to work in advanced practice roles, which may involve conducting comprehensive assessments, ordering investigative tests, and monitoring medications under medical directives.15,23 To enhance physical therapists’ knowledge of and skills in arthritis management, a number of continuing professional development (CPD) programs have been developed. In Canada, The Arthritis Society (TAS) and the Mary Pack Arthritis Program provide standardized workshops for physical therapists who work with patients with arthritis and other orthopedic conditions. The TAS program is a 1-week course that focuses on the assessment of inflammatory arthritis.19,20 The Arthritis Continuing Education (ACE) Program, offered by the Mary Pack Arthritis Program in the province of British Columbia,24 is a 3-day course that focuses on the management of inflammatory arthritis and osteoarthritis (OA). Both programs are taught by rheumatologists and experienced rehabilitation professionals. For physical therapists who want to work in advanced practice roles in rheumatology, extensive training programs, such as the Advanced Clinician Practitioner in Arthritis Care Program offered by St. Michael’s Hospital25 in the province of Ontario, are available.26,27 This 10month program is taught by rheumatologists and requires a significant commitment of resources from the physical therapists and their employers (eg, time off for the therapist to
Number 4
attend courses and internships). To our knowledge, fewer than 1% of physical therapists practicing in orthopedics and rheumatology have completed this type of advanced program in Canada.26 A number of European countries also have developed rheumatology CPD programs for physical therapists, although the organization and availability of these programs vary across countries. In the Netherlands, physical therapists practicing in rheumatology may complete a 10-day course on arthritis management provided by the Dutch Institute of Allied Health Care.28 The training provided by this private organization is accredited by the Royal Dutch Society for Physical Therapy,29 the Dutch professional association for physical therapists. Postprofessional (post– entry-level) training in the physical therapy management of rheumatic diseases at the master’s level also is available at one Dutch university of applied sciences. This clinical master’s level program is accredited by the NederlandsVlaamse Accreditatieorganisatie,30 the accreditation organization of the Netherlands and Flanders. A few studies have demonstrated benefits to patients treated by physical therapists who have successfully completed arthritis CPD training. In a randomized controlled trial (RCT), Bell et al31 found short-term improvement in morning stiffness, selfefficacy, and disease knowledge in patients who received treatment for rheumatoid arthritis (RA) from a TAStrained physical therapist compared with patients on a waiting list. A more recent RCT showed that patients with RA treated by a TAStrained primary therapist were more likely than those treated by a physical therapist or an occupational therapist with no TAS training to achieve a 20% or greater improvement in at least 2 of the following measures: pain, physical function, and knowlApril 2010
Continuing Professional Development and Physical Therapists’ Roles in Arthritis Management edge.14 Primary therapists were physical therapists and occupational therapists who functioned as multiskilled workers (ie, the therapist assumed more roles to provide both physical therapy and occupational therapy treatments and case management).21,22 Primary therapists with a physical therapy background might use their occupational therapist colleagues in a consultative fashion rather than transferring the case to another rehabilitation discipline for completion of the treatment.21,22 This model of care also has been used in pediatric rehabilitation.32 Li et al33 showed that the TAS-trained primary therapist provided a costeffective alternative to traditional rehabilitation treatment. Developing CPD activities such as arthritis training programs requires a substantial investment of resources. However, it remains unclear whether CPD activities have contributed directly to physical therapists assuming more roles in the management of arthritis. To address this knowledge gap, the primary objective of this study was to conduct an exploratory analysis to assess the relationship between the average number of roles assumed by physical therapists in arthritis care and their previous participation in arthritis CPD courses. We hypothesized that participation in CPD activities was a significant predictor of physical therapists assuming a larger number of roles in the management of OA and RA, after accounting for the country of practice, personal characteristics, and arthritis caseload. Our secondary objective was to explore the associations between previous participation in arthritis courses and specific OA and RA roles.
Method The survey instrument was originally developed for the Canadian Physiotherapist Arthritis Care Survey34 and was subsequently translated and April 2010
adapted for physical therapists in the Netherlands. Details of the survey administration in Canada are described elsewhere.34 Briefly, however, individuals were eligible if they were licensed by one of the provincial regulatory colleges to practice physical therapy and were practicing in orthopedics in Canada. Of the 10 regulatory colleges of the physical therapy profession, 9 agreed to participate and subsequently provided assistance to identify eligible physical therapists (n⫽6,994). The college in the province of British Columbia declined to participate due to its internal policy. A computergenerated table of random numbers was used to randomly select 600 physical therapists to receive the questionnaire. Individuals received their survey package in March 2007. Names and addresses of eligible physical therapists were obtained from 8 of the regulatory colleges. One regulatory college (College of Physical Therapists of Alberta) provided a list of computer-generated identification numbers of all eligible physical therapists for randomization due to their confidentiality policy. We then provided the survey packages and reminder letters for the Alberta regulatory college to mail to the selected participants. The questionnaire also was sent to 2 groups of physical therapists in the Netherlands in April 2007: (1) all 211 physical therapists who were members of 1 of 10 regional arthritis networks (referred to as “physical therapists in arthritis care”); eligibility for membership varies across networks, with some networks requiring their members to complete arthritis CPD courses; and (2) 250 physical therapists who were randomly selected from the remaining 20,367 registrants of the Royal Dutch Society for Physical Therapy (referred to as “registered physical therapists”). Thus, a total of 461 Dutch physical therapists received the questionnaire.
The full questionnaire covered 4 areas related to clinical practice, knowledge, and attitude toward physical therapists’ roles in rheumatology: (1) current practice and roles in assessment and treatment; (2) therapists’ confidence in arthritis management; (3) content of rheumatology training; and (4) general opinions on certification, specialization, and extended scope of practice. We defined certification as a “program and process where a learner completes prescribed training and passes an assessment with a minimum acceptable score.” The World Confederation for Physical Therapy defined physical therapist specialization as “the application of advanced clinical competence by a physiotherapist qualified in a defined area of practice within the field of activity recognised as physiotherapy.”35 For extended scope practitioners, we used the definition provided by The Chartered Society of Physiotherapy (United Kingdom), which describes these physical therapists as those “who are working beyond the recognized scope of practice of the profession of interest in innovative or non-traditional roles.”36 These roles may include “requesting investigations (eg, blood tests, scans, nerve conduction studies); using the results of investigations to assist clinical diagnosis and appropriate management of patients; and listing for surgery and referring to other medical and paramedical professionals.”36 Because advanced practice roles for physical therapists have not been legislated in Canada and the Netherlands, physical therapists require facility-specific medical directives to provide treatment that is outside the traditional scope of practice. Examples of medical directives may include physical therapists ordering radiography or laboratory tests on behalf of physicians in the same facility under specific terms and conditions.37
Volume 90
Number 4
Physical Therapy f
631
Continuing Professional Development and Physical Therapists’ Roles in Arthritis Management For the current study, we asked each participant: “Did you take any course(s)/workshop(s) in arthritis assessment and/or management after your entry-level training?” In addition, those who indicated that they had seen patients with OA or RA in the previous month were asked whether they assumed the following roles (Appendix 1): 1. Providing assessment and treatment traditionally provided by a physical therapist. 2. Providing assessment and treatment traditionally provided by other rehabilitation disciplines (eg, occupational therapy interventions). 3. Providing assessment and treatment outside the scope of physical therapist practice (eg, ordering investigative tests, providing injections). 4. Screening patients for physicians. 5. Providing public education. 6. Providing consultation together with another health care professional. 7. Referring patients to medical professionals. 8. Referring patients to other rheumatology rehabilitation professionals. 9. Providing presurgical care. 10. Providing postsurgical services. The items were selected based on consensus of the Canadian research team (consisting of 3 rheumatology researchers, 1 physical therapy educator, 1 rheumatology physical therapist, and 1 rheumatologist34) and then were verified by the Dutch research team (consisting of 2 rehabil632
f
Physical Therapy
Volume 90
itation researchers and 1 rheumatologist/health services researcher38). We believed that these items represented major roles assumed by physical therapists in managing OA and RA in Canada and the Netherlands. This assumption was confirmed by a subsequent systematic review of models of RA care, in which roles of physical therapists, nurses, and other health care professionals were examined.39 The 10 items were presented in the questionnaire without further explanation (Appendix 1). The questionnaire was originally developed in English and was pretested for face and content validity with physical therapists working in orthopedics (n⫽8) or rheumatology (n⫽6). The content subsequently was revised and reviewed by the same volunteers before it was used for the survey. It then was translated into French for physical therapists in the province of Quebec, Canada, and into Dutch for physical therapists in the Netherlands. A rigorous process of forward-backward translation was used to ensure accuracy (details described elsewhere34). Both the Canadian and Dutch surveys used the modified Dillman technique40,41 in order to elicit the fullest participation. For the first mailing, a letter explaining the intent of the study was included with the survey questionnaire. Three weeks later, a reminder postcard was sent to nonrespondents. Second and third reminder letters and another copy of the survey questionnaire were sent to the remaining nonrespondents 6 weeks and 8 weeks after the initial mailing. The Canadian survey was approved by the University of British Columbia Behavioural Research Ethics Board (application number: B06-0719). The Dutch survey received ethics approval from the medical ethics committee of the Leiden University Med-
Number 4
ical Center (application number: 06-097). Statistical Analysis Only participants who indicated that they saw at least one patient with OA in the previous month were included in the OA role analysis, and a similar criterion was applied for the RA role analysis. Participant characteristics were assessed using frequencies, or means and standard deviations, depending on the measure. For the primary objective, we assessed the association between the expected (ie, average) number of roles (minimum⫽0, maximum⫽7) assumed by physical therapists and their participation in arthritis CPD courses using Poisson regression. Only roles 4 to 10, which involved screening patients, providing referral, consultation and public education, and surgical care, were included in the analysis. Roles 1 to 3 were excluded from the model because they could be perceived as including a wide range of activities, including those that were already covered by roles 4 to 10. For example, “providing assessment and treatment traditionally provided by a physical therapist” might include providing public education (role 5), providing consultation with another health care professional (role 6), and providing presurgical and postsurgical care (roles 9 and 10). Osteoarthritis and RA were modeled separately. The Poisson regression model was selected over a linear regression model because of the poor fit of data, especially for RA, in linear regression models. Furthermore, it was selected over binomial models because, in a separate analysis, the Poisson model yielded a lower Akaike’s Information Criterion value, indicating that it was a better fit. Analyses were adjusted for baseline covariates, including sex, age (ⱖ35 years versus ⬍35 years), number of April 2010
Continuing Professional Development and Physical Therapists’ Roles in Arthritis Management
Figure. Survey sampling results. Physical therapists (PTs) might see only patients with osteoarthritis (OA), only patients with rheumatoid arthritis (RA), or both in the previous month.
years since graduation from entrylevel training (ⱕ10 [eg, recent graduates] versus ⬎10), arthritis caseload (⬎40% versus ⱕ40% of patients with OA or RA), and country (Canada versus the Netherlands). The categories were selected based on the categories used in the survey (Appendix 2). We allowed for a possible interaction between participation in arthritis CPD courses and country. We assessed model fit by examining the deviance residual plots and the scaled deviance. Poisson regression coefficients represent effects on log expected role count. If coefficient  represents the effect on log expected role count per unit increase in variable x, then a ␦x change in x additively increases log expected role count by ␦x, or equivalently multiplies the expected role count by exp (␦x) (ie, the expected role count multiplier). The level of statistical significance was set at Pⱕ.05. For the secondary objective, we fit logistic regression models to predict each of the 10 roles assumed by April 2010
physical therapists in OA and RA management, respectively. These models contained the same explanatory variables as the Poisson regression models. The adjusted odds ratio (OR) and 95% confidence interval (CI) were calculated to determine the magnitude of association between previous participation in CPD courses and each specific role.
da⫽224; the Netherlands: physical therapists in arthritis care⫽99, registered physical therapists⫽101) had seen patients with OA in the previous month, and 259 physical therapists (Canada⫽68; the Netherlands: physical therapists in arthritis care⫽101, registered physical therapists⫽90) had seen patients with RA in the previous month (Figure).
Results
Participants’ demographic and practice characteristics are summarized in Table 1. Of those who had seen patients with OA or RA in the previous month, a higher proportion of the Dutch physical therapists in arthritis care (treated patients with OA⫽63.6%; treated patients with RA⫽63.4%) had completed at least one arthritis CPD course compared with the Dutch registered physical therapists (treated patients with OA⫽24.8%; treated patients with RA⫽ 26.7%) or the Canadian physical therapists (treated patients with OA⫽25.9%; treated patients with RA⫽32.4%).
The Canadian survey received 286 completed questionnaires (response rate⫽47.7%). Forty-seven survey packages (7.8%) were returned by the post office. The Dutch survey received 233 replies (overall response rate⫽50.5%; physical therapists in arthritis care: 112/211, response rate⫽53.1%; registered physical therapists: 121/250, response rate⫽48.4%). Four survey packages for the physical therapists in arthritis care (1.9%) and 9 survey packages for the registered physical therapists (3.6%) were returned by the post office. Among the responders, 424 physical therapists (Cana-
Volume 90
Number 4
Physical Therapy f
633
Continuing Professional Development and Physical Therapists’ Roles in Arthritis Management Table 1. Demographic and Practice Characteristicsa Physical Therapists Who Saw Patients With OA, n (%) Canada
The Netherlands
Physical Therapists Who Saw Patients With RA, n (%) Canada
The Netherlands
Physical Therapists in Orthopedic Care (nⴝ224)
Physical Therapists in Arthritis Care (nⴝ99)
Physical Therapists in All Areas of Practice (nⴝ101)
Physical Therapists in Orthopedic Care (nⴝ68)
Physical Therapists in Arthritis Care (nⴝ101)
Physical Therapists in All Areas of Practice (nⴝ90)
157 (70.1)
56 (56.6)
49 (48.5)
50 (73.5)
57 (56.4)
44 (48.9)
67 (29.9)
42 (42.4)
50 (49.5)
18 (26.5)
43 (42.6)
45 (50.0)
0 (0.0)
1 (1.0)
2 (2.0)
0 (0.0)
1 (1.0)
1 (1.1)
20–34
78 (34.8)
21 (21.2)
30 (29.7)
31 (45.6)
21 (20.8)
23 (25.6)
35–49
105 (46.9)
37 (37.4)
41 (40.6)
25 (36.7)
36 (35.6)
37 (41.1)
50–64
38 (17.0)
40 (40.4)
28 (27.7)
11 (16.2)
43 (42.6)
29 (32.2)
Variable Sex Female Male Not stated Age (y)
a
65⫹
3 (1.3)
0 (0.0)
0 (0.0)
1 (1.5)
0 (0.0)
0 (0.0)
Not stated
0 (0.0)
1 (1.0)
2 (2.0)
0 (0.0)
1 (1.0)
1 (1.1)
ⱕ10 y since graduation from entry-level physical therapy training
83 (37.1)
23 (23.2)
28 (27.7)
28 (41.2)
23 (22.8)
23 (25.6)
OA or RA caseload higher than 40% in a typical week
67 (29.9)
16 (16.2)
12 (11.9)
26 (38.2)
16 (15.8)
10 (11.1)
Completed 1 or more post entry-level courses on arthritis
58 (25.9)
63 (63.6)
25 (24.8)
22 (32.4)
64 (63.4)
24 (26.7)
OA⫽osteoarthritis, RA⫽rheumatoid arthritis.
The most common roles reported by participants were to provide traditional physical therapy assessment and treatment and postsurgical management (Tab. 2). Interestingly, 4% of the Canadian physical therapists and 1.5% of the Dutch physical therapists performed tasks outside the scope of physical therapist practice when treating patients with OA. The same was reported by almost 6% of the Canadian physical therapists and 0.5% of the Dutch physical therapists who saw patients with RA. The majority of physical therapists reported assuming ⱕ2 roles when they treated patients with OA or RA (Tab. 3). Poisson regression models demonstrated adequate fit, with deviance 634
f
Physical Therapy
Volume 90
residuals free of patterns against the explanatory variables. Table 4 lists the Poisson regression coefficients and expected role count multipliers. For both the OA and RA models, the interaction term between participation in CPD courses and country was not statistically significant (OA: P⫽.72; RA: P⫽.49); therefore, we dropped the terms in the models. For the OA model, arthritis CPD courses significantly increased (ie, multiplied) the expected number of roles by a factor of 1.32 (95% CI⫽1.11, 1.56) after adjusting for country, personal characteristics, and arthritis caseload. Of the remaining variables, country was a significant predictor, with the Dutch physical therapists showing a lower expected role count compared with
Number 4
those from Canada by a factor of 0.81 (95% CI⫽0.68, 0.97) after adjusting for covariates. For the RA model, CPD courses significantly increased the expected number of roles by a factor of 1.69 (95% CI⫽1.34, 2.13). None of the remaining variables emerged as significant predictors. Table 5 contains the results of logistic regression of arthritis CPD courses on the odds of assuming individual roles after adjusting for the same variables as the role count models. We dropped the nonsignificant interaction in the analyses (P⫽.11–.99). For the management of OA, physical therapists who completed arthritis CPD courses were more likely to refer patients to other rheumatology rehabilitation profesApril 2010
Continuing Professional Development and Physical Therapists’ Roles in Arthritis Management Table 2. Roles in Managing Osteoarthritis and Rheumatoid Arthritis in the Previous Month Reported by Physical Therapistsa Physical Therapists Who Saw Patients With OA, n (%) Canada
Canada
The Netherlands
Physical Therapists in Orthopedic Care (nⴝ224)
Physical Therapists in Arthritis Care (nⴝ99)
Physical Therapists in All Areas of Practice (nⴝ101)
Physical Therapists in Orthopedic Care (nⴝ68)
Physical Therapists in Arthritis Care (nⴝ101)
Physical Therapists in All Areas of Practice (nⴝ90)
Providing assessment and treatment traditionally provided by a physical therapist
217 (96.9)
83 (83.8)
78 (77.2)
64 (94.1)
85 (84.2)
56 (62.2)
Providing assessment and treatment traditionally provided by other rehabilitation disciplines
37 (16.5)
6 (6.1)
5 (5.0)
17 (25.0)
12 (11.9)
1 (1.1)
Providing assessment and treatment that are outside the scope of physical therapist practice
9 (4.0)
1 (1.0)
2 (2.0)
4 (5.9)
1 (1.0)
0 (0.0)
Screening patients for physicians
9 (4.0)
13 (13.1)
14 (13.9)
5 (7.4)
20 (19.8)
11 (12.2)
Providing public education
32 (14.3)
12 (12.1)
7 (6.9)
11 (16.2)
14 (13.9)
6 (6.7)
Providing consultation together with another health care professional
49 (21.9)
22 (22.2)
14 (13.9)
17 (25.0)
28 (27.7)
12 (13.3)
Referring patients to medical professionals
83 (37.1)
18 (18.2)
10 (9.9)
23 (33.8)
41 (40.6)
23 (25.6)
Referring patients to other rheumatology rehabilitation professionals
15 (6.7)
9 (9.1)
2 (2.0)
10 (14.7)
19 (18.8)
11 (12.2)
Providing presurgical care
62 (27.7)
39 (39.4)
31 (30.7)
11 (16.2)
33 (32.7)
17 (18.9)
Providing postsurgical services
155 (69.2)
59 (59.6)
57 (56.4)
25 (36.8)
53 (52.5)
23 (25.6)
Role
a
The Netherlands
Physical Therapists Who Saw Patients With RA, n (%)
OA⫽osteoarthritis, RA⫽rheumatoid arthritis.
sionals (OR⫽8.45; 95% CI⫽2.93, 24.42) and provide presurgical care (OR⫽2.03; 95% CI⫽1.25, 3.29) compared with those who had not completed a course. For the management of RA, physical therapists who completed arthritis courses were more likely to refer patients to other rheumatology rehabilitation professionals (OR⫽4.64; 95% CI⫽1.91, 11.28), provide presurgical care (OR⫽3.98; 95% CI⫽1.91, 8.30), provide public education (OR⫽2.95; 95% CI⫽1.09, April 2010
7.96), provide traditional physical therapy treatment and assessment (OR⫽2.24; 95% CI⫽1.05, 4.78), and provide postsurgical care (OR⫽ 2.12; 95% CI⫽1.15, 3.89) compared with those who had not completed a course. The model for “providing assessment and treatments outside the scope of physical therapist practice” did not converge due to a sparse distribution, especially in the Dutch data (Tab. 2); therefore, logistic regression was not conducted.
Discussion This bi-nation study provides novel data for understanding the value of arthritis CPD courses and the roles of physical therapists in the management of OA and RA. Our analysis supports the hypothesis that participation in arthritis CPD activities is associated with a higher average number of roles assumed by physical therapists. In a systematic review, Davis et al42 concluded that interactive and mixed didactic/interactive
Volume 90
Number 4
Physical Therapy f
635
Continuing Professional Development and Physical Therapists’ Roles in Arthritis Management Table 3. Distribution of the Total Number of Rolesa Assumed by Physical Therapists Who Saw Patients With Osteoarthritis or Rheumatoid Arthritis in the Previous Monthb Physical Therapists Who Saw Patients With OA, n (%) Canada
Physical Therapists Who Saw Patients With RA, n (%)
The Netherlands
Canada
The Netherlands
No. of Roles Assumed by Physical Therapists
Physical Therapists in Orthopedic Care (nⴝ224)
Physical Therapists in Arthritis Care (nⴝ99)
Physical Therapists in All Areas of Practice (nⴝ101)
Physical Therapists in Orthopedic Care (nⴝ68)
Physical Therapists in Arthritis Care (nⴝ101)
Physical Therapists in All Areas of Practice (nⴝ90)
0
32 (14.3)
26 (26.3)
35 (34.7)
20 (29.4)
29 (28.7)
49 (54.4)
1
73 (32.6)
22 (22.2)
22 (21.8)
23 (33.8)
22 (21.8)
11 (12.2)
2
59 (26.3)
26 (26.3)
27 (26.7)
11 (16.2)
16 (15.8)
11 (12.2)
3
35 (15.6)
10 (10.1)
10 (9.9)
5 (7.4)
8 (7.9)
9 (10.0)
4
17 (7.6)
10 (10.1)
6 (5.9)
5 (7.4)
9 (8.9)
9 (10.0)
5
7 (3.1)
3 (3.0)
1 (1.0)
3 (4.4)
9 (8.9)
1 (1.1)
6
1 (0.5)
1 (1.0)
0 (0.0)
0 (0.0)
7 (6.9)
0 (0.0)
7
0 (0.0)
1 (1.0)
0 (0.0)
1 (1.4)
1 (1.0)
0 (0.0)
a
Roles included in the analysis: screening patients for physicians, providing public education, providing consultation together with another health care professional, referring patients to medical professionals, referring patients to other rheumatology rehabilitation professionals, providing presurgical care, and providing postsurgical services. b OA⫽osteoarthritis, RA⫽rheumatoid arthritis.
continuing education sessions significantly improved health professional practice. Our findings provide further evidence that CPD activities may enhance physical therapists’ roles in arthritis care. It should be noted that CPD activities are different from guideline implementation interventions. The former are usually initiated by the learner, with the main goal of maintaining competen-
cies and improving clinical performance.43 Guideline implementation interventions, however, have been described as the “change agent in the health care system,”43(p6) with the main goal of closing gaps in care based on the best evidence. One example of guideline implementation intervention is the use of a social marketing campaign to change health care professionals’ beliefs and
practices in acute low back pain management.44 – 46 In this context, the health care professional is only one component within a mix of factors, including his or her interaction with patients and peers, organizational support, availability of health care system resources, and health policy.43 This complexity may explain why CPD courses, which focus mainly on the clinician’s knowledge
Table 4. Poisson Regression Coefficients and Expected Physical Therapy Role Count Multipliers With 95% Confidence Intervalsa OA Role Count Model Coefficient (95% CI)
Variable Completed post⫺entry-level arthritis course The Netherlands (vs Canada)
0.27 (0.10, 0.44)
Coefficient (95% CI)
Multiplier (95% CI)
1.32b (1.11, 1.56)
0.53 (0.29, 0.76)
1.69 (1.34, 2.13)
⫺0.21 (⫺0.38, ⫺0.03)
0.81 (0.68, 0.97)
⫺0.10 (⫺0.36, 0.17)
0.91 (0.70, 1.19)
0.00 (⫺0.17, 0.16)
1.00 (0.84, 1.18)
⫺0.01 (⫺0.25, 0.22)
0.99 (0.78, 1.25)
Female Age ⱖ35 y
RA Role Count Model
Multiplier (95% CI)
⫺0.04 (⫺0.38, 0.29)
0.96 (0.68, 1.34)
0.35 (⫺0.38, 1.08)
1.42 (0.69, 2.94)
Recent entry-level graduate (ⱕ10 y)
0.03 (⫺0.30, 0.36)
1.03 (0.74, 1.43)
0.19 (⫺0.53, 0.91)
1.21 (0.59, 2.49)
High arthritis caseload (⬎40%)
0.17 (⫺0.01, 0.34)
1.18 (0.99, 1.41)
0.21 (⫺0.05, 0.47)
1.23 (0.95, 1.60)
a
OA⫽osteoarthritis, RA⫽rheumatoid arthritis, CI⫽confidence interval. The model suggests that physical therapists who have completed at least one post– entry-level arthritis course assume 1.32 times more OA roles than those who have not completed any post– entry-level arthritis courses, after accounting for country, sex, age, years after graduation, and arthritis caseload. b
636
f
Physical Therapy
Volume 90
Number 4
April 2010
Continuing Professional Development and Physical Therapists’ Roles in Arthritis Management Table 5. Odds Ratios of Completing Arthritis Continuing Professional Development Courses for Predicting Physical Therapists’ Roles in Managing Osteoarthritis and Rheumatoid Arthritisa OR (95% CI) Predicting OA Role
OR (95% CI) Predicting RA Role
Providing assessment and treatment traditionally provided by a physical therapist
1.80 (0.82, 3.98)
2.24 (1.05, 4.78)
Providing assessment and treatment traditionally provided by other rehabilitation disciplines
1.24 (0.60, 2.57)
0.62 (0.22, 1.75)
Providing assessment and treatment that are outside the scope of physical therapist practice
0.37 (0.04, 3.24)
DNC
Screening patients for physicians
1.02 (0.44, 2.32)
1.49 (0.61, 3.66)
Providing public education
1.31 (0.65, 2.63)
2.95 (1.09, 7.96)
Providing consultation together with another health professional
1.26 (0.72, 2.18)
1.12 (0.56, 2.24)
Referring patients to medical professionals
1.72 (0.99, 2.98)
1.38 (0.74, 2.56)
Referring patients to other rheumatology rehabilitation professionals
8.45 (2.93, 24.42)
4.64 (1.91, 11.28)
Providing presurgical care
2.03 (1.25, 3.29)
3.98 (1.91, 8.30)
Providing postsurgical services
1.17 (0.73, 1.89)
2.12 (1.15, 3.89)
Role
a
Odds ratios were adjusted for country, sex, age, recent entry-level graduate, and arthritis caseload. OR⫽odds ratio, OA⫽osteoarthritis, RA⫽rheumatoid arthritis, DNC⫽did not converge.
and skills, contribute limited benefits in changing clinical practice and patient outcomes when used as a tool for guideline implementation,47,48 but are associated with a larger number of roles assumed by physical therapists in the management of arthritis, as indicated in our findings. Our results also highlight a few trends regarding physical therapists’ roles in the management of OA and RA. First, although the vast majority of participants continued to provide traditional physical therapy treatments, a small proportion of physical therapists reported providing treatments that were outside the traditional scope of practice. This finding may reflect physical therapists’ growing interests in advanced practice roles26 and the increasing support of clinical facilities for physical therapists to work in these roles.15,23,37,49 Second, our results showed that participation in CPD was associated with physical therapists’ referring April 2010
patients to other rheumatology rehabilitation professionals and the provision of presurgical care to patients with OA or RA. Furthermore, those who completed CPD courses were more likely to have provided RArelated physical therapy interventions, postsurgical care, and public education. These findings reflect the similarities and differences in the management of OA and RA. Osteoarthritis is a chronic joint disease with hallmarks including cartilage degeneration, joint pain, and stiffness after prolonged inactivity. Most people with OA can be effectively treated with interventions such as therapeutic exercise, braces, orthoses, and weight management strategies.50,51 For those with severe disease, joint replacement surgery may be required. Physical therapists, in general, have received in-depth training in exercise prescription. In a separate analysis, we found 77.5% of Canadian physical therapists said that their entry-level training adequately covered exercise prescription for OA management, but far fewer re-
ported the same for the assessment and prescription of hand orthoses (9.1%), knee braces (17.7%), and foot orthoses (13.8%).34 It was likely that CPD courses emphasized the use of these interventions and consequently prompted therapists to refer patients to other rehabilitation professionals when they are required. For RA, one of the common chronic systemic inflammatory joint diseases, current guidelines emphasize early medical treatment (ie, the use of disease-modifying anti-rheumatic drugs within 3 months of symptom onset) combined with nonpharmacological interventions, as needed.52 The latter interventions may include exercise, patient education, thermotherapy, and vocational counseling,52 some of which are considered treatments traditionally provided by physical therapists. Results from our study showed that CPD courses might increase the odds of physical therapists providing these treatments to patients with RA. It was
Volume 90
Number 4
Physical Therapy f
637
Continuing Professional Development and Physical Therapists’ Roles in Arthritis Management also encouraging that those who participated in CPD courses were more likely to provide public education. Given the recent research indicating delays in seeking help for initial symptoms of RA,53 it is important that physical therapists contribute to increasing the public’s awareness about arthritis and the importance of early treatments. To meet the challenges of the increasing number of people with arthritis forecast for the next 20 years, a significant amount of work has been done to improve the knowledge and skills of health care professionals who deliver arthritis care. In the United Kingdom, standards for entry-level rheumatology curricula have been developed in nursing, physical therapy, and occupational therapy to ensure that students receive adequate rheumatology content.54 Similar work also has been initiated for medical students in Canada.55 However, although these initiatives are essential for new practitioners, they do not address the continuing learning needs of those already in clinical practice. We argue that there is a need to direct resources to develop arthritis-related CPD activities. In a separate analysis of the Canadian survey,34 we identified important discrepancies in the entry-level rheumatology education received by physical therapists. Areas that were inadequately covered included the assessment of active and damaged joints for RA; back assessment for ankylosing spondylitis; and the assessment and prescription of assistive devices, braces, and orthoses.34 Furthermore, only 19% were satisfied with what they learned about community resources for patients, and only 16% were satisfied with the coverage of professional resources for arthritis management. Yet these skills are essential when providing arthritis care. Currently, physical therapists’ roles in arthritis management concentrate on 638
f
Physical Therapy
Volume 90
the period after diagnosis; however, there is an opportunity to expand these roles to the stage before diagnosis, where physical therapists could be the first point of contact for assessment and could facilitate appropriate management by primary care physicians.39,56 Further development of standards for arthritis CPD activities would serve as the first step for enhancing the roles of physical therapists to serve this population. There are several caveats about the interpretation of these results. First, a portion of physical therapists were drawn from local arthritis networks in the Netherlands but not in Canada. The differences in sampling strategy have inflated the proportion of physical therapists working with a primary focus in arthritis and the amount of CPD taken in the Netherlands. Because of the sampling bias, we recommend against any direct comparison of the practice characteristics between the 2 countries. Second, physical therapists’ roles in OA and RA management were obtained through self-reporting, which might differ from their actual clinical practice. Furthermore, we did not ask therapists to indicate other roles in arthritis management; thus, the list might not capture all the roles currently assumed by physical therapists in the 2 countries. Third, about 50% of the physical therapists did not return the questionnaire, and so the findings may be subjected to response bias. In the Canadian survey, because the information of the selected physical therapists registered with the College of Physical Therapists of Alberta was not available, we were unable to evaluate whether the responders and nonresponders were systematically different. However, when compared with Canadian physical therapists in the workforce in 2007, respondents to the Canadian survey appeared to be younger (those less
Number 4
than 35 years of age⫽34.8%– 45.6% versus 32.3% in the general physical therapist population57) and have a lower proportion of women (70.1%– 73.5% versus 78.7% in the general physical therapist population57). Similar information was not available for a comparison with the Dutch physical therapists. It should be noted that 7.8% of the questionnaires were returned by the postal office in Canada. The response rate might have been higher if these individuals were replaced. Nonetheless, our response rate is comparable to those of other recent physical therapy surveys on practice patterns in North America, with a response rate between 36% and 41%.58,59 Fourth, although we were able to identify therapists who completed arthritis-related CPD courses, we did not know the content of the courses. This was a potential concern for the Canadian survey because the available arthritis courses ranged from short workshops to extensive advanced practice training programs. However, because fewer than 1% of orthopedic physical therapists have completed these extensive training programs, we believe that our findings reflect mainly the roles of those who have completed the shorter courses. Finally, due to differences in health care systems and physical therapist practices across countries, the results may not be directly applicable to jurisdictions outside of Canada and the Netherlands. Nonetheless, our findings may be relevant to health care professionals in the United States, as that country is undergoing health care reform, which may affect the future roles of physical therapists in the management arthritis and other chronic diseases. We recognize that it is considered illegal in the United States for physical therapists to practice outside the scope of physical therapy. On the April 2010
Continuing Professional Development and Physical Therapists’ Roles in Arthritis Management other hand, the profession is beginning to explore issues such as direct access and advanced scope of practice, and their impact on patient care. The recent Summit on Direct Access and Advanced Scope of Practice, co-hosted by the American Physical Therapy Association and the Canadian Physiotherapy Association, has marked the beginning of this endeavor.60 As the roles of physical therapists continue to evolve across countries, we believe that the current study can contribute to the important discussion about physical therapists’ roles and scope of practice by providing evidence about the relationship between CPD activities and the roles in arthritis care.
Conclusion This exploratory study demonstrated the association between arthritis CPD courses and the roles assumed by physical therapists in arthritis care in Canada and the Netherlands. Although a direct causal inference cannot be made, the results may inform health care administrators’ decisions about staff requests to attend CPD courses. As with entry-level physical therapy training, standardized rheumatology CPD curricula for practicing physical therapists will help meet their continuing professional development needs. We recommend that future research should focus on evaluating the effects of CPD on other areas of physical therapist practice and on patients’ outcomes. Dr Li provided concept/idea/research design and fund procurement. Dr Li, Dr Sayre, and Dr Vliet Vlieland provided writing. Dr Li, Ms Hurkmans, and Dr Vliet Vlieland provided data collection. Dr Li and Dr Sayre provided data analysis. Dr Vliet Vlieland provided project management and consultation (including review of manuscript before submission). The authors thank all physical therapy regulatory colleges in Canada, the Royal Dutch Society for Physical Therapy, and the physical therapists who participated in the survey.
April 2010
They also thank the Physiotherapy Foundation of Canada and the Arthritis Health Professions Association for providing financial support for the project and Dr John Verhoef for his assistance with the Dutch survey. Dr Li was supported by a Canadian Institutes of Health Research New Investigator Award and an American College of Rheumatology Research & Education Foundation Health Professional New Investigator Award. Dr Sayre was supported by trainee awards from the Natural Sciences and Engineering Research Council of Canada and the Michael Smith Foundation for Health Research. This article was received December 21, 2008, and was accepted December 28, 2009. DOI: 10.2522/ptj.20080409
References 1 Woolf AD, Akesson K. Understanding the burden of musculoskeletal conditions. BMJ. 2001;322:1079 –1080. 2 Arthritis in Canada: An Ongoing Challenge. H38 – 4/14 –2003E ed. Ottawa, Ontario, Canada: Health Canada; 2003. 3 Koehn C, Dooley A, Hofstetter C, Qualman A. Determining the research priorities of people living with arthritis: listening to patients. The Arthritis Society. 2006. Available at: http://www.arthritis.ca/ look%20at%20research/cap/patient/ default.asp?s⫽1. 4 Li LC, MacKay C, CARE III Local Planning Committee. CARE III Online Patient Survey—Summary of Preliminary Analysis. 2005. Available at: www.arthritis.ca/look %20at%20research/surveys/caresummary/ default.asp?s⫽1. 5 Heiberg T, Finset A, Uhlig T, Kvien TK. Seven-year changes in health status and priorities for improvement of health in patients with rheumatoid arthritis. Ann Rheum Dis. 2005;64:191–195. 6 Bury M. Meanings at risk: the experience of arthritis. In: Anderson R, Bury M, eds. Living With Chronic Illness: The Experience of Patients and Their Families. London, United Kingdom: Unwin and Hyman; 1988:89 –116. 7 Centers for Disease Control and Prevention. Public health and aging: projected prevalence of self-reported arthritis or chronic joint symptoms among persons aged ⬎65 years—United States, 2005–2030. MMWR. 2003;52:489 – 491. 8 Badley EM, Wang PP. Arthritis and the aging population: projections of arthritis prevalence in Canada 1991 to 2031. J Rheumatol. 1998;25:138 –144. 9 Poos MJJC, Gijsen R. Neemt het Aantal Mensen met Artrose Toe Of Af? Volksgezondheid Toekomst Verkenning, Nationaal Kompas Volksgezondheid. Bilthoven, the Netherlands: Rijksinstituut voor Volksgezondheid en Milieu; 2002.
10 van der Linden SJ, Poos MJJC. Neemt het Aantal Mensen met RA Toe Of Af? Volksgezondheid Toekomst Verkenning, Nationaal Kompas Volksgezondheid. Bilthoven, the Netherlands: Rijksinstituut voor Volksgezondheid en Milieu; 2007. 11 Chorus AMJ, van Overbeek K, HopmanRock M. Reumatische Klachten in Nederland: Resultaten Nationale Peiling van het Bewegingsapparaat 2006. Leiden, the Netherlands: TNO; 2007. 12 Submission to Commission on the Future of Health Care in Canada. 2001. Available at: http://www.hc-sc.gc.ca/english/care/ romanow/hcc0148.html. 13 Canadian Medical Directory. Ottawa, Ontario, Canada: Canadian Medical Association; 2009. 14 Li LC, Davis AM, Lineker S, et al. Effectiveness of the primary therapist model for rheumatoid arthritis rehabilitation: a randomized controlled trial. Arthritis Rheum. 2006;55:42–52. 15 Campos AA, Graveline C, Ferguson JM, et al. The physical therapy practitioner (PTP) in pediatric rheumatology: high level of patient and parent satisfaction with services. Physiother Can. 2002;54: 32–36. 16 O’Cathain A, Froggett M, Taylor MP. General practice based physiotherapy: its use and effect on referrals to hospital orthopaedics and rheumatology outpatient departments. Br J Gen Pract. 1995;45:352– 354. 17 Hourigan PG, Weatherley CR. Initial assessment and follow-up by a physiotherapist of patients with back pain referred to a spinal clinic. J R Soc Med. 1994;87:213– 214. 18 Hourigan PG, Weatherley CR. The physiotherapist as an orthopaedic assistant in a back pain clinic. Physiotherapy. 1995;81: 546 –548. 19 Helewa A, Smythe HA, Goldsmith CH, et al. The total assessment of rheumatoid polyarthritis: evaluation of a training program for physiotherapists and occupational therapists. J Rheumatol. 1987;14: 87–92. 20 Stokes BA, Helewa A, Lineker SC. Total assessment of rheumatoid polyarthritis: a postgraduate training program for physical and occupational therapists: a 20 year success story. J Rheumatol. 1997;24: 1634 –1638. 21 Lineker SC, Wood H, Badley EM, et al. Evaluation of the Primary Therapist Model of Service Delivery as Implemented by The Arthritis Society, Consultation and Rehabilitation Service, Phase 1: Therapist Survey. Working Paper 98 –5 ed. Toronto, Ontario, Canada: Arthritis Community Research & Evaluation Unit; 1998. 22 Hurst K. Multi-skilled health carers: nature, purpose and implications. Health Manpow Manage. 1997;23:197–211. 23 Campos AA, Graveline C, Ferguson JM, et al. The physical therapy practitioner: an expanded role for physical therapy in pediatric rheumatology. Physiother Can. 2001;53:282–287.
Volume 90
Number 4
Physical Therapy f
639
Continuing Professional Development and Physical Therapists’ Roles in Arthritis Management 24 Arthritis Continuing Education (ACE) Program. Available at: www.argbc.ca/ practitioner/ace-program. 25 Advanced Clinician Practitioner in Arthritis Care (ACPAC) Program. Available at: www.stmichaelshospital.com/programs/ mobility/acpac.php. 26 Lundon K, Shupak R, Sunstrum-Mann L, et al. Leading change in the transformation of arthritis care: development of an interprofessional academic-clinical education training model. Healthcare Quarterly. 2008;11:59 – 65. 27 A novel pilot project: development of an educational interdisciplinary postgraduate academic and clinical training program in arthritis care for experienced physical and occupational therapists. Presented at CARE IV Conference; 2006; Leeds, England. Available at: www.leeds.ac.uk/ CAREIV/Posters/Rachel%20Shupak.pdf. 28 Dutch Institute of Allied Health Care. Available at: http://www.paramedisch. org/english.html. 29 Royal Dutch Society for Physical Therapy. Available at: www.kngf.nl. 30 Nederlands-Vlaamse Accreditatieorganisatie. Available at: www.nvao.nl. 31 Bell MJ, Lineker SC, Wilkins A, et al. A randomized controlled trial to evaluate the efficacy of community based physical therapy in the treatment of people with rheumatoid arthritis. J Rheumatol. 1998;25: 231–237. 32 Rainforth B. The primary therapist model: addressing challenges to practice in special education. Phys Occup Ther Pediatr. 2002;22:29 –51. 33 Li LC, Maetzel A, Davis AM, et al. Primary therapist model for patients referred for rheumatoid arthritis rehabilitation: a costeffectiveness analysis. Arthritis Rheum. 2006;55:402– 410. 34 Li LC, Westby MD, Sutton E, et al. Canadian physiotherapists’ views on certification, specialisation, extended role practice, and entry-level training in rheumatology. BMC Health Serv Res. 2009;9:88. 35 Position statement: specialisation. World Confederation for Physical Therapy. 2008. Available at: http://www.wcpt.org/node/ 29536. Accessed May 25, 2009. 36 Chartered Physiotherapists Working as Extended Scope Practitioners (ESP): Guidance for Members. London, United Kingdom; The Chartered Society of Physiotherapy; 2003. 37 Robarts S, Kennedy D, MacLeod AM, et al. A framework for the development and implementation of advanced practice roles for physiotherapists that improves access and quality of care for patients. Healthcare Quarterly. 2008;11:67–75.
640
f
Physical Therapy
Volume 90
38 Hurkmans EJ, Li LC, Huizinga TW, Vliet Vlieland TPM. Exercise therapy in rheumatoid arthritis: How confident are physical therapists regarding their knowledge and skills? Arthritis Rheum. 2008;58(9 suppl): S598. 39 Li LC, Badley EM, MacKay C, et al. An evidence-informed, integrated framework for rheumatoid arthritis care. Arthritis Rheum. 2008;59:1171–1183. 40 Dillman DA. Mail and Telephone Surveys: The Total Design Method. New York, NY: John Wiley & Sons Inc; 1978. 41 Salant P, Dillman DA. How to Conduct Your Own Survey. New York, NY: John Wiley & Sons Inc; 1989. 42 Davis D, O’Brien MA, Freemantle N, et al. Impact of formal continuing medical education: Do conferences, workshops, rounds, and other traditional continuing education activities change physician behavior or health care outcomes? JAMA. 1999;282:867– 874. 43 Davis DA. Continuing education, guideline implementation, and the emerging transdisciplinary field of knowledge translation. J Contin Educ Health Prof. 2006;26:5–12. 44 Buchbinder R, Jolley D, Wyatt M. Population-based intervention to change back pain beliefs and disability: three-part evaluation. BMJ. 2001;322:1516 –1520. 45 Buchbinder R, Jolley D. Population-based intervention to change back pain beliefs: three-year follow-up population survey. BMJ. 2004;328:321. 46 Buchbinder R, Jolley D. Effects of a media campaign on back beliefs is sustained 3 years after its cessation. Spine. 2005;30: 1323–1330. 47 Davis D. Does CME work? An analysis of the effect of educational activities on physician performance or health care outcomes. Int J Psychiatr Med. 1998;28:21– 39. 48 Grimshaw JM, Thomas RE, MacLennan G, et al. Effectiveness and efficiency of guideline dissemination and implementation strategies. Health Technol Assess. 2004;8: iii–iiv. 49 Aiken AB, Harrison MM, Atkinson M, Hope J. Easing the burden for joint replacement wait times: the role of the extended practice physiotherapists. Healthcare Quarterly. 2008;11:62– 66. 50 Recommendations for the medical management of osteoarthritis of the hip and knee: 2000 update. American College of Rheumatology Subcommittee on Osteoarthritis Guidelines. Arthritis Rheum. 2000; 43:1905–1915.
Number 4
51 Porcheret M, Jordan K, Croft P; for the Primary Care Rhumatology Society. Treatment of knee pain in older adults in primary care: development of an evidencebased model of care. Rheumatology. 2007;46:638 – 648. 52 American College of Rheumatology Subcommittee on Rheumatoid Arthritis Guidelines. Guidelines for the management of rheumatoid arthritis: 2002 update. Arthritis Rheum. 2002;46:328 –346. 53 Potter T, Mulherin D, Pugh M, et al. Early intervention with disease-modifying therapy for rheumatoid arthritis: where do the delays occur? Rheumatology. 2002;41: 953–955. 54 Hewlett S, Clarke B, O’Brien A, et al. Rheumatology education for undergraduate nursing, physiotherapy and occupational therapy students in the UK: standards, challenges and solutions. Rheumatology. 2008;47:1025–1030. 55 Wadey VMR, Tang ET, Abelseth G, et al. Canadian multidisciplinary core curriculum for musculoskeletal health. J Rheumatol. 2006;34:567–580. 56 MacKay C, Veinot P, Badley EM. Characteristics of evolving models of care for arthritis: a key informant study. BMC Health Serv Res. 2008;8(147). Available at: http:// www.biomedcentral.com/1472– 6963/8/ 147. 57 Workforce Trends of Physiotherapists in Canada, 2007. Ottawa, Ontario, Canada: Canadian Institute for Health Information; 2008. 58 Pantano KJ. Strategies used by physical therapists in the U.S. for treatment and prevention of the female athlete triad. Phys Ther Sport. 2009;10:3–11. 59 Natarajan P, Oelschlager A, Agah A, et al. Current clinical practices in stroke rehabilitation: regional pilot survey. J Rehabil Res Dev. 2008;45:841– 849. 60 American Physical Therapy Association News Release. International Summit Reaches Agreement: Patient Self-Referral to Physical Therapy Improves Public Health. Available at: http://www.apta. org/AM/Template.cfm?Section⫽Media &Template⫽/CM/ContentDisplay.cfm& ContentID⫽65055.
April 2010
Continuing Professional Development and Physical Therapists’ Roles in Arthritis Management Appendix 1. Questions From the Canadian Physiotherapist Arthritis Care Survey About Participants’ Roles in the Management of Osteoarthritis (OA) and Rheumatoid Arthritis (RA)a 11a. Did you see any patient with OA in the previous month? 䡺 Yes (Go to OA column)
䡺 No (Go to 11b)
b. Did you see any patient with RA in the previous month? 䡺 Yes (Go to RA column)
䡺 No (Go to 12)
What was your role when you saw people with OA or RA in the previous month? (Check all that apply)
a
OA
RA
1. Provide assessment and treatment that are traditionally provided by a PT
䡺
䡺
2. Provide assessment and treatment that are traditionally provided by other rehabilitation disciplines (eg, occupational therapy interventions)
䡺
䡺
3. Provide assessment and treatment that are outside the scope of physical therapist practice (eg, ordering investigative tests, providing injection)
䡺
䡺
4. Screen patients and help schedule priority appointments for physicians
䡺
䡺
5. Public education
䡺
䡺
6. Provide consultation together with another health care professional (eg, PT, OT, family physician)
䡺
䡺
7. Refer patients to medical professionals (eg, family physician, rheumatologist)
䡺
䡺
8. Refer patients to other rheumatology rehabilitation professionals
䡺
䡺
9. Presurgical care
䡺
䡺
10. Postsurgical care
䡺
䡺
PT⫽physical therapist, OT⫽occupational therapist.
April 2010
Volume 90
Number 4
Physical Therapy f
641
Continuing Professional Development and Physical Therapists’ Roles in Arthritis Management Appendix 2. Questions From the Canadian Physiotherapist Arthritis Care Survey About Personal and Practice Characteristicsa This section contains questions about the general characteristics of your practice. Please check the appropriate box or fill in the blank as required. 1.
Are you practicing clinically? 䡺 Yes
3
䡺 Full-time
䡺 Part-time
䡺 No (including maternity leave) 3 2.
Go to Question 17
In an average work week, how many patients do you see? _____ Patients per week
3.
4.
What is your primary area of practice? (Check one only) 䡺
Orthopedics Number of years in orthopedics: ________
䡺
Rheumatology Number of years in rheumatology: _______
䡺
Other (Please specify: ___________) Number of years in this area: ___________
In a typical 1-week period, what percentage of patients do you see primarily for osteoarthritis? 䡺 䡺 䡺 䡺 䡺
5.
81%–100% 61%–80% 41%–60% 21%–40% 20% or less
In a typical 1-week period, what percentage of patients do you see primarily for rheumatoid arthritis? 䡺 䡺 䡺 䡺 䡺
81%–100% 61%–80% 41%–60% 21%–40% 20% or less
This section contains questions about your background. The information will be used for data analysis only. Please check the appropriate box or fill in the blank as required. 19.
What is your age? 䡺 䡺 䡺 䡺
20.
Please indicate your gender. 䡺
21. 22.
Entry-level PT (Baccalaureate degree or Diploma) Entry-level PT (Clinical Master degree) Entry-level OT (Baccalaureate degree or Diploma) Entry-level OT (Clinical Master degree) Thesis-based Master degree PhD/DSc Other (Please specify: _________)
Yes No
Are you a member of the Arthritis Health Professions Association (Canada)? Yes No
Are you a member of the Association of Rheumatology Health Professionals (USA)? 䡺 䡺
a
Male
Are you a member of the Canadian Physiotherapy Association Orthopaedic Division?
䡺 䡺 25.
䡺
Please list the degree(s) you have received. (Check all that apply)
䡺 䡺 24.
Female
In what year did you graduate from the entry-level physical therapy program? _________
䡺 䡺 䡺 䡺 䡺 䡺 䡺 23.
20–34 35–49 50–64 ⬎65
Yes No
PT⫽physical therapist, OT⫽occupational therapist.
642
f
Physical Therapy
Volume 90
Number 4
April 2010
Letters to the Editor On “Some factors predict successful short-term outcomes…” Mintken PE, Cleland JA, Carpenter KJ, et al. Phys Ther. 2010;90:26–42. We would like to comment on the article by Mintken et al1 that was published in the January 2010 issue of PTJ. We need clinical articles that validate practice; however, the results of this study will likely never be validated. The trouble comes from the improper use of stepwise regression, specifically overfitting of the predictor variables (k). Overfitting is not a minor problem in the literature; it is a very serious one. Overfitting creates a model that future studies will not likely validate. So far, we have not confronted the problem of overfitting in physical therapy literature very well. And, when we have admitted to the problem of overfitting in discussing study limitations, we have not explained why it can be such a nuisance when interpreting results and what can be done to prevent overfitting in the future. In their study, Mintken et al started off using 46 different baseline predictor variables (Tabs. 1 and 2). These initial 46 variables then were reduced, using multiple t tests or chi-square tests, to 14 variables that were placed into the stepwise regression, which reduced them down to the final 5 predictor variables. This method of reducing the baseline predictor variables capitalizes on chance not just 1 time but 2 different times, thus bypassing the normal rules of probability and increasing the likelihood of a type I error.2,3 The first capitalization on chance occurs when t tests and chi-square tests are performed to
April 2010
Letter 4.10.indd 643
narrow the baseline variable list by determining which predictors have the best relationship with the dependent variable.4 An initial analysis that uses multiple tests is not recommended for pre-variable selection, because potentially important variables can wrongly be rejected.5 If the researcher insists on using P value stopping rules, a higher rejection value (eg, using .50 instead .10) may help solve this potential problem.3 The second capitalization on chance occurs when the researcher discards the earlier 32 variables (those that were not highly related to the dependent variable) and uses only the variables that were significant. The computer-selected 14 variables then are entered into the regression model.3
participants.3 The recommended ratio of number of participants (n) to predictor variables (k) when running a normal (simultaneous or forced) regression usually is 15–20/1.7 When running a stepwise regression, most statistical experts recommend an n/k value of no less than 40/1.2.7 For logistic regression, which Mintken and colleagues used in their study, Peduzzi et al6 advocated an events-per-variable (EPV) value greater than 10, where the EPV value is the number of successful outcomes versus nonsuccessful outcomes divided by the number of predictor variables. Peduzzi et al showed that using EPV values less than 10 often creates major problems, such as biased regression coefficients with wide confidence intervals.
Not using all of the previous independent variables that were considered in the beginning (k=46) of the study creates what Harrell3 called “phantom” degrees of freedom. The problem is that these phantom degrees of freedom usually are forgotten when computing the standard errors and P values, resulting in data that often are too good to be true.2,3 Cohen et al2 recommended that all predictor variables considered for inclusion at the beginning of a study also should be included when running the regression model, meaning not just the variables that were selected by the computer program. A common problem that often occurs later when computing regression coefficients is that the phantom degrees of freedom are forgotten, resulting in overfitting of the data.2–7 Overfitting occurs when a researcher uses too many predictor variables (46 in this study) with too few study
In the study by Mintken et al, the number of successful outcomes was 31, and the number of baseline predictors was 46. The EPV value is 31/46=.67, suggesting that the validity of this model is very much in question. In many of the cases of overfitting, the culprit is the forgotten phantom degrees of freedom, where the researcher does not take into account all of the predictor variables from the very beginning of the study (46 in this study). Not taking these phantom degrees of freedom into account often will create a number-to-predictor (n/k) ratio that is highly inflated, thereby making the ratio look better than it really is.3 Overfitting can create significant validity problems, in that the results of the study may be valid only for the particular sample that was used.3 Therefore, in studies where a clinical prediction rule is developed, care must be taken when applying rules
Volume 90 Number 4 Physical Therapy ■ 643
3/15/10 5:34 PM
Letters to the Editor from overfitted samples to the population. A way to solve the problem of overfitting before the research is performed is to do a thorough review of the literature and use only those predictors that have a strong theoretical and clinical grounding for their use. If this approach is not possible, the researcher can try to reduce the number of variables by using reduction methods such as factor analysis or principal components or, if possible, by adding more participants. The bottom line is that the method used by the authors in this study has almost zero probability of finding the “right” predictors. Michael T. Cibulka, Frank E. Harrell Jr M.T. Cibulka, PT, DPT, MHS, OCS, is Assistant Professor, Physical Therapy Program, Maryville University, St Louis, MO 63141. Address all correspondence to Dr Cibulka at:
[email protected]. F.E. Harrell Jr, is Professor of Biostatistics and Chair, Department of Biostatistics, Vanderbilt University School of Medicine, Nashville, Tennessee. This letter was posted as a Rapid Response on January 25, 2010, at ptjournal.apta.org.
References 1 Mintken PE, Cleland JA, Carpenter KJ, et al. Some factors predict successful shortterm outcomes in individuals with shoulder pain receiving cervicothoracic manipulation: a single-arm trial. Phys Ther. 2010;90:26–42. 2 Babyak MA. What you see may not be what you get: a brief, nontechnical introduction to overfitting in regression-type models. Psychosom Med. 2004;66:411–421. 3 Cohen J, Cohen P, West S, Aiken L. Applied Multiple Regression/Correlation Analysis for the Behavioral Sciences. 3rd ed. Mahwah, NJ: Lawrence Erlbaum Associates; 2003. 4 Harrell F. Regression Modeling Strategies With Application to Linear Models, Logistic Regression and Survival Analysis. New York, NY: Springer-Verlag; 2001. 5 Nick TG, Hardin JM. Regression modeling strategies: an illustrative case study from medical rehabilitation outcomes research. Am J Occup Ther. 1999;53:459–470.
6 Peduzzi P, Concato J, Kemper E, et al. A simulation study of the number of events per variable in logistic regression analysis. J Clin Epidemiol. 1996;49:1373–1379. 7 Tabachnick B, Fidell L. Using Multivariate Statistics. 5th ed. Boston, MA: Pearson Education Inc; 2007. [DOI: 10.2522/ptj.2010.90.4.643]
Author Response We thank Cibulka and Harrell for their response1 to our article.2 As Mark Twain wrote in his autobiography, “Figures often beguile me, particularly when I have the arranging of them myself; in which case the remark attributed to Disraeli would often apply with justice and force: ‘There are three kinds of lies: lies, damned lies, and statistics.’”3 Cibulka and Harrell have raised some valid points that we would like to address in this response. They state that the “results of this study will likely never be validated.” Although it is true that many of the physical therapy research studies that utilize this design are never carried to the next level of validation,4–14 one study validated the findings of a physical therapy derivation study using this same design.15,16 We emphasize that our study is simply the first stage in a stepwise process described by McGinn et al.17 They describe 3 progressive steps in identifying factors that predict a likely response to treatment in a patient, and it is not until the predictive factors have been validated in a new clinical setting that they become appropriate for implementation. The first step involves identification of the predictive factors, the second step involves the validation of the predictive factors, and the final step involves assessing the impact of using these predictive factors on clinical practice.17 Cibulka and
644 ■ Physical Therapy Volume 90 Number 4
Letter 4.10.indd 644
Harrell are correct to point out the need for the very cautious interpretation of this type of preliminary research, as validation is the critical next step to follow-up on a study such as ours. Having said that, for Cibulka and Harrell to state that “the method used by the authors in this study has almost zero probability of finding the ‘right’ predictors” is a bit overreaching. We agree that prospective validation of our findings is important, because individual patient variables and their interrelationships may differ among different populations. We are in the planning stages of developing a prospective randomized controlled trial to put our findings to the test. We acknowledge that overfitting may be a limitation of our study, and we addressed this limitation in our discussion: “It is possible that the statistical procedures used may have resulted in overfitting of the model, which may have resulted in low precision of the prediction accuracy. Therefore, the values for sensitivity, specificity, and LRs [likelihood ratios] presented here may be higher than they actually were.”2 Concato et al18 stated that a large number of outcome events are needed if multiple independent variables are included in the analysis and that the results may not be valid in models having fewer than 10 outcome events per independent variable. We agree with this statement, but we also contend that the relatively narrow confidence intervals in our study do not actually support overfitting, as Concato et al18 stated that “large confidence intervals associated with individual risk estimates may indicate an overfitted model.”
April 2010
3/10/10 10:07 AM
Letters to the Editor from overfitted samples to the population. A way to solve the problem of overfitting before the research is performed is to do a thorough review of the literature and use only those predictors that have a strong theoretical and clinical grounding for their use. If this approach is not possible, the researcher can try to reduce the number of variables by using reduction methods such as factor analysis or principal components or, if possible, by adding more participants. The bottom line is that the method used by the authors in this study has almost zero probability of finding the “right” predictors. Michael T. Cibulka, Frank E. Harrell Jr M.T. Cibulka, PT, DPT, MHS, OCS, is Assistant Professor, Physical Therapy Program, Maryville University, St Louis, MO 63141. Address all correspondence to Dr Cibulka at:
[email protected]. F.E. Harrell Jr, is Professor of Biostatistics and Chair, Department of Biostatistics, Vanderbilt University School of Medicine, Nashville, Tennessee. This letter was posted as a Rapid Response on January 25, 2010, at ptjournal.apta.org.
References 1 Mintken PE, Cleland JA, Carpenter KJ, et al. Some factors predict successful shortterm outcomes in individuals with shoulder pain receiving cervicothoracic manipulation: a single-arm trial. Phys Ther. 2010;90:26–42. 2 Babyak MA. What you see may not be what you get: a brief, nontechnical introduction to overfitting in regression-type models. Psychosom Med. 2004;66:411–421. 3 Cohen J, Cohen P, West S, Aiken L. Applied Multiple Regression/Correlation Analysis for the Behavioral Sciences. 3rd ed. Mahwah, NJ: Lawrence Erlbaum Associates; 2003. 4 Harrell F. Regression Modeling Strategies With Application to Linear Models, Logistic Regression and Survival Analysis. New York, NY: Springer-Verlag; 2001. 5 Nick TG, Hardin JM. Regression modeling strategies: an illustrative case study from medical rehabilitation outcomes research. Am J Occup Ther. 1999;53:459–470.
6 Peduzzi P, Concato J, Kemper E, et al. A simulation study of the number of events per variable in logistic regression analysis. J Clin Epidemiol. 1996;49:1373–1379. 7 Tabachnick B, Fidell L. Using Multivariate Statistics. 5th ed. Boston, MA: Pearson Education Inc; 2007. [DOI: 10.2522/ptj.2010.90.4.643]
Author Response We thank Cibulka and Harrell for their response1 to our article.2 As Mark Twain wrote in his autobiography, “Figures often beguile me, particularly when I have the arranging of them myself; in which case the remark attributed to Disraeli would often apply with justice and force: ‘There are three kinds of lies: lies, damned lies, and statistics.’”3 Cibulka and Harrell have raised some valid points that we would like to address in this response. They state that the “results of this study will likely never be validated.” Although it is true that many of the physical therapy research studies that utilize this design are never carried to the next level of validation,4–14 one study validated the findings of a physical therapy derivation study using this same design.15,16 We emphasize that our study is simply the first stage in a stepwise process described by McGinn et al.17 They describe 3 progressive steps in identifying factors that predict a likely response to treatment in a patient, and it is not until the predictive factors have been validated in a new clinical setting that they become appropriate for implementation. The first step involves identification of the predictive factors, the second step involves the validation of the predictive factors, and the final step involves assessing the impact of using these predictive factors on clinical practice.17 Cibulka and
644 ■ Physical Therapy Volume 90 Number 4
Letter 4.10.indd 644
Harrell are correct to point out the need for the very cautious interpretation of this type of preliminary research, as validation is the critical next step to follow-up on a study such as ours. Having said that, for Cibulka and Harrell to state that “the method used by the authors in this study has almost zero probability of finding the ‘right’ predictors” is a bit overreaching. We agree that prospective validation of our findings is important, because individual patient variables and their interrelationships may differ among different populations. We are in the planning stages of developing a prospective randomized controlled trial to put our findings to the test. We acknowledge that overfitting may be a limitation of our study, and we addressed this limitation in our discussion: “It is possible that the statistical procedures used may have resulted in overfitting of the model, which may have resulted in low precision of the prediction accuracy. Therefore, the values for sensitivity, specificity, and LRs [likelihood ratios] presented here may be higher than they actually were.”2 Concato et al18 stated that a large number of outcome events are needed if multiple independent variables are included in the analysis and that the results may not be valid in models having fewer than 10 outcome events per independent variable. We agree with this statement, but we also contend that the relatively narrow confidence intervals in our study do not actually support overfitting, as Concato et al18 stated that “large confidence intervals associated with individual risk estimates may indicate an overfitted model.”
April 2010
3/10/10 10:07 AM
Letters to the Editor Concato et al18 also brought up the concept of “underfitting,” which occurs when there are too few outcome events. Underfitting may occur when the power for determining relationships is low, resulting in the omission of important variables from the model. We believe that we avoided underfitting the model by having a large number of successful outcomes, which allowed us to identify relationships between potential predictive variables and the outcomes. Given the preliminary nature of our study, we would rather err on the side of overfitting versus underfitting. We understand that our predictive model may describe the relationship between predictors and outcome only in the patient sample we used to develop our model, and it may be that this model will not be able to provide valid predictions in a new subset of patients. Our aim was not to describe the population as a whole. We simply set out to provide a model that gave potential predictors for the population we studied. In theory, our sample offers us the opportunity to learn about the population in general, but, with a small sample size, considerable uncertainty exists. Although our results are interesting, our findings may not replicate well with a larger data set. Our model will have to be applied to a new population in a prospective randomized trial, which we plan on undertaking. We also would like to point out a mistake in reporting the eventsper-variable calculation. Cibulka and Harrell stated that the number of successful outcomes was 31, when in fact it was 49 successful outcomes.
April 2010
Letter 4.10.indd 645
Finally, we argue that, aside from the predictive factors, this study does have important clinical implications. First, many of these patients had experienced pain for a long period of time (mean of more than 400 days). We did not deliver any treatment directly to the shoulder, yet 49 patients (61%) experienced a successful outcome. Given the severity and duration of the symptoms in this patient population and the percentage of successful outcomes, we contend that the results are not likely to be the result of “spontaneous recoveries.” So, the bigger question may be: Which patients with shoulder pain would not benefit from these interventions? Only 3 patients out of 80 experienced a worsening of their symptoms. Based on emerging research, it is clear that manual therapy to the cervicothoracic spine may play an important role in maximizing outcomes in patients with shoulder pain,19–23 and we believe that our study adds to this body of research. It has been our experience in providing education to professional (entry-level) students, residents, fellows, and practicing clinicians that the current standard of physical therapy care for patients with shoulder pain often does not include interventions directed to the cervical and thoracic regions. Based on our research and the research cited above, it may be that manual therapy to the cervicothoracic spine is a “green leafy vegetable” for patients with shoulder pain, meaning that it will help most and hurt very few. Maybe these interventions should be in the “steady diet” of most treatment plans for patients with shoulder pain. Further research is needed, but we are convinced that these techniques are extremely useful in treating patients with shoulder pain.
Again, we would like to thank Cibulka and Harrell for their insightful response to our article, and we look forward to future discussion and research on this topic. Paul E. Mintken, Joshua A. Cleland, Kristin J. Carpenter, Melanie L. Bieniek, Mike Keirns, Julie M. Whitman P.E. Mintken, PT, DPT, OCS, FAAOMPT, is Assistant Professor, Department of Physical Therapy, School of Medicine, University of Colorado Denver, 13121 E 17th Ave, Mailstop C244, Aurora, CO 80045 (USA), and Lead Clinician, Wardenburg Health Center, University of Colorado Boulder, Boulder, Colorado. Address all correspondence to Dr Mintken at:
[email protected]. J.A. Cleland, PT, PhD, is Professor, Department of Physical Therapy, Franklin Pierce University, Concord, New Hampshire; Physical Therapist, Rehabilitation Services, Concord Hospital, Concord, New Hampshire; and Faculty, Manual Physical Therapy Fellowship Program, Regis University, Denver, Colorado. K.J. Carpenter, PT, DPT, is Physical Therapist, Waldron’s Peak Physical Therapy PC, Boulder, Colorado. M.L. Bieniek, PT, DPT, is Rehabilitation Coordinator and Physical Therapist, Concord Hospital. M. Keirns, PT, PhD, is Associate Professor, School of Physical Therapy, Regis University, and Clinical Director, Physiotherapy Associates, Greenwood Athletic Club, Greenwood Village, Colorado. J.M. Whitman, PT, DSc, is Director, Evidence In Motion’s Orthopedic Manual Physical Therapy Program, Louisville, Kentucky, and Assistant Professor, School of Physical Therapy, Regis University. This letter was posted as a Rapid Response on February 12, 2010, at ptjournal.apta.org.
References 1 Cibulka MT, Harrell FE Jr. Letter to the editor on “Some factors predict successful short-term outcomes in individuals with shoulder pain receiving cervicothoracic manipulation: a single-arm trial.” Phys Ther. 2010;90:643–644. 2 Mintken PE, Cleland JA, Carpenter KJ, et al. Some factors predict successful shortterm outcomes in individuals with shoulder pain receiving cervicothoracic manipulation: a single-arm trial. Phys Ther. 2010;90:26–42.
Volume 90 Number 4 Physical Therapy ■ 645
3/10/10 10:00 AM
Letters to the Editor 3 Twain M. Mark Twain’s Autobiography. Vol I. New York, NY: P.F. Collier and Son C.; 1924. 4 Cai C, Pua YH, Lim KC. A clinical prediction rule for classifying patients with low back pain who demonstrate short-term improvement with mechanical lumbar traction. Eur Spine J. 2009;18:554–561. 5 Cleland JA, Childs JD, Fritz JM, et al. Development of a clinical prediction rule for guiding treatment of a subgroup of patients with neck pain: use of thoracic spine manipulation, exercise, and patient education. Phys Ther. 2007;87:9–23. 6 Currier LL, Froehlich PJ, Carow SD, et al. Development of a clinical prediction rule to identify patients with knee pain and clinical evidence of knee osteoarthritis who demonstrate a favorable short-term response to hip mobilization. Phys Ther. 2007;87:1106–1119. 7 Fernandez-de-las-Penas C, Cleland JA, Cuadrado ML, Pareja JA. Predictor variables for identifying patients with chronic tension-type headache who are likely to achieve short-term success with muscle trigger point therapy. Cephalalgia. 2008;28:264–275. 8 Hicks GE, Fritz JM, Delitto A, McGill SM. Preliminary development of a clinical prediction rule for determining which patients with low back pain will respond to a stabilization exercise program. Arch Phys Med Rehabil. 2005;86:1753–1762. 9
Iverson CA, Sutlive TG, Crowell MS, et al. Lumbopelvic manipulation for the treatment of patients with patellofemoral pain syndrome: development of a clinical prediction rule. J Orthop Sports Phys Ther. 2008;38:297–309; discussion 309–312.
10 Lesher JD, Sutlive TG, Miller GA, et al. Development of a clinical prediction rule for classifying patients with patellofemoral pain syndrome who respond to patellar taping. J Orthop Sports Phys Ther. 2006;36:854–866.
11 Raney NH, Petersen EJ, Smith TA, et al. Development of a clinical prediction rule to identify patients with neck pain likely to benefit from cervical traction and exercise. Eur Spine J. 2009;18:382–391. 12 Sutlive TG, Lopez HP, Schnitker DE, et al. Development of a clinical prediction rule for diagnosing hip osteoarthritis in individuals with unilateral hip pain. J Orthop Sports Phys Ther. 2008;38:542–550. 13 Wainner RS, Fritz JM, Irrgang JJ, et al. Development of a clinical prediction rule for the diagnosis of carpal tunnel syndrome. Arch Phys Med Rehabil. 2005;86:609–618. 14 Whitman JM, Cleland JA, Mintken PE, et al. Predicting short-term response to thrust and nonthrust manipulation and exercise in patients post inversion ankle sprain. J Orthop Sports Phys Ther. 2009;39:188–200. 15 Childs JD, Fritz JM, Flynn TW, et al. A clinical prediction rule to identify patients with low back pain most likely to benefit from spinal manipulation: a validation study. Ann Intern Med. 2004;141:920–928. 16 Flynn T, Fritz J, Whitman J, et al. A clinical prediction rule for classifying patients with low back pain who demonstrate short-term improvement with spinal manipulation. Spine. 2002;27:2835–2843.
20 Bergman GJD, Winters JC, Groenier KH, et al. Manipulative therapy in addition to usual medical care for patients with shoulder dysfunction and pain: a randomized, controlled trial. Ann Intern Med. 2004;141:432–439. 21 Boyles RE, Ritland BM, Miracle BM, et al. The short-term effects of thoracic spine thrust manipulation on patients with shoulder impingement syndrome. Man Ther. 2009;14:375–380. 22 Winters JC, Sobel JS, Groenier KH, et al. Comparison of physiotherapy, manipulation, and corticosteroid injection for treating shoulder complaints in general practice: randomised, single blind study. BMJ. 1997;314:1320–1325. 23 Walser RF, Meserve BB, Boucher TR. The effectiveness of thoracic spine manipulation for the management of musculoskeletal conditions: a systematic review and meta-analysis of randomized clinical trials. J Man Manip Ther. 2009;17:237– 246. [DOI: 10.2522/ptj.2010.90.4.644]
17 McGinn TG, Guyatt GH, Wyer PC, et al; Evidence-Based Medicine Working Group. Users’ guides to the medical literature, XXII: how to use articles about clinical decision rules. JAMA. 2000;284:79–84. 18 Concato J, Feinstein AR, Holford TR. The risk of determining risk with multivariable models. Ann Intern Med. 1993;118:201–210.
646 ■ Physical Therapy Volume 90 Number 4
Letter 4.10.indd 646
19 Strunce JB, Walker MJ, Boyles RE, Young BA. The immediate effects of thoracic spine and rib manipulation on subjects with primary complaints of shoulder pain. J Man Manip Ther. 2009;17:230– 236.
April 2010
3/10/10 10:00 AM
Clarification
Clarification Weppler CH, Magnusson SP. “Increasing muscle extensibility...?” Phys Ther. 2010;90:438–449. Weppler and Magnusson (“Increasing Muscle Extensibility: A Matter of Increasing Length or Modifying Sensation?” March 2010) offer clarification on points made on page 443 of their article. Specifically, they clarify that they were referring to research with subjects who were neurologically intact and asymptomatic and that the biomechanical effects of contract/relax stretching involve both a sensory modification and a viscoelastic effect. Revised parts are bolded below: Experimental evidence does not support any of these assertions in studies involving subjects who were neurologically intact and asymptomatic.13,14,54,55 Stretch reflexes have been shown to activate during very rapid and short stretches of muscles that are in a mid-range position, producing a muscle contraction of short duration.54 However, most studies of subjects who were neurologically intact and asymptomatic and whose muscles were subjected to a long, slow, passive stretch into end-range positions, did not demonstrate significant activation of stretched muscles.14,54,56,57 Even studies that simulated ballistic (cyclic and high-velocity) stretching demonstrated no evidence of significant stretch reflex activation of muscles both in human26 and animal23 models. In a study that evaluated the effects of a single “contract-relax” stretch,25 no significant electromyographic activity was observed during stretch application and no shift of passive torque/angle curves was observed aside from what could be attributed to viscoelastic stretch relaxation. Several short-term (3 and 6 weeks’ duration) stretching studies45,58 demonstrated similar findings: no significant electromyographic activity during initial and final testing and no shift of passive torque/ angle curves. The increase in end-range joint angles, therefore, could not be attributed to neuromuscular relaxation.14,25,45,58
Quality. Value. Peace of mind. All under one roof. Safety First® offers your patients a variety of ADA approved products that enhance lives by assisting in short and long term care. Our quality products are available at affordable prices at your local Home Depot® or www.homedepot.com/bathsafety
[DOI: 10.2522/ptj.20090012.cx]
647 ■ Physical Therapy Volume 90 Number 4
Clarification_4.10.indd 647
April 2010
3/15/10 5:35 PM
Scholarships, Fellowships, and Grants News from the Foundation for Physical Therapy Recent Publications by Foundation-Funded Researchers “Changes in Two Children with Cerebral Palsy After Intensive Suit Therapy: A Case Report,” by Bailes AF, Greve K, and Schmitt LC, is featured in the Spring 2010 issue of Pediatric Physical Therapy (22[1]:76–85). Amy Feldman Bailes, PT, MS, PCS, is a recipient of a 2009 Promotion of Doctoral Studies (PODS) I award, and Laura Schmitt, PT, DPT, OCS, SCS, ATC, received PODS II awards in 2004 and 2005.
E
“The Impact of a Sloped Surface on Low Back Pain During Prolonged Standing Work: A Biomechanical Analysis,” by Nelson-Wong E and Callaghan JP, was published online in Applied Ergonomics on January 27, 2010. Erika Nelson-Wong, PT, DPT, MSPT, received a 2005 Kendall Scholarship, a 2006 PODS I, and a 2008 PODS II. “Neural Correlates of Impaired Functional Independence in Early Alzheimer’s Disease,” by Vidoni ED, Honea RA, and Burns JM, is featured in Journal of Alzheimer’s Disease (2010;19[2]:517–527). Eric Vidoni, PT, PhD, is the recipient
Become
xce
!
PTi o n a l
...
With a modest monthly gift as little as $21, you’ll build the future of physical therapy.
So, go green and become ExcePTional today!
FoundationForPhysicalTherapy.org
Research
Results
648 ■ Physical Therapy Volume 90 Number 3
Foundation 04.10.indd 648
Recognition
of a 2009 New Investigator Fellowship Training Initiative. “The Challenge of Pain in Children and Adolescents With Cerebral Palsy,” an editorial by Ann VanSant, PT, PhD, FAPTA, is featured in the Spring 2010 issue of Pediatric Physical Therapy (22[1]:1). VanSant received Foundation awards in 1983 and 1988. “Measuring Function in Young Children With Impairments,” by Tucker CA and Watson KE, is available in the Spring 2010 issue of Pediatric Physical Therapy (22[1]:51). Carole Tucker, PT, PhD, PCS, RCEP, received Foundation awards in 1995 and 1996.
Now Accepting Nominations to Join SRC The Foundation for Physical Therapy is currently accepting nominations to serve on the Scientific Review Committee. This all-volunteer committee is vital to the awarding of more than 25 scholarships, fellowships, and grants each year. If you know of someone who could serve as an application reviewer, or if you have an interest in volunteering yourself, please visit the Foundation’s Web site under “Grants, Fellowships, and Scholarships” to review the criteria for membership. If you qualify and
Save the Date Break out your zoot suits and flapper dresses! The Foundation is taking you back to the 1920s at their annual gala, “Puttin’ on the Ritz,” on Thursday, June 17, 2010, during APTA’s Annual Conference in Boston. Tickets are on sale now. Visit the Foundation’s Web site for details.
March 2010
3/15/10 5:35 PM
Scholarships, Fellowships, and Grants
have the desire to make an impact on the future of physical therapy education and research, submit your full CV to the Scientific Program Administrator.
22nd Annual Foundation Split Raffle Did you know that funds raised through the Foundation’s Annual Split Raffle support doctoral scholarships for today’s emerging physical therapist researchers? The purchase of a split raffle ticket serves as an investment in the strength and future of the physical therapy profession with a chance to win one of nine $2,000 prizes or the $10,000 grand prize. Contact Barbara Malm for more information on how to participate or visit the Foundation’s Web site for complete rules.
Do You Have Some Good News You’d Like To Share?
Win a Trip to Hawaii! The Foundation is sponsoring the “Aloha Getaway” Sweepstakes, where you could win a trip for 2 to Hawaii! Every $10 donation to the Foundation will receive 1 entry into the drawing; a $40 donation receives 5 entries. If you weren’t able to enter the sweepstakes at our booth at CSM, you can enter online at the Foundation’s Web site, www. FoundationforPhysicalTherapy.org. No Purchase Necessary to Enter or Win: The “Aloha Getaway” Sweepstakes is open to all legal residents of the United States, age 18 years or older as of January 22, 2010. Sweepstakes entry begins January 22, 2010, and ends June 25, 2010. The staffs of the Foundation and APTA are not eligible to participate. View the complete official rules at the Foundation’s Web site, which govern the sweepstakes. Void where prohibited.
2010 Miami–Marquette Challenge Raises Funds for Research This year, students participating in the 2010 Miami–Marquette Challenge pledged to raise $200,000 in support of the Foundation. The student-led fundraiser supports research grants and PODS I and II scholarships for emerging physical therapist researchers. Students hope to meet this year’s goal and expand their role by helping to fund the new $300,000 Clagett Family Research Grant. Donations must be received by the May 5, 2010 Challenge deadline to count for this year’s fundraiser. For a list of participating schools and ways to get involved, visit the Foundation’s Web site or contact Barbara Malm at 800/875-1378, ext 8502, for more information. [DOI: 10.2522/ptj.2010.90.4.648]
If you’d like to include any announcements in the Foundation’s section, contact our communications assistant at abegailmatienzo@ apta.org.
March 2010
Foundation 04.10.indd 649
Volume 90 Number 3 Physical Therapy ■ 649
3/10/10 10:50 AM