Physical Therapy- Journal of the APTA, Volume 90, Issue 3 (March 2010)

March 2010  Volume 90  Number 3

Research Reports 338

Balance Impairment in Older Adults

348

Improving Postnatal Well-Being

356

Effects of Resistance Training on Physical Activity

367

Pediatric Endurance and Limb Strengthening

382

Motor Skill Development in Infants With Down Syndrome

391

Rehabilitation Improves Function in Facial Paralysis

398

Interhemispheric Modulation Induced by Cortical Stimulation and Motor Training

Technical Report 411

Case Reports 420

Physical Therapy in the Emergency Department

427

Aerobic Exercise in Spinal Cord Injury Rehabilitation

Perspective 438

Increasing Muscle Extensibility

CARE V Conference Series 450

cover_03.10.indd 1

Patellar Taping in Patients With Patellofemoral Pain

Patient-Centered Research

2/12/10 4:32 PM

Business Growing?

Referen

If your business is growing, you need a solid professional liability plan to help protect your interests so your business will flourish.

ce

A 3 D B A 3code PD

The Professional Liability Insurance program offered through HPSO gives you 24/7 protection for your business, assets, and employees against covered allegations of professional malpractice, including:

For Physical Therapy and Rehabilitation group Practices

 Professional Liability Insurance  Legal Defense for covered claims  License Protection  Deposition Representation  Defendant Expense Benefit for lost wages  And more

Optional Coverages:  General Liability  Additional Insured Endorsement  Medical Director Endorsement  Additional Office Locations

Start protecting your business interests first. To find out more about coverage offered through HPSO:

Call 1-888-288-3534 • www.hpso.com/ptj3

This program is underwritten by American Casualty Company of Reading, Pennsylvania, a CNA company, and is offered through the Heatlhcare Providers Service Organization Purchasing Group. Coverages, rates and limits may differ or may not be available in all states. All products and services are subject to change without notice. This material is for illustrative purposes only and is not a contract. It is intended to provide a general overview of the products and services offered. Only the policy can provide the actual terms, coverages, amounts, conditions and exclusions. CNA is a service mark and trade name registered with the U.S. Patent and Trademark Office. Healthcare Providers Service Organization is a division of Affinity Insurance Services, Inc.; in CA (License #0795465), MN & OK, AIS Affinity Insurance Agency, Inc.; and in NY, AIS Affinity Insurance Agency. ©2010 Affinity Insurance Services, Inc. PTJ0310

Ads_3.10.indd 2

2/12/10 2:51 PM

It’s a matter of time. Spend less time hunting and more time healing. Start your search for evidence at www.hookedonevidence.org. APTA’s Hooked on Evidence database provides members with  article extracts and patient clinical scenarios of commonly seen conditions. Each extract provides information about the participants, treatments, and outcomes that you can apply in your own practice.   Bookmark our site. Let Hooked on Evidence lead you to the articles that apply best to your patients. And benefit from instant access to evidence of the effectiveness of physical therapy interventions.

www.hookedonevidence.org

Ads_3.10.indd 317

2/12/10 2:51 PM

Next month—in PTJ or online at ptjournal.apta.org •

Perturbation-Based BalanceTraining Program for Older Adults

•

HRQOL and Robotic-Assisted Therapy for Hand Function

•

Supervised Exercise Program for People With Bleeding Disorders and Hemophilic Arthritis

• Intensive Physical Training After Stroke •

Electromyographic Comparison of Exercises With Dumbbells and Elastic Tubing

•

Assessment of Risk of Recurrent Falls in Elderly People

•

Kinematics of Rising From a Chair

•

Construct Validity of Muscle Force Tests

•

Cold Modalities and Nerve Conduction

• Therapeutic Exercise Program for Patients Osteoarthritis of the Hip •

Physical Therapist Management of Chronic Fatigue Syndrome/ Myalgic Encephalomyelitis

•

Ethics in Nonpharmacological Treatment Research

•

Continuing Professional Development in Arthritis Management

•

And much more!

“Just had to RAVE about PTJ Online! I received the email ToC alert, and clicked on the article I wanted. Not only could I cut and paste the references for an upcoming talk, but with a click of the mouse I was transported to the FREE FULL TEXT of these reference articles. Magic! I also listened to 2 podcasts while I was making dinner!” —Janet Peterson, PT, DPT, MA

Visit ptjournal.apta.org for enhanced features, 0, 19 . f 2 ll 1 r o of a d # y journal o n p t ac 5 s a ra t f top nal the e c r pa he ou al id im in t n j ysic ldw ) an J is atio ph or ing ith PT ilit all ls w ank b g a r W ha n rn R re amo jou (JC

including articles published ahead of print! Physical Therapy (PTJ)—APTA’s peer-reviewed scholarly

Ads_3.10.indd 336

2/12/10 3:17 PM

Guidelines for

Clinicians Guidelines for the

Provision of Physical Therapy in the Home Second Edition

NEW: N EW Guidelines for the Provision of Physi Physical Therapy in the Home, 2nd Edition Include in this updated publication from the APTA Home Health Section are guidelines on the Included roles of the PT and PTA, orientation to agencies, scheduling, discharge planning, and documentation; a bbibliography; and a list of online resources. (ISBN 978-1-931369-59-6, 24 pages, 2008) Order No. N P-131 Regular R egullar price: $48 APTA A PTA Member M price: $29 Special S peciaal Home Health Section Member price: $20 (phone or fax orders only)

Domestic Violence Guidelines

G iddelinnes for Gu or

These guidelines give you needed background on the issues and the roles that you have in identifying, treating, and advocating for people who are abused. Take and pass the exam at the end of each booklet and earn .2 CEU. Regular price: $22.00 each APTA Member price: $12.95 each

Recognizin

es for Guidelin izing

n Recog

and

anndd an

e

Car ding Provi tims of for Vic stic Dome ce n Viole

g

Prro oviid ding Care for Viic fo cttiim ms off Elld der Abuse

OR BUY ALL 3 Domestic Violence Guidelines and SAVE!

Guidelines for for

Order No. KIT-ABUSE Regular price: $55.00 APTA Member price: $32.95

Providing Provi rovidi idi ding i Car C Care forr Victims of fo Child Ab Abuse

Guidelines for Recognizing and Providing Care for Victims of Elder Abuse (ISBN 978-1-931369-39-8, 50 pgs, 2007) Order No. P-160-07

Ads_3.10.indd 337

Guidelines for Recognizing and Providing Care for Victims of Domestic Violence (ISBN 978-1-887759-31-1, 40 pgs, 2005) Order No. P-138

Recognizing R Re eco cogn gniiz izin izin ing g and

Guidelines for Recognizing and Providing Care for Victims of Child Abuse (ISBN 978-1-887759-80-9, 36 pgs, 2005) Order No. P-159

2/17/10 3:54 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

320 ■ Physical Therapy Volume 90 Number 3

Masthead_3.10.indd 320

March 2010

2/12/10 4:33 PM

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_3.10.indd 321

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

March 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 3 Physical Therapy ■ 321

2/12/10 4:33 PM

Editorial Retaining Trust

W

e all are buried in paperwork and red tape, but we also live in an environment where trust is low and suspicion is high. To ensure even greater transparency in our published work, PTJ is presenting a new and improved conflictof-interest (COI) policy. We are adopting the COI policy and form developed by the International Journal of Medical Journal Editors (http://www.icmje.org/).1 Four types of disclosures are included in this policy: (1) Authors’ associations with commercial entities that provided support for the work reported in the submitted manuscript. (2) Authors’ associations with commercial entities that have an interest in the general area of the submitted manuscript. (3) Any financial associations involving the author or author’s spouse and children younger than 18 years of age. (4) Any nonfinancial associations that may be relevant to the submitted manuscript. Although PTJ has had a COI policy for many years, our adoption of this new policy will contribute to the standardization of a policy among many of the leading biomedical journals. A recent survey of 256 biomedical journals revealed that 89% had COI policies, but there was variation in the definition of COI and in the requirements for disclosure.2 The policy that PTJ has adopted provides standard definitions and disclosure requirements, and the electronic disclosure form (now available at http://ptjournal.apta.org/misc/coi_ disclosure.pdf) can be completed and uploaded to PTJ’s manuscript submission site. This approach should allow authors to save the form and update it as needed for future submissions. Our new policy goes one step further: we are asking Reviewers and Editorial Board members to identify potential conflicts with the topic of the manuscript. This is our attempt to recognize and address any potential bias that our review team may have. At this time, there is limited financial association between most of our authors and commercial entities, and, in those few cases where there is an association, that does not mean that the research has been “tainted.” It is important, however, that the reader is aware of a potential conflict and feels free to evaluate the meaningfulness of the results in the context of a financial association. More importantly, as physical therapists move to develop more interdisciplinary relationships to assist in the translation of basic science and engineering, opportunities to develop relationships with commercial entities are looming, and we want to be prepared. Thank you for helping PTJ to share all the relevant information so that our readers have increased confidence in the findings that authors report.

To comment, submit a Rapid Response to this editorial posted online at ptjournal.apta.org.

322 ■ Physical Therapy Volume 90 Number 3

editorial_Craik_3.10.indd 322

March 2010

2/12/10 4:34 PM

Editorial

Special Thanks On another note, I would like to take this opportunity to thank an Editorial Board Member Extraordinaire. G. Kelley Fitzgerald, PT, PhD, OCS, has served on our Editorial Board for more than 5 years. During this time, he shepherded hundreds of manuscripts and worked with a committee to enhance our case reports. He called last year at this time to say that he had taken on additional responsibilities at the University of Pittsburgh and that he needed to preserve time to oversee his own, very exciting research endeavors. But we managed to hang on to him until he finally begged to leave his PTJ post. Kelley has been an insightful, and delightful, member of our team. He brought new authors to PTJ and was always in touch with our readers, often suggesting meaningful ways to improve the procedures that we use for review of manuscripts and dissemination of content. Never one afraid to express his opinion, Kelley always respected a sound rationale. Kelley, thank you for the endless hours you spent helping to improve the relevance of PTJ for our readers, and, in particular, for our readers who asked for more immediate clinical relevance. Rebecca L. Craik, PT, PhD, FAPTA Editor in Chief References 1 Drazen JM, Van Der Weyden MB, Sahni P, et al. Uniform format for disclosure of competing interests in ICMJE journals. JAMA. 2010;303:75–76. 2 Blum JA, Freeman K, Dart RC, et al. Requirements and definitions in conflict: conflict of interest policies of medical journals. JAMA. 2009;302:2230–2234. [DOI: 10.2522/ptj.2010.90.3.322]

March 2010

editorial_Craik_3.10.indd 323

Volume 90 Number 3 Physical Therapy ■ 323

2/12/10 4:35 PM

Editorial Improving the Evidence Base for Physical Therapy Disability Interventions

D

isability is an emerging global issue, as reflected by the first joint World Health Organization (WHO)–World Bank Report on Disability, Rehabilitation, and Inclusion, scheduled to be released by WHO in 2010. Some estimates suggest that 10% of the world’s population has some form of disability—but that figure excludes the families who are affected by a member’s disability.1 According to 2005 US Census Bureau statistics, 54 million (18.7%) people in the United States have some level of disability.2 That number will grow significantly over the next 20 years as the baby-boom generation enters late life, when the risk of disability is greatest. People with disabilities comprise the largest and arguably the most important health care consumer group in the United States; however, as the Institute of Medicine and others have warned,3,4 far too little progress has been made over the past 2 decades in removing obstacles that limit what many people with physical and cognitive impairments can achieve. Part of the challenge might be that disability is complex, multifactorial, and difficult to define. In addition, there is a persistent lack of evidence about strategies to reduce activity limitations and improve participation for people with disability, which hinders the development of sound policy and intervention options. Despite the challenges, this is a time of great innovation: many technological developments provide the opportunity for new strategies and interventions to enhance participation in people with disability. New technologies have the potential to reduce environmental barriers, such as “smarter” homes and buildings that allow tasks to be performed with minimal human input and mobility and communication equipment that minimizes the impact of underlying impairments on function. The usual delivery of therapy services is enhanced by telerehabilitation, which facilitates training via the Web, telephone, or other technology over great distances. New social media are allowing more people with disabilities to connect with one another and support each other in real time in a way that was impossible before. The field of physical therapy can be central to ensuring an optimal future for people with disability across the globe. As Jules Rothstein, Editor in Chief Emeritus, wrote in PTJ’s 1994 special issue on Physical Disability: We began, and I hope will always be seen as, professionals who seek to deal not just with pathologies and impairments, but also with the disabling consequences of diseases, trauma, and developmental anomalies. In essence we became an identifiable group not when we administered massages, heat, exercises, or electrical stimulation, but rather when we chose to collect a body of knowledge and clinical approaches that focused on eliminating disability, when we sought to bring meaningful function back into the lives of those who sought our services.5(p375)

To comment, submit a Rapid Response to this editorial posted online at ptjournal.apta.org.

There is a scarcity of available scientific evidence about the patterns of health conditions and the health care needs of people with disability and about effective interventions for prevention, treatment, and rehabilitation. In addition to the lack of evidence about what works to improve activity and participation of those with disabilities, there also is little evidence about the costs incurred. With the aim of helping to advance science and practice in these areas, PTJ invites original contributions to a special issue, to be published in late 2010/early 2011. Improving the Physical Therapy Evidence Base for Disability Interven-

324 ■ Physical Therapy Volume 90 Number 3

editorial_Jette_3.10.indd 324

March 2010

2/17/10 2:56 PM

Editorial tion will capture the latest research, perspectives, and scholarship. We seek to aggregate and disseminate high-quality disability research on such topics as: • What are the patterns of need for physical therapy in people with disabilities? • What are the key physical therapy interventions that will improve levels of activity and participation for people with disabilities? • What new interventions or technological developments have preliminary evidence of efficacy in improving participation in people with disability? • What is the current state of understanding of the etiology, prevention, and treatment of secondary conditions for people with disability? • Are physical therapy services, technological aids, and universal design features that promote good health and that maximize societal participation accessible to people with disability? • Are there effective ways to reduce environmental barriers for people with disability? • Are there important breakthroughs and advances in the conceptualization, definition, and measurement of disability? • Are there special issues and needs for people aging with a disability compared to those who develop a disability as they age? • What rehabilitation approaches have some evidence of efficacy when used in developing countries? We encourage scientists conducting research relevant to the nexus of physical therapy and disability to contribute to building this crucial knowledge base. PTJ’s special issue will provide a global forum for presenting authoritative empirical research, theoretical development and perspectives, clinical case studies, and related innovative developments. Those interested in contributing to the special issue are asked to forward an outline of your proposed topic for consideration to [email protected]. Alan M. Jette, PT, PhD, FAPTA Nancy Latham, PT, PhD References 1 Mont D. Measuring disability prevalence. SP Discussion Paper No. 0706. Washington, DC: The World Bank; 2007. 2 Brault M. Americans with disabilities: 2005. Current Population Reports. Washington, DC: US Census Bureau; 2008:3. 3 Field M, Jette AM, eds; Committee on Disability in America. The Future of Disability in America. Washington, DC: The National Academies Press; 2007. 4 Disability: beyond the medical model [editorial]. The Lancet. 2009;374:1793. 5 Rothstein JM. Disability and our identity [editorial]. Phys Ther. 1994;74:375–378. [DOI: 10.2522/ptj.2010.90.3.324]

March 2010

editorial_Jette_3.10.indd 325

Volume 90 Number 3 Physical Therapy ■ 325

2/12/10 4:45 PM

In Tribute Geoffrey D. Maitland, 1924–2010 Geoffrey D. Maitland, a physical therapy visionary and innovator for more than 4 decades, died on January 22, 2010, in Australia. A pioneer in the use of mobilization for pain modulation, his models for practice and his descriptions of examination and treatment techniques are still used as methodological standards by manual therapy researchers. Maitland authored the classic texts Vertebral Manipulation, now in its 7th edition, and Peripheral Manipulation, now in its 4th edition, both of which have been translated into several languages, including Japanese, Spanish, and German. He wrote extensively for the Australian Journal of Physiotherapy and other journals worldwide. He promoted “research by physical therapists for physical therapists” at a time when prescribing physical therapy interventions was viewed as the role of the physician. In 1965, Maitland presented the first 3-month course on “Manipulation of the Spine,” based at the South Australian Institute of Technology. In 1974, this course developed into the 1-year postgraduate diploma in manipulative physiotherapy and subsequently became a master’s degree course. In addition to serving on various physical therapy committees and boards in Australia, Maitland contributed to the bigger, international picture. In 1974, he participated in founding the International Federation of Orthopaedic Manipulative Therapy (IFOMPT), a branch of the World Confederation for Physical Therapy (WCPT). The founders of IFOMPT wanted to establish a benchmark for the teaching of manual therapy internationally, and the 2008 IFOMPT Educational Standards Document is the culmination of those early efforts, forming the basis of manual therapy education program in member countries. Maitland also inspired the establishment of the International Maitland Teacher’s Association (IMTA). Maitland received numerous honors throughout his career. He was named a Member of the British Empire in 1981 and received the Mildred O. Elson Award from WCPT in 1995 for his life’s work. He was awarded honorary membership in the American Physical Therapy Association in 1992. Maitland’s legacy is reflected in the work of several researchers who have built on his decision-making process and developed it into a structured and evidence-based clinical reasoning framework and who have advanced knowledge, skills, and strategies that address neurogenic and other pain mechanisms. He highlighted the need for deep and broad theoretical knowledge to support and inform clinical practice. In addition to his foundational work in manipulation and mobilization, Maitland has been heralded for fostering patientcentered care and awareness of “the nature of the person” and its impact on treatment. He believed in “the body’s capacity to inform” both verbally and nonverbally and was said to detect the almost imperceptible nuances of the patient’s responses. Maitland died almost 1 year after the death of his wife Anne, who was often present at his lectures and gave him honest feedback. From the beginning, Maitland and his wife developed a robust internal moderation system to ensure quality control and quality assurance of his work. —Submitted by Kevin Banks, Chairman, and Members of The International Maitland Teacher’s Association

326 ■ Physical Therapy Volume 90 Number 3

In tribute.indd 326

March 2010

2/17/10 2:53 PM

Health Policy in Perspective Comparative Effectiveness Research: Opportunities and Challenges for Physical Therapy Janet K. Freburger, Timothy S. Carey

I

n February 2009, President Obama signed the American Recovery and Reinvestment Act of 2009, which, among other initiatives, appropriated $1.1 billion for comparative effectiveness research (CER). Of that $1.1 billion, $300 million was allocated to the Agency for Healthcare Research and Quality (AHRQ); $400 million to the National Institutes of Health (NIH); and $400 million to the Secretary of the Department of Health and Human Services. The purposes of the appropriations are to: (1) conduct, support, or synthesize research that compares the clinical outcomes, effectiveness, and appropriateness of items, services, and procedures that are used to prevent, diagnose, or treat diseases, disorders, and other health conditions; and (2) encourage the development and use of clinical registries, clinical data networks, and other forms of electronic health data that can be used to generate or obtain outcomes data.1

As applied to health care, CER is simply the evaluation of the impact of different options that are available for treating a given medical condition for a particular set of patients.2 Many believe that CER is a mechanism for decreasing health care costs and improving health care quality, the primary goals of health care reform.2,3 This belief is fueled by 2 major factors: (1) the limited evidence available to patients and providers about what treatments work best for which patients and (2) the fact that there are significant geographic

March 2010

JournalHEalthPolicy.indd 327

differences in health care spending within the United States—but few apparent differences in healthrelated outcomes. The Institute of Medicine (IOM) estimates that less than half of all health care treatments delivered in our nation are supported by adequate evidence.4 Furthermore, many practice guidelines—even those thoughtfully conceived and developed by experts—are based on limited evidence.5 For example, of the 23 Osteoarthritis Research Society International (OARSI) guidelines for the treatment of hip and knee osteoarthritis, 6 are based on expert opinion, and several are based on only moderate evidence.6 Even if evidence is available on the effectiveness of a treatment, we often do not know whether the treatment is effective for all subgroups of patients.2,4,5 Nor do we typically know whether the added benefits of a more expensive treatment warrant the added costs. Sometimes effective treatments are underutilized in all patients or in specific subpopulations, such as minority populations or groups with a low socioeconomic status. For example, our study published in 2009 found that exercise, one of the moderately effective treatments for chronic low back pain, is underutilized and that individuals with less education are less likely to be prescribed exercise.7 Even after accounting for differences in health or “need,” wide geographic variation in the use of

various types of health care has been repeatedly documented in the literature.8–10 This variation in health care use is attributed to the uncertainty about the effectiveness of many types of care. Weinstein and colleagues11 illustrated this concept clearly in their study of lumbar spine surgery rates among Medicare beneficiaries. They found significant geographic variation in surgery rates and attributed this to the uncertainty regarding the effectiveness of lumbar spine surgeries. In contrast, rates for surgical repair of hip fracture, an intervention with strong evidence of effectiveness and minimal tradeoffs between risks and benefits, varied little across the United States. Use of procedures and treatments that have more uncertainty in terms of effectiveness or that have tradeoffs between risks and benefits may vary based on (1) patients’ values and attitudes toward risk and (2) providers’ attitudes, experiences, incentives, and judgments about the treatment. Although most physical therapy interventions carry little risk, the effectiveness of many of our interventions is far from certain. The aim of CER is to improve health care quality by providing patients, health care providers, and other stakeholders with better information about the risks and benefits of different treatment options. In addition to identifying effective treatments, CER should educate about tests and treatments of limited effectiveness. Because there is so much uncertainty about the effectiveness and relative effec-

Volume 90 Number 3 Physical Therapy ■ 327

2/12/10 5:46 PM

Health Policy in Perspective tiveness of medical treatments, many believe that CER could eventually alter the way in which medicine is practiced and lead to lower health care spending without having adverse effects on health.2

Defining Comparative Effectiveness Research As CER has come to the forefront as a mechanism to improve health care, several public and private agencies have created definitions for the term “comparative effectiveness research.”5 Although these definitions differ slightly in their wording, they all emphasize that CER involves the direct comparison of the effectiveness of 2 or more interventions in “real world settings” and that its purpose is to inform patients, providers, and decision makers about what interventions are most effective for what patients under what circumstances. The definition from the IOM states: Comparative effectiveness research is the generation and synthesis of evidence that compares the benefits and harms of alternative methods to prevent, diagnose, treat, and monitor a clinical condition or to improve the delivery of care. The purpose of CER is to assist consumers, clinicians, purchasers, and policy makers to make informed decisions that will improve health care at both the individual and population levels.5

What is lacking from the IOM definition is a specific mention of costs. There is disagreement as to whether costs should be incorporated into CER.3,12 Costs can be defined in different ways and from different perspectives (eg, patient, private insurer, society). Because costs can vary dramatically depending on the perspective, there is concern that results of cost analyses may be

used inappropriately to make decisions regarding pricing, coverage, and reimbursement. Opponents also argue that reducing uncertainty through clinical effectiveness research—rather than through cost effectiveness research—is a more valuable use of scarce research dollars. Proponents of including costs in CER argue that costs are already a part of the decision-making process regarding coverage and health care services and that not including costs does not adequately address the problem of rising health care costs.

Is Comparative Effectiveness Really Something Novel? In a broad sense, CER is nothing new. Scientists have been comparing the effectiveness of different interventions for decades. Relative to efficacy research, however, CER is lagging behind. Efficacy studies are randomized, controlled trials comparing an intervention to a control (often a placebo or a sham treatment) on a carefully selected group of subjects under controlled conditions. Although these types of studies are considered the “gold standard” for determining the effects of an intervention, their findings often are not generalizable to broader populations and settings or to certain subgroups that may have been excluded from the study. Subgroups often excluded from efficacy trials include the elderly and those with multiple coexisting conditions. Comparative effectiveness research utilizes a variety of data sources, including systematic reviews of existing literature and analysis of secondary data, such as claims data, patient registries, and electronic health records. Study designs include nonexperimental studies, such as prospective cohort studies and practical experimental clinical

328 ■ Physical Therapy Volume 90 Number 3

JournalHEalthPolicy.indd 328

trials (eg, head-to-head pragmatic randomized trials, clustered randomized trials). The AHRQ has been supporting CER for many years through its Effective Health Care Program (EHC).13 The EHC program consists of Evidence Based Practice Centers (EPCs), the Developing Evidence to Inform Decisions about Effectiveness (DEcIDE) network, Centers for Education and Research on Therapeutics (CERTs), and the John M. Eisenberg Clinical Decisions and Communications Science Center. Each of these centers performs a variety of activities related to the various aspects of CER. The Eisenberg Center, in particular, focuses on distilling and disseminating the EHC reports for providers and the public. Although many of the activities of the EHC have not specifically focused on physical therapy–related issues or interventions, one EPC review currently underway is the Comparative Effectiveness of NonOperative and Operative Treatments for Rotator Cuff Repair. Several reports on methodologies for CER are also available and may be of use for physical therapy researchers interested in learning more about CER. In addition, 3 of the 14 CERTs focus on topics particularly relevant to physical therapists: musculoskeletal disorders, therapies for older adults and the effects of aging, and pediatric therapeutics. A final key aspect of the EHC program is that it seeks strong involvement of all stakeholders (eg, patients, clinicians, researchers, health care payers, professional organizations). Interested individuals can nominate topics to be researched and can submit comments on drafts of proposed research and reports through the EHC Web site (www.effective healthcare.ahrq.gov).

March 2010

2/12/10 5:47 PM

Health Policy in Perspective • Compare the effectiveness of primary prevention methods, such as exercise and balance training, versus clinical treatments in preventing falls in older adults at varying degree of risk. • Establish a prospective registry to compare the effectiveness of treatment strategies for low back pain without neurological deficit or spinal deformity. • Compare the effectiveness of school-based interventions involving meal programs, vending machines, and physical education, at different levels of intensity, in preventing and treating overweight and obesity in children and adolescents. • Compare the effectiveness of various strategies (eg, clinical interventions, selected social interventions [such as improving the built environment in communities and making healthy foods more available], combined clinical and social interventions) to prevent obesity, hypertension, diabetes, and heart disease in at-risk populations such as the urban poor and American Indians. • Compare the effectiveness of interventions (eg, community-based multi-level interventions, simple health education, usual care) to reduce health disparities in cardiovascular disease, diabetes, cancer, musculoskeletal diseases, and birth outcomes.

Figure 1. Priorities in the Insititute of Medicine’s top quartile that are relevant to physical therapists.

What Does Comparative Effectiveness Research Mean for the Physical Therapy Profession? There is no question that CER is relevant to physical therapy. In reviewing the IOM’s list of initial national priorities for CER,5 we identified 5 topics in their upper quartile of priority that were particularly rel-

evant to physical therapy (Fig. 1). In addition, several of the remaining 75 topics were relevant. Many of the priority conditions and populations that are a focus of AHRQ’s CER projects also are of interest to physical therapy (Table). Priority conditions and populations are cross-cutting themes in the strategic framework of the Federal Coordinating Council

Table. Agency for Healthcare Research and Quality: Priority Conditions and Populations Priority conditions

Priority populations

March 2010

JournalHEalthPolicy.indd 329

• Arthritis and nontraumatic joint disorders • Cancer • Cardiovascular disease, including stroke and hypertension • Dementia, including Alzheimer disease • Depression and other mental health disorders • Developmental delays, attention-deficit hyperactivity disorder, and autism • Diabetes mellitus • Functional limitations and disability • Infectious diseases, including HIV/AIDS • Obesity • Peptic ulcer disease and dyspepsia • Pregnancy, including preterm birth • Pulmonary disease/asthma • Substance abuse • Racial and ethnic minorities • Persons with disabilities • Children • Elderly • Patients with multiple and chronic conditions • Disadvantaged and/or under-represented populations for the specific condition being studied

for CER14 (Fig. 2). This framework groups investments and activities of CER into 4 major categories and serves as a useful point of departure for considering physical therapists’ efforts and roles in this area. Research Research is needed to better understand the comparative effectiveness of physical therapy interventions relative to other medical/surgical interventions and relative to other physical therapy interventions. To better understand the usefulness of physical therapy within the context of the larger health care system, the comparison of physical therapy interventions to other medical/ surgical interventions is important. Such research across disciplines is challenging, however, because it requires multidisciplinary, collaborative efforts that may take time to develop and that require understanding different disciplinary perspectives. Physical therapy also needs to look within its own practices to determine which interventions are most effective and for whom. Too often the term “physical therapy” has been used as a poorly described bundle of heterogeneous services, similar to the way the term “usual care” is applied. Physical therapist researchers also need to consider that certain interventions, although often delivered by physical therapists, are not necessarily specific to the profession. Interventions such as exercise and manipulation may be delivered by various health care providers. In addition to demonstrating the comparative effectiveness of such interventions, it would be useful to determine whether certain types of providers are more or less effective in delivering those interventions. The IOM report and AHRQ emphasize that CER can compare systems of care as well as specific treat-

Volume 90 Number 3 Physical Therapy ■ 329

2/17/10 2:57 PM

Health Policy in Perspective but will facilitate growth in this type of physical therapy researcher.

Figure 2. Strategic framework for Comparative Effectiveness Research (CER). Adapted with permission from Federal Council on Coordination of CER.14

ments. Comparing the effectiveness of treatments delivered by different providers is an example of comparing systems of care. Other systems analyses could include comparing the effectiveness of different combinations of treatments or of different numbers of provider visits. In addition to focusing CER on priority conditions and populations, some suggest targeting interventions that exhibit significant geographic variation in use.14 We know little about practice variation in physical therapy. Some of our research suggests that physician referral to physical therapy for the treatment of musculoskeletal conditions varies by census region,15,16 as does use of physical therapy by community-based Medicare beneficiaries.17 More research in this area would be useful to identify conditions and services to target for future research. Another area that will be particularly useful to target are the research gaps identified in newer systematic reviews. Planned CER

initiatives will endeavor to provide more rapid-turnaround research to fill these gaps and address policyrelevant research needs. Human and Scientific Capital Investing in the future of researchers and training individuals in CER is important to maintain the enterprise. Physical therapists are expanding beyond traditional PhD programs with more individuals pursuing degrees in health services research and epidemiology. This should continue to remain a priority for the profession. Because training in these areas opens the door to many career opportunities, one challenge will be keeping these individuals connected to the profession. Academic physical therapy program chairs need to recognize the value of having such individuals connected with their departments. Exposing professional students to the area of health services research, and to CER in particular, not only will increase their understanding of the value of this type of research

330 ■ Physical Therapy Volume 90 Number 3

JournalHEalthPolicy.indd 330

The Foundation for Physical Therapy’s recent establishment of a new research funding opportunity— the New Investigator Fellowship Training Initiative in Health Services Research—is one avenue for training of physical therapists in CER and health services research. The AHRQ will also be increasing training opportunities in CER over the next several years through its T32 mechanism for predoctoral and postdoctoral fellowships and through K12 institutional career development awards. The latter awards will provide institutions with several 75%-supported positions for faculty start-up funds and career development. A final human capital path is through the 46 Clinical Science and Translational Award (CTSA) institutions that have been established throughout the country via NIH funding. Each CTSA institution supports several research career development positions. The physical therapy profession should help disseminate information about these different training opportunities, keep abreast of new opportunities, and encourage graduate students and junior faculty to take advantage of these opportunities. CER Data Infrastructure Data infrastructure is a key aspect of CER and includes the development of patient registries and distributed health data networks. Currently, research too often consists of a one-time effort in which the data-gathering infrastructure is not maintained over time. An enhanced infrastructure will allow ongoing analyses of care patterns and outcomes over time. A patient registry is “…an organized system that uses observational data methods to collect uniform

March 2010

2/17/10 2:57 PM

Health Policy in Perspective data (clinical and other) to evaluate specified outcomes for a population defined by a particular disease, condition, or exposure and serves a predetermined scientific, clinical or policy purpose(s).”18 A distributed health data network, or federated data network, is a system that allows secure, remote analysis of a collection of geographically and organizationally separate databases that are treated as one entity and viewed through a single user interface.19 Such networks facilitate sharing and interchanges of data among autonomous databases, such as electronic health records located within different organizations. Distributed data networks are particularly appealing because they allow data holders to retain physical control over their data, thereby eliminating some of the barriers related to confidentiality and proprietary interests. The establishment of APTA Connect and APTA’s vision for a national outcomes database have been timely, strategic moves in the context of CER data infrastructure. The Focus on Therapeutic Outcomes (FOTO) database also has the potential to facilitate CER in physical therapy.20 Because electronic data collection serves as the basis for distributed health data networks and patient registries, the profession should strive to increase the adoption of electronic health records (EHRs). Health information technology is a rapidly moving field, and interoperability issues will be critical. In addition, the profession should keep abreast of the development of distributed data networks and registries that are relevant in regard to physical therapy populations and identify ways in which to partner with stakeholders involved in these networks and registries. Finally, the profession should consider how to partner with key stakeholder groups to develop a distributed data network or patient registry focused March 2010

JournalHEalthPolicy.indd 331

on issues and populations that are relevant to physical therapy. Dissemination and Translation of CER Research articles and federal reports are often lengthy, with substantial technical language. Providers need brief abstracts of the best available evidence. Professional societies can serve as one dissemination conduit, as can hospitals and outpatient practice administration. An as-yet underused but exciting dissemination tool will be the incorporation of CER findings into EHRs—for example, reminding providers of the most effective interventions for the patient’s condition. Although straightforward in concept, such dissemination tools will be challenging. We will need to keep information up-to-date, work with EHR vendors, and avoid overloading providers with prompts and reminders.

Conclusions A considerable amount of money and effort currently are being invested in building the foundation for a CER enterprise. Physical therapy will be well served by joining the effort. The time seems right for several reasons: • The evidence base for practice has grown enough that we can synthesize evidence across multiple studies for many physical treatments. Sometimes the evidence base will be sufficiently strong to arrive at conclusions regarding treatment effectiveness; in other cases, we will be able to arrive at much more specific recommendations for future research. • The movement for reform of the health care system is both an opportunity and a challenge to providers to demonstrate that their treatments are effective and provide value to the health care system.

• The rapid dissemination and interoperability of EHRs should speed both the conduct of comparative effectiveness research as well as its dissemination. Although the physical therapy profession has lagged somewhat behind other health professions in their use of EHRs, that will likely change in the very near future. • Patient-centered care has been and continues to be a primary focus of physical therapy. The physical therapist works with the patient to develop a plan of care and treatment goals that are consistent with the patient’s values and needs. This attribute of care will serve the profession well as we attempt to gain a better understanding of what works best, for whom, and when. We have seen many positive changes in the delivery of physical therapy in response to the evidence-based practice movement. CER has the potential to transform health care delivery even more dramatically. Getting in on the ground floor of the CER movement will accelerate advancement of the physical therapy profession, lead to the delivery of higher quality care, and ultimately improve the health of individuals with activity and functional limitations. J.K. Freburger, PT, PhD, is Research Associate and Fellow at the Cecil G. Sheps Center for Health Services Research, and Research Scientist at the Institute on Aging at the University of North Carolina, in Chapel Hill North Carolina. T.S. Carey, MD, MPH, is Director at the Cecil G. Sheps Center for Health Services Research, and Professor in the Departments of Medicine and Social Medicine. The authors would like to thank Kendra Heatwole Shank, MS, OTR/L for her assistance in the preparation of this manuscript. This work was partially supported through multiple task order contracts from the Agency for Healthcare Research and Quality to the RTIUNC Evidence-based Practice Center. DOI: 10.2522/ptj.2010.90.3.327

Volume 90 Number 3 Physical Therapy ■ 331

2/17/10 2:57 PM

Health Policy in Perspective References 1 American Recovery and Reinvestment Act of 2009, Pub L No. 111-5. 123 Stat. 115–521. 2 Congressional Budget Office. Research on the Comparative Effectiveness of Medical Treatments: Issues and Options for an Expanded Federal Role. December 2007. Available at: http://www.cbo.gov/ ftpdocs/88xx/doc8891/12-18-Compara tiveEffectiveness.pdf. Accessed on October 28, 2009. 3 Kaiser Permanente Foundation. Explaining Health Reform: What Is Comparative Effectiveness Research? Focus on Health Reform. October 2009. Available at: http:// www.kff.org/healthreform/upload/7946. pdf. Accessed October 28, 2009. 4 Olsen LA, Aisner D, McGinnis JM, eds. The Learning Healthcare System: Workshop Summary (IOM Roundtable on EvidenceBased Medicine). Washington, DC: National Academies Press; 2007. Available at: http://www.nap.edu/catalog.php?record_ id=11903#toc. Accessed October 28, 2009. 5 Committee on Comparative Effectiveness Research Prioritization, Board on Health Care Services, Institute of Medicine. Initial National Priorities for Comparative Effectiveness Research. Washington, DC: National Academies Press; 2009. Available at: http://books.nap.edu/openbook. php?record_id=12648. Accessed October 28, 2009. 6 Zhang W, Moskowitz RW, Nuki G, et al. OARSI recommendations for the management of hip and knee osteoarthritis, part II: OARSI evidence-based, expert consensus guidelines. Osteoarthritis Cartilage. 2008;16(2):137–162.

7 Freburger JK, Carey TS, Holmes GM, et al. Exercise prescription for chronic back or neck pain: who prescribes it? Who gets it? What is prescribed? Arthritis Rheum. 2009;61(2):192–200. 8 Wennberg JE, Fisher ES, Goodman DC, Skinner JS. Tracking the Care of Patients With Severe Chronic Illness: The Dartmouth Atlas of Health Care. Lebanon, NH: The Dartmouth Institute of Health Policy and Clinical Practice; 2008. Available at http://www.dartmouthatlas. org/atlases/2008_Chronic_Care_Atlas.pdf. Accessed October 28, 2009. 9 Fisher ES, Bynum JP, Skinner JS. Slowing the growth of health care costs—lessons from regional variation. N Engl J Med. 2009;360(9):849–852. 10 Fisher ES, Wennberg DE, Stukel TA, et al. The implications of regional variations in Medicare spending, part 2: health outcomes and satisfaction with care. Ann Intern Med. 2003;138(4):288–298. 11 Weinstein JN, Lurie JD, Olson PR, et al. United States’ trends and regional variations in lumbar spine surgery: 1992-2003. Spine. 2006;31(23):2707–2714. 12 Gluck ME. Research Insights: Incorporating Costs Into Comparative Effectiveness Research. Washington, DC: Academy Health; 2009. Available at: http://www. academyhealth.org/files/publications/ ResearchInsightsCER.pdf. Accessed October 28, 2009. 13 Agency for Healthcare Research and Quality. Effective Health Care Program. Available at: http://effectivehealthcare.ahrq. gov/. Accessed October 28, 2009.

332 ■ Physical Therapy Volume 90 Number 3

JournalHEalthPolicy.indd 332

14 Federal Coordinating Council for Comparative Effectiveness Research. Report to the President and Congress. Washington, DC: Department of Health and Human Services; 2009. Available at http://www.hhs. gov/recovery/programs/cer/cerannualrpt. pdf. Accessed October 28, 2009. 15 Freburger JK, Carey TS, Holmes GM. Physician referrals to physical therapists for the treatment of spine disorders. Spine J. 2005;85(5):530–541. 16 Freburger JK, Holmes GM, Carey TS. Physical referrals to physical therapy for the treatment of musculoskeletal conditions. Arch Phys Med Rehabil. 2003;84(12):1839– 1849. 17 Freburger JK, Holmes GM. Physical therapy use by community-based older people. Phys Ther. 2005;85(1):19–33. 18 Gliklich RE, Dreyer NA, eds. Registries for Evaluating Patient Outcomes: A User’s Guide. Rockville, MD: Agency for Healthcare Research and Quality; April 2007. AHRQ Publication No. 07-EHC001-1. 19 Pace WD, West DR, Valuck RJ, et al. Distributed Ambulatory Research in Therapeutics Network (DARTNet): Summary Report. Rockville, MD: Agency for Healthcare Research and Quality; July 2009. Available at: http://effectivehealthcare.ahrq.gov/ ehc/products/53/151/2009_0728DEcIDE_ DARTNet.pdf. Accessed October 28, 2009. 20 Deutscher D, Hart DL, Dickstein R, et al. Implementing an integrated electronic outcomes and electronic health record process to create a foundation for clinical practice improvement. Phys Ther. 2008;88:270–285. Epub 2007 Nov 27.

March 2010

2/12/10 5:48 PM

Special Communication The Added Value of Confidence Intervals Paul W. Stratford

S

ince its formal introduction in 1991, evidence-based medicine/practice has received considerable attention. Defined as “the conscientious, explicit, and judicious use of best evidence in making decisions about the care of individual patients,”1 evidence-based practice embraces the integration of best research evidence, clinical expertise, and patient values.2 Clinicians are active participants not only in applying their expertise, but also in seeking out and interpreting research evidence. To allow the optimal transfer of information from research report to clinical practice, researchers must present their findings in an easy-to-understand format that provides the maximum amount of information efficiently.

sonal beliefs, risk of an adverse event, cost, and the feasibility of providing the intervention, test, or measure in practice. Because clinicians—and researchers for that matter—are likely to have different opinions concerning the magnitude of a clinically important difference, it is essential that authors provide their results in a manner that informs the widest audience and that the audience is equipped to interpret the information. Confidence intervals (CIs) convey both statistical and clinical information to assist clinicians with their decision making. The following example is offered to illustrate the communicative advantage that CIs have over P values.

Clinical Illustration

When interpreting the results from studies investigating the merits of competing therapeutic interventions, the reliability or validity of clinical measurements, or the causal association of putative risk factors, clinicians and researchers are interested in the answers to 2 important questions: (1) Are the results likely due to chance? and (2) Are the findings clinically important? The former question considers statistical significance, and the latter question addresses clinical significance.

This illustration considers the results from 4 hypothetical randomized clinical trials. For each study, the research question was: In patients with low back pain of 8 to 12 weeks’ duration, is there a difference between interventions A and B in the proportion of successful outcomes following 2 weeks of treatment? A successful outcome was defined as a 2-point reduction in pain, as measured by an 11-point (0–10) numeric pain rating scale. The investigators formed the following hypotheses:

Statistical significance is dictated by tradition, with a critical P value of .05 typically being the requisite minimal value. Statistical significance is influenced by sample size, sample variability, and the magnitude of the observed effect. In contrast to the arbitrary standard for statistical significance, clinical importance is influenced by per-

Null hypothesis: There will be no difference in the proportion of successes (π) between groups A and B (Hnull: πA = πB).

March 2010

Alternate hypothesis: There will be a difference in the proportion of successes between groups A and B (Halternate: πA ≠ πB).

Consistent with the alternate hypothesis, a 2-tailed test of significance was applied, and the critical P value for rejecting the null hypothesis was set at P≤.05. The investigators reported P values and 95% CIs on the betweengroup difference in the proportion of patients with a successful outcome. To complete the vignette, suppose we are treating patients with intervention B; however, we would consider acquiring the skills necessary to implement treatment A if there is a good chance that it would truly increase the proportion of patients who would achieve a successful outcome by 0.15 (15%), which is considered, for the purposes of this example, to be the minimal clinically important difference (MCID). Two studies produced negative results (P>.05), and the other 2 studies yielded positive results (P<.05). Table 1 displays the results from the negative studies, and Table 2 provides the findings from the positive studies. Table 1 shows that the proportion of patients who had a successful outcome was identical for both studies (0.30 for treatment A, 0.25 for treatment B) and that the between-group differences in success proportions for studies 1 and 2 were not statistically significant (P>.05). In addition, the difference in the observed success proportions of 0.05 was much less than the designated MCID value of 0.15. Armed with this information alone, as is the case when investigators present P values only, 2 questions arise: (1) Is the sample size sufficiently large to detect a difference if one truly exists? and

Volume 90 Number 3 Physical Therapy ■ 333

The Added Value of Confidence Intervals Table 1.

Results From the Negative Studies Difference in Success (95% Confidence Interval)

Study No.

Sampe Size per Group

Treatment A (Successes/n)

Treatment B (Successes/n)

1

100

30/100=0.30

25/100=0.25

0.05 (−0.073, 0.173)

.428

2

200

60/200=0.30

50/200=0.25

0.05 (−0.037, 0.137)

.263

Study No.

Sampe Size per Group

Treatment A (Successes/n)

Treatment B (Successes/n)

3

30

18/30=0.60

9/30=0.30

0.30 (0.06, 0.54)

.037

4

100

60/100=0.60

30/100=0.30

0.30 (0.17, 0.43)

<.001

P

Table 2.

Results From the Positive Studies

(2) Is it likely that a study conducted with a larger sample size would yield a between-group difference as large as the defined MCID value? These questions can be answered by directing our attention to the CIs reported in the tables and shown graphically in the Figure. For study 1, the 95% CI extends from −.073 to .173. Not only does this interval include the value of zero, but it also includes the defined MCID value of 0.15. The answers to the questions posed above are: (1) the sample size is too small, and (2) we cannot rule out the possibility that the true difference between treatment groups is 0.15. Applying a similar reasoning process to the results for study 2, we see that the 95% CI does not include the defined MCID value of 0.15. Thus, although we cannot rule out that a true difference exists between interventions A and B, we can be reasonably certain that the magnitude of the true difference is less than 0.15, and a subsequent study is not necessary to answer our question concerning clinical importance.

In contrast to the results shown in Table 1, the observed difference in the proportion of successful outcomes for studies 3 and 4 reported in Table 2 is 0.30, and the results are statistically significant. This result can be interpreted as providing strong support for a true difference in success proportions between interventions A and B. However, these findings alone do not tell us whether it is likely that the true between-group difference represents a clinically important difference according to our standard. To gain an insight into clinical significance, we again direct our attention to the CIs. Referring to study 3, we see that the CI is quite wide and that it includes the value of 0.15. Thus, although this study provides strong evidence that a true difference exists in favor of treatment A, it has too few participants to help us determine whether the true difference meets or exceeds our defined MCID value of 0.15. Turning our attention to the CI reported for study 4, we see that the lower confidence limit of 0.17 exceeds our defined value

334 ■ Physical Therapy Volume 90 Number 3

Difference in Success (95% Confidence Interval)

P

for MCID. The results from study 4 can be interpreted as providing strong evidence in support of a clinically important difference in favor of treatment A. In the preceding examples, we considered a between-group difference in success proportions of 0.15 to be clinically important; others may disagree with this decision. The beauty of CIs is that they allow each individual to apply a unique value for MCID. For example, a colleague may consider a difference in success proportions of 0.20 to be clinically important. For this individual, the 2 negative trials reported in Table 1 virtually exclude the possibility that the treatment effect is as great as 0.20. However, neither of the 2 positive trials reported in Table 2 excludes the chance that the true treatment effect is less than 0.20. For this individual, a subsequent larger trial is necessary. The insights offered in the preceding scenarios are not new, but rather they emphasize the advantages of CIs described by Sim and Reid.3 They provided recommenMarch 2010

The Added Value of Confidence Intervals

dations concerning the reporting and interpretation of CIs that are as relevant today as they were a decade ago. PTJ’s Editorial Board encourages you to read the article by Sim and Reid. If you are a researcher, provide CIs; if you are clinician, take advantage of the rich information provided by CIs. P.W. Stratford, PT, MS, is Professor, School of Rehabilitation Science, and Associate Member, Department of Clinical Epidemiology and Biostatistics, McMaster University, Ontario, Canada. Address all correspondence to Mr Stratford at: stratfor@ mcmaster.ca.

References

Difference in proportion of success between patients

1

Figure. Difference in proportion of success between treatments. Red vertical line indicates the minimal clinically important difference (MCID). Horizontal lines represent 95% confidence intervals.

Sackett DL, Rosenberg WMC, Muir Gray JA, et al. Evidence-based medicine: what it is and what it isn’t. BMJ. 1996;312:71– 72. 2 Sackett DL, Straus SE, Richardson WS, et al. Evidence-based Medicine: How to Practice and Teach EBM. 2nd ed. Toronto, Ontario, Canada: Churchill Livingstone; 2000. 3 Sim J, Reid N. Statistical inference by confidence intervals: issues of interpretation and utilization. Phys Ther. 1999;79:186–195. [DOI: 10.2522/ptj.2010.90.3.333]

March 2010

Volume 90 Number 3 Physical Therapy ■ 335

Research Report Balance Impairment as a Risk Factor for Falls in Community-Dwelling Older Adults Who Are High Functioning: A Prospective Study S.W. Muir, BScPT, PhD, is Postdoctoral Fellow, Department of Medicine, Division of Geriatric Medicine, Schulich School of Medicine and Dentistry, University of Western Ontario, Parkwood Hospital, Room A-283, 801 Commissioners Rd East, London, Ontario, Canada N6C 5J1. Address all correspondence to Dr Muir at: susan.muir@ uwo.ca. K. Berg, PT, PhD, is Chair and Associate Professor, Department of Physical Therapy, University of Toronto, Toronto, Ontario, Canada. B. Chesworth, BScPT, MClScPT, PhD, is Assistant Professor, School of Physical Therapy, and Faculty of Health Sciences, Department of Epidemiology and Biostatistics, Schulich School of Medicine and Dentistry, University of Western Ontario. N. Klar, PhD, is Associate Professor, Department of Epidemiology and Biostatistics, Schulich School of Medicine and Dentistry, University of Western Ontario. M. Speechley, PhD, is Associate Professor, Department of Epidemiology and Biostatistics, Schulich School of Medicine and Dentistry, University of Western Ontario. [Muir SW, Berg K, Chesworth B, et al. Balance impairment as a risk factor for falls in communitydwelling older adults who are high functioning: a prospective study. Phys Ther. 2010;90:338 – 347.] © 2010 American Physical Therapy Association

Susan W. Muir, Katherine Berg, Bert Chesworth, Neil Klar, Mark Speechley

Background. Screening should have simple and easy-to-administer methods that identify impairments associated with future fall risk, but there is a lack of literature supporting validation for their use. Objective. The aim of this study was to evaluate the independent contribution of balance assessment on future fall risk, using 5 methods to quantify balance impairment, for the outcomes “any fall” and “any injurious fall” in community-dwelling older adults who are higher functioning.

Design. This was a prospective cohort study. Methods. A sample of 210 community-dwelling older adults (70% male, 30% female; mean age⫽79.9 years, SD⫽4.7) received a comprehensive geriatric assessment at baseline, which included the Berg Balance Scale to measure balance. Information on daily falls was collected for 12 months by each participant’s monthly submission of a falls log calendar.

Results. Seventy-eight people (43%) fell, of whom 54 (30%) sustained an injurious fall and 32 (18%) had recurrent falls (ⱖ2 falls). Different balance measurement methods identified different numbers of people as impaired. Adjusted relative risk (RR) estimates for an increased risk of any fall were 1.58 (95% confidence interval [CI]⫽1.06, 2.35) for self-report of balance problems, 1.58 (95% CI⫽1.03, 2.41) for one-leg stance, and 1.46 (95% CI⫽1.02, 2.09) for limits of stability. An adjusted RR estimate for an increased risk of an injurious fall of 1.95 (95% CI⫽1.15, 3.31) was found for self-report of balance problems.

Limitations. The study was a secondary analysis of data. Conclusions. Not all methods of evaluating balance impairment are associated with falls. The number of people identified as having balance impairment varies with the measurement tool; therefore, the measurement tools are not interchangeable or equivalent in defining an at-risk population. The thresholds established in this study indicate individuals who should receive further comprehensive fall assessment and treatment to prevent falls.

Post a Rapid Response or find The Bottom Line: www.ptjournal.org 338

f

Physical Therapy

Volume 90

Number 3

March 2010

Balance Impairment in Older Adults

A

lthough impairment in balance is acknowledged as a major predictor of falls, systematic reviews have been limited in recommending specific clinical assessment scales of balance.1– 4 There are many risk factor studies in which the results from univariate analysis have driven the regression modeling procedures, but few prospective epidemiologic studies have focused a priori on balance impairment as the exposure of interest. Additional limitations in the literature evaluating balance impairment as the exposure of interest include the failure to take account for multicollinearity among covariates and the inclusion of multiple measures of balance in the same regression model.5,6 Measures of balance may be highly related to other variables such as lower-extremity strength (force-generating capacity), and a thorough evaluation of the independent contribution of balance requires evaluating these other relationships to accurately determine the true magnitude of effect. Postural stability is a complex process that involves the rapid, automatic integration of information from the vestibular, somatosensory, visual, and musculoskeletal systems, in the presence of cognition, which includes attention and reaction time.7,8 The measurement tools used to evaluate balance in the clinical setting are a means of quantifying the working capacity of the sum of the components that enable postural stability. The goal of screening is to identify individuals who are at an elevated risk for falling and who should receive further assessment. Screening should have simple and easy-to-administer measurement tools that identify impairments associated with future fall risk. We hypothesized that the clinical assessment of balance can identify future fall risk even after the adjustment for other factors related to an March 2010

increased fall risk. Measurement scales that involve a difficult balance task (eg, using a reduced base of support and movement) should be the most successful, as they reduce ceiling effects. The primary objective of the study was to evaluate the independent contribution of the clinical assessment of balance on future fall risk, using 5 methods to quantify balance impairment, for the outcome of “any fall” in community-dwelling older adults who are higher functioning. A secondary objective was to perform the same analysis for the outcome of “any injurious fall.”

Method Participants This study was a secondary analysis of data from phase II of the Project to Prevent Falls in Veterans (PPFV). The PPFV, funded jointly by Veterans Affairs Canada and Health Canada, was approved by the University of Western Ontario Research Ethics Board for Health Sciences Research Involving Human Subjects. The sampling and data collection procedures for phase I of the PPFV have been described in detail elsewhere.9 Briefly, in phase I of the PPFV, a questionnaire was mailed to 3 simple random samples of addresses of 1,000 Canadian veterans of WWII and the Korean War living in southwest Ontario. Canadian veterans of WWII and the Korean War, as determined by Veterans Affairs Canada, and their caregivers who were living independently in the community and who were able to understand and provide responses to the questionnaire were eligible. Participants from 2 regions in phase I of the PPFV became eligible for participation in phase II, a risk factor modification trial. Participants from phase I who consented to be recontacted and who self-reported at least one modifiable risk factor for falling were stratified into 5 groups

by the number of modifiable risk factors from the questionnaire (lowerextremity weakness; 4 or more prescription medications; and balance, foot, and vision problems). Randomization was performed within each stratum to 1 of 2 groups: the specialized geriatric services (SGS) group or the family physician (FP) group. The SGS group received a comprehensive geriatric assessment performed by a geriatrician or physical therapist and was provided with individual recommendations for fall risk factor reduction. The intervention was the assessment and did not involve the provision of any follow-up treatment. A letter summarizing the risk factors identified on the mailed questionnaire was sent to the participants in the FP group, and a similar letter was sent to each participant’s family physician. Any treatment was left at the discretion of the family physician. People with no reported modifiable risk factors for falling on the questionnaire formed an open arm in the study. This group received the same comprehensive geriatric assessment as the SGS group and was given educational materials on fall prevention and healthy living. There were no statistically or clinically significant differences among the study arms in the proportion falling, the number of falls, or the time to first fall. Because there were no detectable differences on any outcome among the study groups, the opportunity existed to use the data collected to evaluate the association between balance and fall risk.

Available With This Article at ptjournal.apta.org • Audio Abstracts Podcast This article was published ahead of print on January 7, 2010, at ptjournal.apta.org.

Volume 90

Number 3

Physical Therapy f

339

Balance Impairment in Older Adults In phase II of the PPFV, 210 people received the comprehensive geriatric assessment. Four people who were less than 60 years of age were excluded from the analysis, ensuring all people in the sample were older than 60 years of age, and 2 people had incomplete balance measurement data. Twenty-two people did not have complete follow-up information on fall outcome data, leaving a sample of 182 participants for analysis in this study. Reasons for missing follow-up data were: no fall calendars submitted (n⫽15), dropped out because not interested in participating anymore (n⫽6), and deceased (n⫽1). Assessment The baseline comprehensive fall assessment utilized the basic version of the interRAI Community Health Assessment (CHA), a subset of the Minimum Data Set for Home Care (MDSHC) version 2.0.10,11 Reliability and validity of all items of the MDS-HC in community settings have been reported.12,13 A study supplement included the Berg Balance Scale (BBS), a valid and reliable scale for the assessment of balance in older adults.14 –16 The BBS consists of 14 balance tasks scored on a scale of 0 to 4, where 0 indicates the inability to perform a task and 4 indicates that the task is performed independently. The maximum possible score of 56 indicates no identified balance difficulties. The individual items of the BBS represent assessment of different aspects of balance, including maintaining a fixed position, dynamic balance, and movement over a fixed base. No single measurement tool can evaluate all aspects of or mastery of balance control, but it may be possible that a single tool can optimally identify people at an elevated risk for falling in a screening format. Individual items of the BBS, which have the potential to be independent mea340

f

Physical Therapy

Volume 90

surement tools to predict falls in a screening situation, and additional measures of postural stability from the mailed questionnaire and interRAI assessment were available within the data set for analysis. The present analysis, with a priori specified associations for exposure and confounding, evaluated whether simple and easy methods of assessing postural stability were independently associated with the outcome of any fall. The following clinical assessment methods were used to evaluate balance.

Limits of stability in standing. The participant was asked to lift an arm to 90 degrees with fingers stretched out in front and then, without changing foot position, reach as far forward as possible. The maximum distance, measured as the distance forward that the finger reached while the individual was in the most forward lean position, was recorded. A distance of less than 25.4 cm (10 in) was denoted as impairment. The validity and reliability of this test have been established.22,24,26 –29

One-leg stance with eyes open. Each participant was asked to stand on his or her preferred leg. The ability to assume the position independently and maintain it for longer than 10 seconds was recorded. The need for assistance to assume the starting position or inability to maintain the position for 10 seconds was considered impaired ability. The validity and reliability of this test have been established.17–22

Self-report of balance impairment. On a self-administered questionnaire, the following question was asked, “Sometimes people get dizzy or light-headed, and lose their balance. Other people report a loss of balance in their legs. Do you ever feel that you are losing your balance, other than when you feel dizzy or light-headed? By that, we mean do you feel the problem is in your legs rather than your head?”30 A “yes” response to this question denoted impairment.

Tandem stance with eyes open. The heel of one foot was placed in front of and touching the toes of the other foot. The ability to independently assume this position and maintain it for 30 seconds was recorded. The need of assistance to assume the starting position or inability to remain in balance for 30 seconds was considered impaired ability. The validity and reliability of this test have been established.18,22–25 Observational gait assessment. The participant was asked by the assessor to ambulate at a comfortable speed, and the presence of any of the following features during an observational gait assessment was used to denote an unsteady or impaired gait: steppage; leg-length discrepancy; waddling, antalgic, ataxic, spastic, or frontal-lobe gait; or vestibular or parkinsonian movements.

Number 3

A new variable was created as the summed number of balance tests with impairment from the individual tests of one-leg stance, tandem stance, limits of stability, and unsteady gait. Each test represents an evaluation of a different aspect and challenge to postural stability; the association of the composite may be superior to each test alone. Data Collection Prospective information on daily falls was collected for 12 months by each participant’s monthly submission of a falls log calendar. A fall was defined as the person coming to rest unintentionally on the floor or ground. An injurious fall was defined as a fall resulting in an injury that required the person to see a physician. Participants who indicated falling in the previous month were March 2010

Balance Impairment in Older Adults interviewed by telephone to obtain detailed information about the specifics of the fall. Data collection for the baseline comprehensive geriatric assessment was started in May 2002, and collection of 1-year follow-up information on prospective falls was completed in January 2004.

the Hosmer-Lemeshow test for goodness of fit. As falls were a common outcome, the adjusted odds ratios will overestimate the RR; thus, a modified Poisson regression model was applied to the final logistic models to directly obtain adjusted relative risk values.32

Data Analysis Descriptive statistics for study participant characteristics using data from the questionnaire and comprehensive geriatric assessment, including demographics and measures of functional ability, were calculated.

The presence of biologic interaction, as described by Rothman and Greenland,33 among balance exposure, history of falls, and number of prescription medications and their association with the outcomes of any fall and any injurious fall was evaluated. The calculation of the biologic interaction necessitated the creation of a composite variable with 4 levels for balance exposure/history of falls and balance exposure/number of prescription medications: (ab, unexposed to both factors; Ab, exposed to non– balance factor only; aB, exposed to balance factor only; and AB, exposed to both non– balance factor and balance factor).34 The RR was calculated for each category using a modified Poisson regression adjusting for age, sex, and treatment group in the PPFV. A biologic interaction effect was defined as the departure from additivity of the absolute effects, and the excess RR caused by the interaction (RERI) of the 2 terms was calculated using the equation notation from Hosmer and Lemeshow34:

First, a univariate analysis between each measure of balance and the dependent variable (any fall, treated as a dichotomous variable) was performed to generate unadjusted relative risk (RR) estimates and 95% confidence intervals (CIs). Adjustment for confounding, based on clinical significance and previously identified fall risk factors, was accomplished by controlling for age (ⱖ80 years), sex, number of prescription medications (ⱖ4), history of falls in the previous 12 months, and the treatment group in the PPFV. Analysis of multicollinearity among the independent variables was performed using the phi coefficient (⌽) to test for correlation among categorical variables. For pairs of variables that were highly correlated (⌽ⱖ.30),31 only one variable was entered into the modeling procedure, and then the effect of placing correlated variables in the same model was evaluated. The same analyses were repeated for the outcome of any injurious fall, also treated as a dichotomous (yes/no) variable. The presence of multiplicative effect measure modification was evaluated for each model, prior to the calculation of adjusted effect measure estimates, using logistic regression. Regression diagnostics were performed on the full adjusted analyses using March 2010

共1兲 RERI ⫽ RR (AB) ⫺ RR (Ab) ⫺ RR(aB) ⫹ 1 An RERI of zero means no interaction, but an RERI of 0.5 would indicate that, because of interaction between the 2 variables, the RR is 0.5 greater than expected, based on the addition of the 2 risks.34 The percentage of falls among those participants with both exposures that was attributable to the interaction (AP(AB)%) was calculated as follows34:

共2兲 AP(AB)% ⫽ (RERI/RR(AB)) ⫻ 100% All analyses were performed using SAS, version 8.2.*

Results Baseline characteristics of the sample are presented in Table 1. In the complete sample, the average age was 79.9 years (SD⫽4.7) and 127 participants (70%) were male. Over the 1-year follow-up, 78 people (43%) fell, of whom 54 (30%) sustained an injurious fall and 32 (18%) had recurrent falls (ⱖ2 falls). Different balance measurement tools identified different numbers of people as impaired (Tab. 2). The assessment of both one-leg stance and tandem stance, which produce the narrowest base of support, identified the largest number of participants as having deficits (63% and 43%, respectively). The distribution of deficits per score level of the sum of balance impairments variables were: (1) 85% had impairment in one-leg stance; (2) 88% had impairment in one-leg stance and tandem stance; and (3) 64% had combined impairment in one-leg stance, tandem stance, and limits of stability, and another 28% had combined impairment in one-leg stance, tandem stance, and unsteady gait. Fallers had a higher prevalence of all measured impairments compared with nonfallers. A trend of decreasing total BBS score was demonstrated, with an increase in number of impaired test results. In unadjusted analyses, each method of assessing balance demonstrated that impairment was associated with an increased risk for any fall, with only tandem stance not statistically significant. Multiple measures of postural stability were not included in * SAS Institute Inc, PO Box 8000, Cary, NC 27513.

Volume 90

Number 3

Physical Therapy f

341

Balance Impairment in Older Adults Table 1. Baseline Characteristics Presented for the Whole Sample (n⫽182) and Then Stratified by Fall Status at the End of the Study Fall Status at End of Study Variable Age (y), X (SD) Sex (male) History of falling in the previous 3 mo

Whole Sample (nⴝ182)

Fallers (nⴝ78)

Nonfallers (nⴝ104)

79.9 (4.7)

81.0 (3.8)

79.0 (5.1)

127 (70%)

60 (77%)

67 (64%)

46 (25%)

31 (40%)

15 (14%)

50.7 (7.7)

48.9 (9.1)

52 (6.1)

25 (14%)

18 (23%)

7 (7%)

133 (73%)

64 (82%)

69 (66%)

Clinical assessmenta Berg Balance Scale score, X (SD) Lower-extremity weakness No. of prescription medications (ⱖ4) Vision impairment

18 (10%)

8 (10%)

10 (10%)

Use of mobility device (cane or walker)

23 (13%)

12 (15%)

11 (11%)

Unsteady gait

31 (17%)

18 (23%)

13 (13%)

9 (5%)

5 (6%)

4 (4%)

18 (10%)

11 (14%)

7 (7%)

2 (1%)

2 (3%)

0 (0%)

11 (14%)

7 (9%)

4 (4%)

Cognitive impairment Fear of falling Decrease in basic activities of daily living in previous 3 mo Any dependence in basic activities of daily living a

Data on participant characteristics were obtained from a comprehensive geriatric assessment using the interRAI Community Health Assessment basic version, a study-specific supplement for prescription medications, and the Berg Balance Scale. Lower-extremity weakness was defined as the inability to stand up from a chair without using the arms of the chair. Vision impairment was defined as the inability to read fine detail in adequate light, with glasses if needed, or the presence of eye pathology (eg, cataracts, glaucoma, macular degeneration, diabetic retinopathy). Use of mobility device was defined as use of any aid for primary mode of locomotion; an unsteady gait was assessed with an observational analysis of gait quality. Cognitive impairment was defined as anything less than participants being able to make independent decisions that were consistent, reasonable, and safe in organizing their day. Fear of falling was defined as the person reported limiting going outdoors due to fear of falling. Dependence in basic activities of daily living was defined as anything less than fully independent in the activities of bed mobility, locomotion at home, dressing of upper body, dressing of lower body, personal hygiene, eating, bathing, and toilet use.

the same regression model due to the variables being highly correlated. The adjusted RR estimates demonstrated a statistically significant association with increased risk for any fall for self-report of balance problems (RR⫽1.58; 95% CI⫽1.06, 2.35), one-leg stance (RR⫽1.58; 95% CI⫽1.03, 2.41), and limits of stability (RR⫽1.46; 95% CI⫽1.02, 2.09) (Tab. 3). For the outcome of any injurious fall, although all RR estimates were associated with an increased risk, only the RR estimate for self-report of balance problems (RR⫽1.95; 95% CI⫽1.15, 3.31) was statistically significant. The number of tests positive for an impairment (one-leg stance, tandem stance, limits of stability, and unsteady gait) demonstrated a statistically significant association per unit increase in the score (from 0 to 4) for both outcomes, such that for each test with 342

f

Physical Therapy

Volume 90

impairment, the risk for any fall and for any injurious fall increased by 20% and 23%, respectively. The presence of multiplicative effect measure modification was evaluated, and the estimated effect of interaction was not statistically significant for either outcome of any fall or any injurious fall. The presence of biological interaction, or additive interaction, was evaluated between the balance exposures that were significant in adjusted analyses and the risk factors of history of falling in the previous 12 months and the number of prescription medications (ⱖ4). Due to the interaction, the RR was greater than expected, based on the addition of the 2 factors alone for the outcome of any fall: self-report of balance problems and history of falls, self-report of balance problems and number of prescription medications,

Number 3

one-leg stance and number of prescription medications, and limits of stability and number of prescription medications (Tab. 4). For the outcome of any injurious fall, the RR was greater than expected for the 2-factor combinations of self-report of balance problems and history of falls and self-report of balance problems and number of prescription medications. A sensitivity analysis to evaluate the robustness of the measures of association using the best- and worst-case scenarios was performed by assuming that all individuals with missing fall outcome data either had or did not have the outcome of falling. The magnitude of association in the adjusted analyses did not change by greater than 10%, except for unsteady gait, which changed by 13% assuming all participants with missMarch 2010

Balance Impairment in Older Adults Table 2. Frequency of Impairment in Balance by Clinical Measurement Scale Assessed at Baseline of the Study for the Complete Sample and Stratified by Fall Status at the End of the Study, and Mean Total Berg Balance Scale Scores for People Identified as Impaired and Nonimpaired in Each Category No. of Participants With Impairment

Balance Measure

Mean Total Berg Balance Scale Scores, Impaired (SD)/ Nonimpaired (SD)

Complete Sample (nⴝ182)

Fallers (nⴝ78)

Nonfallers (nⴝ104)

Absolute Risk for Any Fall Among Those Identified With Impairment on the Test

Self-report of balance problemsa

45.2 (8.1)/53.2 (4.2)

53 (30%)

30 (30%)

23 (23%)

57%

One-leg stance

47.9 (8.3)/55.3 (2.9)

114 (63%)

58 (74%)

56 (54%)

51%

Tandem stance

45.6 (8.8)/54.6 (3.1)

79 (43%)

39 (50%)

40 (39%)

49%

Limits of stability

42.0 (9.8)/53.4 (4.2)

43 (24%)

24 (31%)

19 (18%)

56%

Unsteady gait

38.6 (9.3)/53.2 (4.2)

31 (17%)

18 (23%)

13 (13%)

58%

0

55.7 (1.3)

63 (35%)

20 (26%)

43 (41%)

32%

1

53.8 (3.7)

34 (19%)

15 (19%)

19 (18%)

44%

2

50.7 (3.2)

41 (23%)

18 (23%)

23 (22%)

44%

3

45.8 (6.6)

25 (14%)

12 (15%)

13 (13%)

48%

4

34.8 (8.7)

19 (10%)

13 (17%)

6 (6%)

68%

Sum of balance impairments

Fall Status at End of Study

b

a

Sample size n⫽177 due to 5 participants who did not complete this information on a mailed questionnaire from phase I of the Project to Prevent Falls in Veterans study; Berg Balance Scale is a functional measure of balance status with a maximum score of 56, indicating no impairment. b The number of clinical tests showing impairment on one-leg stance, tandem stance, limits of stability, and unsteady gait.

Table 3. Unadjusted and Adjusted Relative Risk Estimates (With 95% Confidence Intervals) for Balance Impairment as Measured by 5 Different Methods on the Outcomes of “Any Fall” and “Any Injurious Fall” Calculated From a Modified Poisson Regression Any Fall

a b c

Any Injurious Fall

Balance Measurement

Unadjusted Analysis

Fully Adjusted Analysisa

P Value for Fully Adjusted Model

Self-report of balance problems

1.58 (1.14, 2.19)

1.58 (1.06, 2.35)

.024b b

Unadjusted Analysis

Fully Adjusted Analysisa

P Value for Fully Adjusted Model

1.76 (1.14, 2.73)

1.95 (1.15, 3.31)

.013b

One-leg stance

1.73 (1.15, 2.61)

1.58 (1.03, 2.41)

.035

Tandem stance

1.30 (0.93, 1.82)

1.26 (0.76, 1.90)

.18

1.55 (0.93, 2.59)

1.35 (0.79, 2.31)

.27

1.40 (0.90, 2.19)

1.46 (0.85, 2.49)

.17

Limits of stability

1.44 (1.02, 2.01)

1.46 (1.02, 2.09)

.039b

1.49 (0.94, 2.36)

1.54 (0.94, 2.51)

.084

Unsteady gait

1.46 (1.02, 2.09)

1.24 (0.86, 1.80)

.25

1.71 (1.07, 2.73)

1.51 (0.90, 2.54)

.12

Sum of balance impairmentsc (per unit increase in score)

1.18 (1.05, 1.32)

1.20 (1.04, 1.39)

.013b

1.20 (1.02, 1.40)

1.23 (1.01, 1.49)

.035b

Adjusted for age (ⱖ80 years), sex, number of prescription medications (ⱖ4), history of falls in the previous 12 months, and treatment group. Statistically significant at P⬍.05. The number of clinical tests showing impairment on one-leg stance, tandem stance, limits of stability, and unsteady gait, for a possible score of 0 to 4.

ing data did not fall. The measures that were statistically significant in the final adjusted analyses remained statistically significant in the scenario that people with missing fall outcome data did not fall for both fall outcomes. March 2010

Discussion The clinical assessment of balance is used as a means for measuring the integrity of the postural stability system, which involves the integration of information from somatosensory, musculoskeletal, visual, and vestibu-

lar systems and cognition. This study demonstrated the clinical assessment of balance impairment using simple and easy-to-administer measures encompassing self-report of balance problems, one-leg stance, and limits of stability, which were

Volume 90

Number 3

Physical Therapy f

343

Balance Impairment in Older Adults Table 4. Values of Excess Relative Risk and Percentage of Falls Among Study Participants With Both Exposures That Was Attributable to Their Interaction

tering multiple clinical assessment scales also may be seen as a limitation for implementation in a screening situation, but it does provide a more comprehensive evaluation of different aspects of postural stability, and the number of tests with impairment is associated with both fall outcomes.

Some prospective studies have shown the assessment of one-leg stance not to be an independent predictor of future falls.35– 40 There are some methodological issues with the regression modeling procedures that may have affected the findings. The lack of an a priori objective to evaluate one-leg stance, a variation in the threshold used to denote impairment, and lack of evaluation of multicollinearity among covariates may have influenced this measurement tool being found not significant in adjusted analyses. Conversely, Vellas et al21 did find the assessment of one-leg stance time to be an independent predictor of injurious falls. The present study adds further information to support the assessment of one-leg stance time as independently associated with falls in a sample of individuals who are high functioning. Bergland et al40 found one-leg stance was not associated with injurious falls in a sample of older women.

Importantly, not all of the methods of evaluating balance impairment are independently associated with future falls. The use of an observational gait assessment and tandem stance were not statistically associated with an increased risk for any fall and should be discouraged from use in a screening situation to identify individuals who are at risk for falls. If the intention of assessment was to identify risk for an injurious fall, none of the clinical tests was statistically significantly associated with this outcome. This study also demonstrated that the number of people identified as having balance impairment varied with the measurement tool used to clinically assess balance. The measurement tools, therefore, are not interchangeable or equivalent in defining an at-risk population. This information has implications for clin-

The self-report of balance problems is on par with clinical measures of balance as a predictor of future fall risk. The self-report of balance problems was statistically significantly associated with history of falls, both for any fall and for any recurrent falls, and may not add additional information than that obtained from a report of a fall history. That said, the self-report of balance problems appears to be an important piece of information that should be sought during a self-report history for its association with any fall and any injurious fall. A positive report of problems should be enough to warrant further comprehensive evaluation and may be such an important factor, as it encompasses problems experienced over a prolonged period of time and in many situations that cannot be captured in a single test.41

Excess Relative Riska

Percentage of Falls Attributable to the Interaction

Self-report of balance problems ⫹ history of falls

0.06

2

Self-report of balance problems ⫹ no. of prescription medications (ⱖ4)

0.47

20

One-leg stance ⫹ no. of medications (ⱖ4)

0.19

7

Limits of stability ⫹ no. of prescription medications (ⱖ4)

0.50

19

Risk Factors Outcome: any fall

Outcome: any injurious fall Self-report of balance problems ⫹ history of falls

0.22

5

Self-report of balance problems ⫹ no. of prescription medications (ⱖ4)

0.95

25

a Relative risk estimate adjusted for age (ⱖ80 years), sex, and treatment group in the Project to Prevent Falls in Veterans.

shown to be independent predictors of any fall over the subsequent year. None of the clinical assessment scales alone was associated with an injurious fall, although the self-report of balance problems was associated with an increased risk. When multiple clinical assessment scales were considered together, the risk for any fall or any injurious fall increased as the number of tests indicating impairment increased. As systematic reviews have been limited in their ability to recommend specific measurement tools, this study presents important information that can be transferred directly to clinical practice to facilitate the identification of communitydwelling older adults at an elevated risk for falling. As this population of older adults was high functioning with very few functional limitations, the test maneuvers that identified future fall risk were challenging tasks that would decrease a ceiling effect to facilitate exposing deficits. Additionally, the time requirement to administer the complete BBS may not be possible for a screening situation, but the identified task items could be performed in this situation. Adminis344

f

Physical Therapy

Volume 90

ical practice, as well as for the comparison of results across studies.

Number 3

March 2010

Balance Impairment in Older Adults Covinsky et al42 also found that the self-report of either dizziness or balance problems was associated with increased fall risk. They found an odds ratio of 1.83, which may overestimate the RR, and the inclusion of symptoms of dizziness does not separate problems of balance from potential cardiac-mediated symptoms. The findings from our study reinforce the importance, both for its inclusion and its diagnostic weight, of the self-report of balance problems in screening communitydwelling older adults. Testing the limits of stability in this study was equivalent to the Functional Reach Test (FRT). The present findings are consistent with previous work that showed the FRT is associated with future fall risk.27 The FRT, evaluated as a continuous variable of the maximum reaching distance (in centimeters), has not been found to be associated with falls in adjusted analyses.37,40,43 Duncan et al27 found a distance of less than 25.4 cm as the critical threshold that denotes the onset of compromised ability, a finding corroborated in the present study. The regression modeling maintained power to find an association, but potential reasons for the lack of an association include: validity of the measurement tool; the threshold value; chance; balance impairment being on the causal pathway for falls, with a resultant attenuation of association; and the inability of the measurement tool to detect a response to a challenge of the postural control systems. The analyses in this study used an a priori causal model where balance was treated as the sum result of multiple systems; control over confounders deliberately did not include variables such as vision impairment or lower-extremity muscle weakness, as their effects are subsumed within the measure of balance. Power was acceptable for the outMarch 2010

come of any injurious fall, but sensitivity analyses did not change the magnitude and statistical significance of the findings. We have confidence in the results found for this outcome. The evaluation of biologic interaction was a novel contribution to the falls literature, establishing that the joint effects of some variables exceed that expected from the sum of the individual components. Information on biologic interaction has important implications for the prevention and treatment of falls. If the joint additive effect of 2 factors is in excess of the sum of the individual effects, then reduction in either factor would reduce the risk of the other factor producing falls. The strength of the interaction was not consistent across the measures of balance impairment. The self-report of balance problems and the limits of stability when each measure was combined with the measure of polypharmacy produced the strongest evidence of effect measure modification. This finding suggests that a review of medications should be a standard procedure to limit fall risk, particularly in people with identified balance impairments. The present study’s participants represented a relatively highfunctioning group of older adults with few functional limitations, although fall events were still a common occurrence over the follow-up period. Risk for any fall in this sample (43%) was higher than the population average of 30% to 35%. Risk for any fall may have been higher in this group of older adults because they participated at their usual level of functioning, not curtailed or modified activity, which may have exposed them to more situations that could result in a fall. In our study, we used a general definition of an injurious fall, including potentially minor and major injuries, which resulted in

a fall risk of 30%. Comparison with other studies on fall injuries is difficult due to the many ways that injuries are reported in the literature. Bergland and Wyller,40 however, found that 24% of falls resulted in an injury, which is comparable to values obtained in our study, but is surprising as their sample comprised only older women. A potential limitation of the present study is that prospective falls data were derived from an intervention study. The participants’ exposure to falls prevention information may have produced a decrease in the number of falls, attenuating the association between balance impairments and new falls and producing conservative estimates. The use of data from 2 arms of an intervention study is not a design flaw, as there was no detected treatment effect from the intervention, which provides reasonable grounds to combine data for prognostic research.42 The treatment group in the PPFV was included as a covariate in the adjusted analyses to fully account for any differences that the intervention group may have introduced. Sensitivity analysis including and excluding the PPFV treatment group variable did not change the statistical significance of associations or the magnitude of association in any important way. Another potential limitation is the lack of generalizability of the results to the target population of all older adults at risk for falling. First, the sample may not be representative of the general population of community-dwelling older adults because they were volunteers. The men were from a representative sample of male veterans, but they may have been different from their contemporaries who were not veterans due to possible unique health consequences from active military service. This study used the gold standard for

Volume 90

Number 3

Physical Therapy f

345

Balance Impairment in Older Adults collecting fall occurrence information, but there is still a possibility of underreporting of fall events, in particular injurious falls. The self-report of injurious falls was not validated with medical records or documentation of a clinic visit for medical attention, and thus the RR estimates may be conservative. As all individuals volunteered to participate in the study, we felt there would not be a reporting bias by sex, and we are unaware of any research that would suggest male veterans would be differentially inclined to underreport or overreport falls compared with the average male older adult. Participants were aware the study was supported by Veterans Affairs Canada, and this partnership may have further ensured accurate reporting of events. The strengths of the study include the prospective design and the use of a reliable method to collect falls data in a large sample of communitydwelling older adults.44 The contact with study participants when falls were identified on calendars enhanced information collection, and the contact with participants who did not submit calendars aided in maximizing follow-up of study participants. Finally, all participants received the same standardized, comprehensive geriatric assessment at baseline.

Conclusions The self-report of balance problems, the one-leg stance test, and the limits of stability test are associated with increased risk for any fall in community-dwelling older adults who are high functioning. The selfreport of balance problems also is strongly associated with an individual sustaining an injurious fall. Balance impairment in the presence of polypharmacy increases fall risk beyond simple addition of the individual risks producing biologic interaction. A finding of impairment using 346

f

Physical Therapy

Volume 90

the thresholds established in this study for these simple and easy-toadminister screening measures indicates an individual who should be referred for further comprehensive fall risk assessment and interventions to prevent falls. Dr Muir provided concept/idea/research design and data analysis. All authors provided writing. Dr Speechley provided data collection and fund procurement. A poster presentation of this research was given at the XIXth International Congress of the International Association of Gerontology and Geriatrics; July 5–9, 2009; Paris, France. This study was a secondary analysis of data from phase II of the Project to Prevent Falls in Veterans (PPFV). The PPFV, funded jointly by Veterans Affairs Canada and Health Canada, was approved by the University of Western Ontario Research Ethics Board for Health Sciences Research Involving Human Subjects. This article was received May 19, 2009, and was accepted September 15, 2009. DOI: 10.2522/ptj.20090163

References 1 Scott V, Votova K, Scanlan A, Close J. 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. 2 Perell KL, Nelson A, Goldman RL, et al. Fall risk assessment measures: an analytic review. J Gerontol A Biol Sci Med Sci. 2001;56:M761–M766. 3 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;43:664 – 672. 4 National Institute of Clinical Excellence. Clinical Guideline 21: The Assessment and Prevention of Falls in Older People, 2004. Available at: www.nice.org.uk. Accessed June 21, 2005. 5 Boulgarides LK, McGinty SM, Willett JA, Barnes CW. Use of clinical and impairment-based tests to predict falls by community-dwelling older adults. Phys Ther. 2003;83:328 –339. 6 Brauer SG, Burns YR, 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. 7 Horak FB. Postural orientation and equilibrium: what do we need to know about neural control of balance to prevent falls. Age Ageing. 2006;35(suppl 2):ii7–ii11.

Number 3

8 Lord S, Sherrington C, Menz H, Close J. Falls in Older People: Risk Factors and Strategies For Prevention. Cambridge, United Kingdom: Cambridge University Press; 2007. 9 Speechley M, Belfry S, Borrie MJ, et al. Risk factors for falling among communitydwelling veterans and their caregivers. Can J Aging. 2005;24:261–274. 10 interRAI Community Health Assessment (CHA). Available at: www.interrai.org. Accessed May 6, 2007. 11 Morris JN, Bernabei R, Ikegami N, et al. RAI-Home Care (RAI-HC) Assessment Manual for Version 2.0: Primer on Use of the Minimum Data Set–Home Care (MDS-HC) Version 2.0 and the Client Assessment Protocols (CAPS), for Use in Canada and the United States. Washington, DC: interRAI Corp; 2002. 12 Landi F, Tua E, Onder G, et al. Minimum data set for home care: a valid instrument to assess frail older people living in the community. Med Care. 2000;38: 1184 –1190. 13 Morris JN, Fries BE, Steel K, et al. Comprehensive clinical assessment in community setting: applicability of the MDS-HC. J Am Geriatr Soc. 1997;45:1017–1024. 14 Berg KO, Wood-Dauphine´e SL, Williams JI, Gayton D. Measuring balance in the elderly: preliminary development of an instrument. Physiother Can. 1989;41: 304 –311. 15 Berg KO, Maki BE, Williams JI, et al. Clinical and laboratory measures of postural balance in an elderly population. Arch Phys Med Rehabil. 1992;73:1073–1080. 16 Berg KO, Wood-Dauphine´e SL, Williams JI, Maki B. Measuring balance in the elderly: validation of an instrument. Can J Public Health. 1992;83(suppl 2):S7–S11. 17 Briggs RC, Gossman MR, Birch R, et al. Balance performance among noninstitutionalized elderly women. Phys Ther. 1989;69:748 –756. 18 Iverson BD, Gossman MR, Shaddeau SA, Turner ME Jr. Balance performance, force production, and activity levels in noninstitutionalized mean 60 to 90 years of age. Phys Ther. 1990;70:348 –355. 19 Frandin K, Sonn U, Svantesson U, Grimby G. Functional balance tests in 76-year-olds in relation to performance, activities of daily living and platform tests. Scand J Rehabil Med. 1995;27:231–241. 20 Tinetti ME. Performance-oriented assessment of mobility problems in elderly patients. J Am Geriatr Soc. 1986;34:119 –126. 21 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. 22 Franchignoni F, Teslo L, Martino MT, Ricupero C. Reliability of four simple, quantitative tests of balance and mobility in healthy elderly females. Aging Clin Exp Res. 1998;10:26 –31. 23 Bergland A, Jarnlo GB, Laake K. Validity of an index of self-reported walking for balance and falls in elderly women. Adv Physiother. 2002;4:65–73.

March 2010

Balance Impairment in Older Adults 24 Murphy MA, Olson SL, Protas EJ, Overby AR. Screening for falls in communitydwelling elderly. J Aging Phys Act. 2003; 11:66 – 80. 25 Winograd CH, Lemsky CM, Nevitt MC, et al. Development of a physical performance and mobility examination. J Am Geriatr Soc. 1994;42:743–749. 26 Duncan PW, Weiner DK, Chandler J, Studenski S. Functional reach: a new clinical measure of balance. J Gerontol A Biol Sci Med Sci. 1990;45:M192–M197. 27 Duncan PW, Studenski S, Chandler J, Prescott B. Functional reach: predictive validity in a sample of elderly male veterans. J Gerontol A Biol Sci Med Sci. 1992;47: M93–M98. 28 Franzen H, Hunter H, Landreth C, et al. Comparison of functional reach in fallers and nonfallers in an independent retirement community. Phys Occup Ther Geriatr. 1998;15:33– 40. 29 Weiner DK, Duncan PW, Chandler J, Studenski S. Functional reach: a marker of physical frailty. J Am Geriatr Soc. 1992; 40:203–207. 30 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. 31 Fleiss JL. Statistical Methods for Rates and Proportions. 2nd ed. New York, NY: John Wiley & Sons; 1981.

March 2010

32 Zou GY. A modified Poisson regression approach to prospective studies with binary data. Am J Epidemiol. 2004;159: 702–706. 33 Rothman KJ, Greenland S. Modern Epidemiology. 2nd ed. Philadelphia, PA: Lippincott Williams & Wilkins; 1998. 34 Hosmer DW, Lemeshow S. Confidence interval estimation of interaction. Epidemiology. 1992;3:452– 456. 35 Koski K, Luukinen H, Laippala P, Kivela SL. Risk factors for major injurious falls among the home-dwelling elderly by functional abilities. Gerontology. 1998;44: 232–238. 36 Northridge ME, Nevitt MC, Kelsey JL. Nonsyncopal falls in the elderly in relation to home environment. Osteoporos Int. 1996; 6:249 –255. 37 Hill K, Schwarz J, Flicker L, Carroll S. Falls among healthy, community-dwelling older women: a prospective study of frequency, circumstances, consequences and prediction accuracy. Aust N Z J Public Health. 1999;23:41– 48. 38 Vellas BJ, Wayne SJ, Garry PH, Baumgartner RN. A two-year longitudinal study of falls in 482 community-dwelling elderly adults. J Gerontol A Biol Sci Med Sci. 1998;53A:M264 –M274. 39 Nevitt MC, Cummings SR, Kidd S, Black D. Risk factors for recurrent nonsyncopal falls. JAMA. 1989;261:2663–2668.

40 Bergland A, Wyller TB. Risk factors for serious fall related injury in elderly women living at home. Inj Prev. 2004;10:308 –313. 41 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. 42 Covinsky KE, Kahana E, Kahana B, et al. History and mobility exam index to identify community-dwelling elderly persons at risk of falling. J Gerontol A Biol Sci Med Sci. 2001;56:M253–M259. 43 Delbaere K, Van den Noortgate N, Bourgois J, et al. The Physical Performance Test as a predictor of frequent fallers: a prospective community-based cohort study. Clin Rehabil. 2006;20:83–90. 44 Ganz DA, Higashi T, Rubenstein LZ. Monitoring falls in cohort studies of community-dwelling older people: effect of the recall interval. J Am Geriatr Soc. 2005;53:2190 –2194.

Volume 90

Number 3

Physical Therapy f

347

Research Report

An Exercise and Education Program Improves Well-Being of New Mothers: A Randomized Controlled Trial Emily Norman, Margaret Sherburn, Richard H. Osborne, Mary P. Galea E. Norman, BPhysio, PGrad Cert (Exercise for Women), MPhysio, is Grade 4 Physiotherapist (Continence and Women’s Health) and Manager of Allied Health Clinical Research, Eastern Health, Ferntree Gully, Victoria, Australia. M. Sherburn, BAppSci (Physio), MWomen’s Health, PhD, is Senior Lecturer, School of Physiotherapy, University of Melbourne, Parkville, Victoria, Australia. R.H. Osborne, BSc, PhD, is Professor of Public Health, School of Health and Social Sciences, Deakin University, Burwood, Victoria, Australia. M.P. Galea, BAppSci (Physio), BA, PhD, is Professor of Clinical Physiotherapy, School of Physiotherapy, University of Melbourne, Grattan Street, Parkville, Victoria 3010, Australia. Address all correspondence to Dr Galea at: [email protected]. [Norman E, Sherburn M, Osborne RH, Galea MP. An exercise and education program improves wellbeing of new mothers: a randomized controlled trial. Phys Ther. 2010;90:348 –355.] © 2010 American Physical Therapy Association

Objective. The purpose of this study was to evaluate the effect of a physical therapy exercise and health care education program on the psychological well-being of new mothers.

Design. This was a randomized controlled trial. Participants. Primiparous and multiparous English-speaking women ready for discharge from The Angliss Hospital postnatal ward were eligible for this study. Women who were receiving psychiatric care were excluded. One hundred sixty-one women were randomized into the trial. Intervention. The experimental group (n⫽62) received an 8-week “Mother and Baby” (M&B) program, including specialized exercise provided by a women’s health physical therapist combined with parenting education. The other group (education only [EO], n⫽73) received only the same educational material as the experimental group. Main Outcome Measures. Psychological well-being (Positive Affect Balance Scale), depressive symptoms (Edinburgh Postnatal Depression Scale), and physical activity levels were assessed at baseline, after 8 weeks (post-program), and then 4 weeks later. Results. There was significant improvement in well-being scores and depressive symptoms of the M&B group compared with the EO group over the study period. More specifically, there was a significant positive effect on well-being scores and depressive symptoms at 8 weeks, and this effect was maintained 4 weeks after completion of the program. The number of women identified as “at risk” for postnatal depression pre-intervention was reduced by 50% by the end of the intervention.

Limitations. Although this study provides promising short-term (4-week) outcomes, further work is needed to explore whether the intervention effects are maintained as sustained psychological and behavioral benefits at 6 months. Conclusions. A physical therapy exercise and health education program is effective in improving postnatal well-being. Routine use of this program may reduce longer-term problems such as postnatal depression.

Post a Rapid Response or find The Bottom Line: www.ptjournal.org 348

f

Physical Therapy

Volume 90

Number 3

March 2010

Improving Postnatal Well-Being

G

iving birth involves many changes in a woman’s physical, emotional, and social health. Increased levels of sex hormones cause laxity of ligaments and muscles to prepare a woman’s body for the growing baby and birth process. This laxity makes certain movements and physical tasks difficult and uncomfortable. Major changes of both hormonal and social origin occur with pregnancy and continue into the postnatal period. Decisions regarding career and when to finish work, as well as factors such as relationships, sleep deprivation, and the availability of support from family and friends can affect a woman’s selfesteem and perception of herself as a new mother and influence her parenting ability.1

Postnatal depression (PND) is a major health issue affecting up to 13% of all new mothers throughout the world, with most cases commencing in the first 3 months of the postnatal period.2 Postnatal depression is thought to evolve from neuroendocrine changes, such as pregnancy stress, and personality predisposition, as well as a combination of many other factors.3 Yet its duration is thought to be determined more by sociocultural factors such as selfesteem of the mother, the childbirth experience, and the availability of support and local services.1 It has been suggested that a multidisciplinary treatment approach should be taken to manage and reduce the risk for PND.1 Group physical therapy exercise programs may assist in the management of this condition through a mixture of social support and the effects of exercise. The literature also suggests that future trials must determine which approaches will be most successful, costeffective, and feasible to implement in the wider community.4 However, to date there have been no randomized controlled trials (RCTs) evaluatMarch 2010

ing the benefits of group physical therapy exercise approaches to improve psychological health outcomes of women postnatally. The Mother and Baby (M&B) Program was developed in the Physiotherapy Department of The Angliss Hospital and has been delivered regularly for the past 7 years. The program was developed in response to a Women’s Health Strategy developed by the Victorian Government’s Department of Human Services over the period 2001 to 2004, which identified a large number of isolated new mothers in Melbourne’s Outer East. It comprises an 8-week program of 1 hour of group physical therapy exercise with mothers and their babies and a 30-minute education session delivered by health care professionals. In this article, we report on an RCT evaluating the benefits of the M&B Program on the psychological health outcomes of postnatal women. We hypothesized that women participating in the M&B Program would have higher well-being scores and lower risk for postnatal depression compared with women who were not participating in the program. Such a program is feasible to deliver and, if shown to be effective in reducing the risk for PND, it would make a significant contribution to the health of new mothers.

Setting and Participants Recruitment for the study was conducted between June 2005 and June 2006. All primiparous and multiparous women ready for discharge from The Angliss Hospital postnatal ward and who spoke and read English independently were invited to participate, regardless of the type of delivery. Women were excluded if they had a diagnosis of a psychiatric disorder medicated and managed by a general practitioner or a psychiatrist or if they needed hospitalization. The CONSORT flowchart (Figure) shows the details of the study. Randomization and Interventions Women were assigned randomly to either the M&B group or the EO group using a computer-generated random numbers list. Allocation was managed by one of the researchers (M.P.G.) and stratified according to parity (primiparous or multiparous) in blocks of 16 participants. Group allocation was concealed in consecutively numbered, sealed, opaque envelopes that were opened by the physical therapist conducting the M&B Program. Due to the nature of the intervention, blinding of participants was not possible. The EO group received written educational material mailed to them every week over 8 weeks. Education topics covered baby massage, nutrition for mothers, introducing solids,

Method Ethics approval for this study was obtained from the institutional review boards at The Angliss Hospital and the University of Melbourne, and all participants gave informed consent. Design Overview This study was an RCT investigating the effect of the M&B Program, commencing at 6 to 10 weeks postpartum, on the psychological well-being of new mothers.

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 January 7, 2010, at ptjournal.apta.org.

Volume 90

Number 3

Physical Therapy f

349

Improving Postnatal Well-Being exercise sessions was adapted for each woman depending on the type of delivery and her recovery. Participants also had a 30-minute education session delivered by health care professionals, including physical therapists, dietitians, speech pathologists, health psychologists, and midwives. In addition, the M&B group received the same written educational material as the EO group. In the last week of the program, all the speakers and the women and their babies gathered together for afternoon tea. Both groups received a booklet containing diagrams of all the exercises provided over course of the program, as well as a list of local gyms and community resources to assist them in continuing their exercise at home. Outcome Measures and Follow-up All participants completed questionnaires at 6 to 10 weeks postpartum (baseline), at the end of the intervention (8 weeks from baseline), and then 4 weeks following the intervention (12 weeks from baseline). The questionnaire booklet contained a well-being scale (the Positive Affect Balance Scale [PABS]), a depression scale (Edinburgh Postnatal Depression Scale [EPNDS]), and questions regarding the amount of physical activity (minutes per week) undertaken.

Figure. CONSORT flowchart. M&B⫽“Mother and Baby” Program, EO⫽education only.

adjusting to a new lifestyle, communicating with the baby, sun care for the baby, and play development. Contact details of health care personnel also were included in this written information.

350

f

Physical Therapy

Volume 90

The M&B Program was conducted once per week for 8 weeks at The Angliss Hospital. Each week, women undertook 1 hour of group exercise with their babies, facilitated by a physical therapist, which involved cardiovascular and strength components. Each of the 8

Number 3

The primary outcome measure was the PABS. This 10-question scale indicates psychological reactions (positive and negative) of people in the general population to events in their daily lives.5 The scale is described as an indicator of an individual’s ability to cope with the stresses of everyday living, and the “positive affect” questions in particular have demonstrated a relationship with social participation, satisfaction with social life, and engagement in activities. Responses to the questions, which are March 2010

Improving Postnatal Well-Being made according to a 3-point scale of “never,” “sometimes,” or “often,” reflect experiences in the previous week. A positive affect score, a negative affect score, and a total affect balance score can be calculated. This study used the positive affect score (ranging from 5 to 15), with 15 being the maximum (most positive) score. The components of this total positive score include feeling “excited or interested,” “proud,” “pleased to accomplish,” “on top of the world,” and “things were going your way.” This scale is a reliable measure of psychological well-being.6 The secondary outcome measures were the EPNDS7 and amount of physical activity. The EPNDS is specifically used for the screening of postnatal depression risk factors.7 It consists of 10 questions, with responses on a 4-point scale. Women are asked to circle the response that most closely represents how they have felt over the previous 7 days. A score of 13 or more generally is used internationally as a cutoff for depressive symptoms and the need for further investigation regarding PND.1,7 This scale is not a diagnostic tool; rather, it identifies risk factors for postnatal depression. It was selected for the study because it is the only outcome measure available to screen women in the postnatal period. It was not used as a primary outcome measure because the main purpose of the study was to measure women’s psychological well-being in the postnatal period, not postnatal depression. However, improving women’s psychological well-being, in turn, might reduce the risk for postnatal depression (ie, lead to lower scores on the EPNDS). Three questions were asked of the women regarding type, duration, and frequency of physical activity they undertook at baseline, 8 weeks, and 12 weeks of the study period. These questions were based on the March 2010

American College of Sports Medicine and American Heart Foundation’s most recent exercise guidelines, which recommend that the average adult who is healthy should participate in 30 minutes of moderateintensity aerobic exercise 5 times weekly and 8 to 10 strength training exercises with 8 to 12 repetitions of each exercise twice weekly.8 The total duration of formal physical activity per week was determined (minutes per week) and compared between the 2 groups across the 3 time periods. The questionnaire booklet was identified by a code number for each participant to conceal group allocation from the researcher (E.N.) who undertook the scoring. Data Analysis A sample size calculation based on a pilot study of the M&B Program indicated that a total of 134 participants would be sufficient to detect a clinically important difference of 1.3 units on the PABS with 80% power and an alpha level of .05. All data were analyzed using SPSS software, version 11.5.* Data were examined for deviation from normal distribution. Analysis was by intention-totreat. The PABS and EPNDS scores and amount of exercise for both groups were compared over the 3 time periods using a mixed-model analysis of covariance, with group as a between-subjects factor, time as a within-subjects factor and baseline score, and parity and age as covariates. Linear regression was used to examine the factors predicting PABS and EPNDS scores. Nonparametric tests were used for other analyses. Imputation of missing data at follow-up was by last observation carried forward.

* SPSS Inc, 233 S Wacker Dr, Chicago, IL 60606.

Role of the Funding Source The Angliss Hospital and the University of Melbourne provided financial support for the provision of the M&B Program and statistical advice.

Results Recruitment for the study was conducted between June 2005 and June 2006. One hundred sixty-one women were randomly assigned to either the M&B group or the EO group (Figure). The demographics of the cohort are shown in Table 1. Eighteen women assigned to the M&B group and 8 women assigned to the EO group did not receive the allocated intervention for the reasons outlined in Table 2. These participants were not included in the analysis. The remaining participants who commenced the intervention (M&B group, n⫽62; EO group, n⫽73) were all included in the analysis, including those who dropped out at 8 weeks (n⫽3) and 12 weeks (n⫽2) (Figure). There were no adverse events. The overall dropout rate for those commencing the study was 3.7%. There was a significant betweengroup difference in PABS scores over time. Post-hoc t tests showed a significant difference between baseline and 8 weeks, but not between 8 and 12 weeks, indicating that PABS scores remained relatively stable for both groups over this time period (Tab. 3). Comparison of the EPNDS scores over the time periods revealed a significant main effect, with the women in the M&B group showing a reduced risk for PND over time. There was no significant change in EPNDS scores between 8 and 12 weeks (Tab. 3). At baseline, the proportion of women who were at risk for depression, as indicated by a score of 13 or more on the EPNDS, was 22% in the

Volume 90

Number 3

Physical Therapy f

351

Improving Postnatal Well-Being Table 1. Demographics of “Mother and Baby” (M&B) Program Group (n⫽62) and Education Only (EO) Group (n⫽73) at Baselinea M&B Group Characteristic

EO Group

X

SD

Range

X

SD

Range

P

Mother’s age (y)

29.3

4.0

20–41

30.1

5.3

17–39

.32

Weight (kg)

73.4

14.8

50–125

71.0

13.2

52–110

.34

Baby’s age (wk)

7.3

1.3

6–10

8.0

1.5

6–10

.35

Age of oldest child (y)

1.2

2.7

0–15

1.7

3.1

0–14

.36

n

%

n

%

Primiparous

42

68

46

63

.5

Cesarian birth

34

55

27

37

.02a

Vaginal birth

28

45

46

63

.02a

14

22

19

26

.34

0

0

2

3

.42

White collar

24

39

21

29

.35

Blue collar

24

39

31

42

.36

1

2

1

1

.35

Occupation Homemaker Student

Employment Full-time

a

Part-time–casual

4

6

6

8

.34

Maternity leave

36

58

34

47

.43

Unemployed

21

34

32

44

.02a

Significant at P⬍.05.

M&B group and 16% of the EO group. After the 8-week program, the proportion of women in the M&B group with an increased risk for depression was reduced by 50% to 11%, whereas in the percentage in the EO group remained unchanged. There was no significant difference in the amount of formal exercise undertaken by the 2 groups at baseline, and this did not change over time

(Tab. 3). For all outcome measures, there were no significant effects of age or parity. The PABS scores at baseline, the amount of exercise performed at 8 and 12 weeks, and group membership all independently contributed to the prediction of the PABS scores at 8 weeks and 12 weeks. A greater amount of exercise at 8 weeks and 12 weeks, higher baseline PABS scores, and membership in the exer-

Table 2. Reasons for Not Commencing the Allocated Interventiona Never Commencing Intervention

a

M&B Group (nⴝ18)

Baby unwell

n⫽2

n⫽1

Unable to be contacted

n⫽5

n⫽7

Moving or going overseas

n⫽5

Too busy with other children

n⫽6

M&B⫽“Mother and Baby” Program, EO⫽education only.

352

f

EO Group (nⴝ8)

Physical Therapy

Volume 90

Number 3

cise group were all significantly associated with higher PABS scores, and accounted for 47% (46% adjusted) and 54% (53% adjusted) of the variance in PABS scores. The EPNDS scores at baseline and group membership, but not the amount of exercise performed at 8 weeks or 12 weeks, independently contributed to the prediction of the EPNDS scores at 8 and 12 weeks. Higher baseline EPNDS scores and membership of the M&B group were significantly associated with greater EPNDS scores, and accounted for 55% (54% adjusted) and 52% (50% adjusted) of the variance in EPNDS scores.

Discussion This study is the first RCT to investigate the effects of group physical therapy exercise and health care education on new mothers’ well-being March 2010

Improving Postnatal Well-Being Table 3. Descriptive Statistics for “Mother and Baby” Program (M&B) and Education Only (EO) Groups Across Time for All Outcome Measures (Positive Affect Balance Scale, Edinburgh Postnatal Depression Scale, and Physical Activity) M&B Group (nⴝ62)

EO Group (nⴝ73)

X (SD), 95% CIa

X (SD), 95% CI

Baseline

10.72 (2.19), 10.10–11.24

10.67 (2.17), 10.16–11.21

8 weeks

11.82 (2.08), 11.24–12.37

10.47 (2.26), 9.96–11.01

.007b,*

12 weeks

11.93 (2.31), 11.38–12.45

10.49 (1.91), 10.01–11.01

.580b,#

Baseline

8.00 (6.16), 6.54–9.57

6.75 (5.44), 5.32–8.08

8 weeks

5.47 (5.11), 4.19–6.92

6.75 (5.51), 5.42–7.95

⬍.0001c,*

12 weeks

4.73 (5.27), 3.49–6.23

6.54 (5.61), 5.03–7.59

.194c,#

165 (168), 121–209

141 (148), 106–176

Time

P

Positive Affect Balance Scale

Edinburgh Postnatal Depression Scale

Physical activity (minutes per week) Baseline 8 weeks

188 (126), 156–221

154 (153), 118–191

12 weeks

176 (110), 147–205

155 (173), 114–196

.87d

a

CI⫽confidence interval. Analysis of covariance (ANCOVA) adjusted for baseline score, age, and parity: significant main effect (F1,134⫽18.065, P⫽.001); *significant between-group effect (post hoc t test: baseline vs 8 weeks); #nonsignificant between-group effect (post hoc t test: 8 weeks vs 12 weeks). c ANCOVA adjusted for baseline score, age, and parity: significant main effect (F1,134⫽12.688, P⬍.0001); *significant between-group effect (post hoc t test: baseline vs 8 weeks); #nonsignificant between-group effect (post hoc t test: 8 weeks vs 12 weeks). d ANCOVA adjusted for baseline score, age, and parity: nonsignificant main effect (F1,134⫽0.14, P⫽.87). b

and their risk for PND. Previous studies have shown that general exercise improves mood states in younger9 and older women,10 improves wellbeing,11 and leads to a reduction in depressive symptoms in mothers diagnosed with PND.12 No clinical trials to date have been conducted postnatally in women who are healthy, nor has the effect of a physical therapy exercise program been investigated. This RCT has shown that the M&B Program, comprising education plus exercise, resulted in a greater improvement in mothers’ well-being than an education-only intervention. A significant reduction in maternal symptoms of depression and, therefore, a decreased number of women at risk for PND also were observed. The proportion of women who were at risk for depression, as measured by the EPNDS (22% of the M&B group; 16% of the EO group) is similar to the prevalence rate of PND internationally (10%–20%).3 Neither March 2010

the mothers’ age nor the number of children affected the findings. Surprisingly, there was no significant difference in the amount of exercise performed by the 2 groups. Importantly, the analyses suggest that a higher level of physical activity at 8 weeks was associated with increased psychological well-being, irrespective of the group to which the women were allocated. However, being part of the intervention group was predictive of increased well-being and reduced depression scores, regardless of the amount of physical activity. Sampselle et al13 reviewed the benefits of physical activity on the psychological well-being of women postnatally. Although they did not conduct a clinical trial, their interviews suggested that women who were more active reported more satisfaction with motherhood and with their partners. These observations are supported by reports that a single session of exercise can result in

both an increase in positive mood states (eg, feeling positive, more energetic, happy and more refreshed) and a decrease in negative mood states (eg, tension, anxiety, confusion).11 These results are consistent with the findings of our study. In the general population, aerobic exercises (eg, walking, running, cycling, swimming) have a stronger effect on mood than anaerobic activities (eg, weight training).11,14 Specific guidelines on how exercise can affect well-being in different populations, including in women postnatally, have been lacking. In our study, the physical therapy exercise component of the M&B Program included both modified aerobic and anaerobic components every week during the program. The women in the M&B Program reported that the physical activity they undertook separately from the intervention was aerobic in nature, such as walking, cycling, and swimming.

Volume 90

Number 3

Physical Therapy f

353

Improving Postnatal Well-Being Offering all women specific physical therapy exercises and advice has been recommended to assist them with recovery from giving birth.15 Although all women participating in the study were taught pelvic-floor exercises as part of routine postnatal care, the M&B Program provided an opportunity for participants to have individualized instruction on pelvicfloor exercises if they requested it. The program also included information on good bladder and bowel habits, assessment of and appropriate exercise for abdominal muscles, and upper- and lower-body strengthening exercises to assist mothers with picking up other children and performing household tasks. All of these components may contribute to the effectiveness of, and adherence to, such a program.15 The M&B Program is consistent with recommendations given to general practitioners to prescribe exercise to patients with depression. Nicoloff and Schwenk16 suggested that supervision, positive reinforcement, and feedback were important, especially early in the exercise program. The excellent program adherence (more than 85%) and the very low dropout rate from the M&B Program demonstrate that it was feasible for the mothers to attend. Moreover, many participants wrote thank-you notes indicating that they enjoyed the sessions and were still continuing to meet with other participants in their group on a social basis. Although parity was not a factor in the effectiveness of the program, the provision of child care would make it easier for women with more than one child to attend. The physical therapist emphasized the importance of exercise every week and the mothers had the opportunity to develop social networks with other new mothers while exercising. Although these factors are most likely to be important elements of this effective intervention, it was not possible to 354

f

Physical Therapy

Volume 90

determine which component of the M&B Program had an effect on wellbeing. Other studies involving exercise such as pram (stroller) walking have found the social interaction in such groups to be effective for new mothers.17–19 However, pram walking exercise programs have been shown to improve fitness levels and reduce depression symptoms in women with PND significantly more than social support alone.12 In our study, the proportion of women in the M&B group with an increased risk for depression (22%) was reduced by 50% after 8 weeks to 11%. The size of this effect is similar to that found by Hiscock and Wake,20 who used a behavioral infant sleep intervention to improve infant sleep and maternal mood. In this study, the M&B group had 18% more cesarian deliveries than the EO group yet still showed a significantly greater increase in wellbeing. Despite earlier reports that cesarian delivery was associated with increased risk of postnatal depression, a recent study by Patel et al21 showed that this was not the case. Reducing the risk of depressive symptoms in new mothers is important, considering that only approximately 25% of PND cases are reported and 63% may be treated as late as 1 year postpartum.3 All of the mothers in this study had had their babies 6 to 10 weeks prior to participating in the study, which is the most common time for symptoms of mild PND to occur.3 Thus, the program was conducted at the most relevant period for these women, when they were most vulnerable to developing PND. The M&B intervention provided by health care professionals offers a potentially cost-effective way of improving the well-being of new mothers and of assisting those with depressive symptoms by providing

Number 3

an opportunity for these mothers to develop social networks. This intervention could be implemented immediately by appropriate professionals in an outpatient setting or community health center at very little cost. The M&B Program could help to manage the stigma associated with PND and perhaps prevent PND in those women who may be at risk.22 Armstrong and Edwards12 found that participants with PND symptoms in their pram walking programs preferred to be with women who were going through similar difficulties, as the group understood their situation. Future studies measuring depressive symptoms at baseline and establishing the effectiveness of exercise interventions similar to the M&B Program on women with depressive symptoms and over a longer period of time would complement these findings.

Conclusions This rigorous RCT has shown that in a cohort of postnatal women attending a maternity service, the combination of exercise, supervised by a physical therapist, and face-to-face health care education can improve the well-being of new mothers and reduce the risk for PND. The findings of this study can be generalized to similar groups of women who are healthy and attending a typical maternity service. Ms Norman, Dr Osborne, and Dr Galea provided concept/idea/research design, writing, and data analysis. Ms Norman provided data collection, participants, and facilities/ equipment. Ms Norman and Dr Galea provided project management. Dr Galea provided fund procurement and clerical support. Ms Norman, Dr Sherburn, and Dr Galea provided institutional liaisons. All authors provided consultation (including review of manuscript before submission). Ethics approval for this study was obtained from the institutional review boards at The

March 2010

Improving Postnatal Well-Being Angliss Hospital and the University of Melbourne. This work was presented at the National Conference of the Australian Physiotherapy Association; October 4 –11, 2007; Cairns, Queensland, Australia. This study was supported by funding from The Angliss Hospital and the Rehabilitation Sciences Research Centre, University of Melbourne. Trial registry: National Institutes of Health. Trial register number: NCT00361478. This article was received April 29, 2009, and was accepted September 21, 2009. DOI: 10.2522/ptj.20090139

References 1 Milgrom J, Martin PR, Negri LM. Treating Postnatal Depression: A Psychological Approach for Treating Health Care Practitioners. London, United Kingdom: John Wiley & Sons Ltd; 1999. 2 Dennis C. Psychosocial and psychological interventions for prevention of postnatal depression: a systematic review. BMJ. 2005; 331:1– 8. 3 Rice MJ, Records K, Williams M. Postpartum depression: Identification, treatment, and prevention in primary care. The Clinical Letter for Nurse Practitioners. 2001; 5:1– 4.

March 2010

4 Battle CL, Zlotnick C. Prevention of postpartum depression. Psychiatr Ann. 2005; 35:590 –598. 5 Bradburn NM. The Structure of Psychological Well-Being. Chicago, IL: Aldine; 1969. 6 Moriwaki SY. The affect balance scale: a validity study with aged samples. J Gerontol. 1974;29:73–78. 7 Cox JL, Holden JM, Sagovsky R. Detection of postnatal depression: development of the 10-item Edinburgh postnatal depression scale. Br J Psychiatry. 1987;150: 782–786. 8 Haskell W, Lee I, Pate R, 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. 9 Annesi JJ. Mood states of formerly sedentary younger and older women at weeks 1 and 10 of a moderate exercise program. Psychol Rep. 2004;94(3 pt 2):1337–1342. 10 Lee C, Russell A. Effects of physical activity on emotional well-being among older Australian women: cross-sectional and longitudinal analyses. J Psychosom Res. 2003; 54:155–160. 11 Berger BG, Motl RW. Exercise and mood: a selective review and synthesis of research employing the profile of mood states. J App Sport Psychol. 2000;12:69 –92. 12 Armstrong K, Edwards H. The effects of exercise and social support on mothers reporting depressive symptoms: a pilot randomized controlled trial. Int J Mental Health Nurs. 2003;12:130 –138. 13 Sampselle CM, Seng J, Yeo S. Physical activity and postpartum well-being. J Obstet Gynecol Neonatal Nurs. 1999;28:41– 49.

14 Berger BG. Psychological benefits of an active lifestyle: what we know and what we need to know. Am J Kinesiol Phys Ed. 1996;48:330 –353. 15 Sapsford R, Bullock-Saxton J, Markwell S. Women’s Health: A Textbook for Physiotherapists. London, United Kingdom: WB Saunders Company Ltd; 1998. 16 Nicoloff G, Schwenk MD. Using exercise to ward off depression. Phys Sports Med. 1995;23:44 –58. 17 Harrison MJ, Neutfield A. Women in transition: access and barrier to social support. J Adv Nurs. 1995;21:858 – 864. 18 Sefi S. Running a post-natal exercise group. Health Visit. 1987;60:197. 19 Sefi S, Macfarlene A. Child health clinics: why mothers attend. Health Visitor. 1985; 58:129 –130. 20 Hiscock H, Wake M. Randomised controlled trial of behavioural infant sleep intervention to improve infant sleep and maternal mood. BMJ. 2002;324:1– 6. 21 Patel RR, Murphy DJ, Peters TJ. Operative delivery and postnatal depression: a cohort study. BMJ. 2005;330:879. 22 Currie J, Boxer E, Devlin E. Pramwalking as postnatal exercise and support: an evaluation of the Stroll Your Way To Wellbeing program and supporting resources in terms of individual participation rates and community group formation. Austr J Midwifery. 2001;14:21–25.

Volume 90

Number 3

Physical Therapy f

355

Research Report Progressive Resistance Training Improves Overall Physical Activity Levels in Patients With Early Osteoarthritis of the Knee: A Randomized Controlled Trial Joshua N. Farr, Scott B. Going, Patrick E. McKnight, Shelley Kasle, Ellen C. Cussler, Michelle Cornett J.N. Farr, MS, is Research Assistant, Department of Physiological Sciences, University of Arizona, 1713 E University Blvd #93, Tucson, AZ 85721 (USA). Address all correspondence to Mr Farr at: jfarr@ email.arizona.edu. S.B. Going, PhD, is Professor, Department of Nutritional Sciences, University of Arizona. P.E. McKnight, PhD, is Assistant Professor of Psychology, Department of Psychology, George Mason University, Fairfax, Virginia. S. Kasle, PhD, is Research Assistant Professor, Arizona Arthritis Center, College of Medicine, University of Arizona. E.C. Cussler, MS, is Data Manager, Department of Physiological Sciences, University of Arizona. M. Cornett, RN, is Senior Research Nurse, Arizona Arthritis Center, College of Medicine, University of Arizona. [Farr JN, Going SB, McKnight PE, et al. Progressive resistance training improves overall physical activity levels in patients with early osteoarthritis of the knee: a randomized controlled trial. Phys Ther. 2010;90:356 –366.] © 2010 American Physical Therapy Association

Background. Prescription of resistance training (RT) exercises is an essential aspect of management for knee osteoarthritis (OA). However, whether patients with knee OA who are randomly assigned to receive RT simply substitute RT for other modes of physical activity remains unclear. Objective. The aim of this study was to determine the effect of a structured RT intervention on overall levels of moderate- and vigorous-intensity physical activity (MVPA) in patients with early-onset knee OA. The study compared patients with early-onset OA who participated in an RT program, those who participated in a self-management (SM) program, and those who participated in both RT and SM. Because participants randomly assigned to receive the RT intervention may simply switch activity modes, resulting in little net effect, we assessed total MVPA in addition to tracking changes in strength (force-generating capacity).

Design and Intervention. This study was a randomized controlled trial comparing the effectiveness of SM alone, RT alone, and combined RT⫹SM on MVPA in patients with early OA of the knee.

Setting. The study was conducted on a university campus, with patient recruitment from the local community.

Participants. The participants in this study were 171 patients (74% women, 26% men) with knee OA. They had a mean age of 55.1 (SD⫽7.1) years, a mean body mass index of 27.6 (SD⫽4.2) kg/m2, and radiographic status of grade II OA (and no higher) in at least one knee, as defined by the Kellgren and Lawrence classification. They wore an accelerometer while awake (X⫽14.2 [SD⫽2.2] hours) for 5 to 7 contiguous days (X⫽6.8 [SD⫽0.5] days) at baseline and at 3 and 9 months of intervention. Results. The participants engaged in MVPA a mean of 26.2 (SD⫽19.3) minutes per day at baseline. Both groups significantly increased their MVPA from baseline to 3 months (RT group by 18% [effect size (d)⫽0.26]; SM group by 22% [effect size (d)⫽0.25]), but only the RT group sustained those changes at 9 months (RT group maintained a 10% increase [effect size (d)⫽0.15]; SM group maintained a 2% increase [effect size (d)⫽0.03]). A significant group ⫻ time interaction for MVPA indicated that the RT group maintained higher MVPA levels than the SM group. Limitations. Lack of direct measures of energy expenditure and physical function was a limitation of the study.

Post a Rapid Response or find The Bottom Line: www.ptjournal.org 356

f

Physical Therapy

Conclusions. Patients with early-onset OA of the knee can engage in an RT program without sacrificing their overall MVPA levels. These results support the value of RT for management of knee OA.

Volume 90

Number 3

March 2010

Effects of Resistance Training on Physical Activity

T

he primary goals of knee osteoarthritis (OA) treatment are to reduce pain and improve function and quality of life. Declining enthusiasm for cyclo-oxygenase 2 inhibitors for knee OA pain relief and unsuccessful clinical trials of diseasemodifying OA drugs have contributed to increased interest in nonpharmacologic treatments for OA.1 Resistance training (RT) exercise programs and educational selfmanagement (SM) programs are 2 mainstays of nonpharmacologic treatment. Physical activity (PA) refers to any bodily movement that results in energy expenditure. Physical activity is an essential recommendation included in all guidelines for management of knee OA.2– 4 Moreover, PA is recommended by the US Centers for Disease Control and Prevention (CDC) and the American College of Sports Medicine (ACSM) for general health to reduce risks of obesitylinked health problems, including diabetes and cardiovascular disease,5,6 which often coexist with knee OA. Work group recommendations from the 2002 Exercise and Physical Activity Conference (EPAC)7 advise patients with knee OA to accumulate 30 minutes of at least moderateintensity (ⱖ3 metabolic equivalents [METs]*) PA on at least 3 days of the week. The expert EPAC panel concluded that promotion of PA in adults with arthritis should emphasize aerobic moderate- and vigorousintensity physical activity (MVPA, ⱖ3 METs) and muscle strengthening resistance exercise. In a more recent statement, an expert consensus panel provided evidence-based recommendations for practical delivery of exercise therapy for patients with knee OA, stating that “both general (aerobic fitness training) and local (strengthening) exercises are essen-

* 1 MET⫽3.5 mL O2䡠kg⫺1䡠min⫺1.

March 2010

tial, core aspects of management for every patient with knee OA.”8(p69) In recent years, it has become clear that RT can have a positive effect on resting energy expenditure (REE), total free-living energy expenditure (TEE), and activity-related energy expenditure (AEE). Withers et al9 compared REE, TEE, and AEE of chronically active women who engaged in RT and chronically inactive women, aged 49 to 70 years. They reported that the chronically active women had increased REE, TEE, and AEE compared with the chronically inactive women. Hunter et al10 addressed this concern in elderly men and women who were healthy, aged 61 to 77 years. They found increases in REE, TEE, and AEE in response to 26 weeks of RT and showed that the TEE increase remained significant even after adjustment for the energy expenditure of the RT. These findings suggest that RT has value in increasing energy expenditure and lipid oxidation rates in older adults. A potential concern when structured RT programs are prescribed is that participants may simply switch activity modes, resulting in a decrease in aerobic MVPA. For example, Goran and Poehlman11 and Meijer et al12 both observed a compensatory decrease in free-living PA levels of older adults after engaging in RT programs. However, we found no studies that have addressed this concern in a patient population such as patients with early OA of the knee. Although controlling mode is desirable for study purposes, in clinical and public health settings, replacement of one mode with another may defeat efforts to increase overall MVPA. In contexts such as the present study, participants randomly assigned to receive RT might engage in less overall MVPA, substituting RT for other modes of MVPA. Alternatively, if participants randomly assigned to receive RT increased or at

least maintained their MVPA levels, they would benefit from both RT and aerobic MVPA. However, if RT inhibited participants from achieving recommended MVPA levels, the net result could interfere with exercise interventions aimed at improving cardiovascular function, insulin action, energy metabolism, and psychological health in patients with OA of the knee.13,14 Therefore, in the present analysis, we aimed to determine the effect of a structured RT intervention on overall daily levels of activity by using accelerometry to measure MVPA in individuals with early-onset knee OA who participated in an RT program and in those who participated in an SM program. We hypothesized that in addition to improving muscle strength (forcegenerating capacity), the RT groups would maintain similar levels of MVPA compared with the SM group.

Method Design Overview The data used for this analysis were obtained from the Multidimensional Intervention for Early Osteoarthritis of the Knee Study (the Knee Study), a randomized clinical trial comparing the effectiveness of SM alone, RT alone, and combined RT⫹SM on relevant knee OA outcomes over 24 months. After preliminary analyses of the 3 groups, the RT groups (RT alone and RT⫹SM) were collapsed into a single group and compared with the SM only group to test the

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 January 7, 2010, at ptjournal.apta.org.

Volume 90

Number 3

Physical Therapy f

357

Effects of Resistance Training on Physical Activity Table 1. Baseline Descriptive Characteristics of Participants by Intervention Groupa

Characteristic

SM Group (nⴝ57)

RT Group (nⴝ52)

RTⴙSM Group (nⴝ62)

Collapsed RT and RTⴙSM Group (nⴝ114)

Noncompleters (nⴝ83)

72

73

79

76

83

Female (%) Age (y), X (SD) Height (cm), X (SD)

55.8 (6.1)

55.5 (7.3)

54.2 (7.3)

54.7 (7.3)

55.5 (7.7)

169.5 (8.1)

169.6 (10.6)

167.2 (9.4)

168.3 (10.0)

166.7 (10.1)

Weight (kg), X (SD)

80.6 (13.7)

80.1 (19.4)

76.1 (13.4)

77.8 (16.6)

78.2 (15.2)

BMI (kg/m2), X (SD)

28.0 (4.0)

27.5 (4.5)

27.2 (4.2)

27.3 (4.3)

28.1 (4.3)

Normal (⬍25)

30%

35%

32%

33%

33%

Overweight (25–30)

40%

36%

40%

39%

39%

Obese (⬎30)

30%

29%

28%

28%

28%

Grade 11 unilateral knee OA

59%

46%

56%

52%

55%

Grade II bilateral knee OA

41%

54%

44%

48%

45%

Knee OA severity

a

SM⫽self-management, RT⫽resistance training, RT⫹SM⫽resistance training ⫹ self-management, BMI⫽body mass index, OA⫽osteoarthritis.

question of whether participants randomly assigned to the RT group would substitute RT for MVPA. The analysis presented here was based on data from the first phase (baseline to 9 months) of the interventions described. Outcome measurements were obtained at baseline and at 3 and 9 months of intervention. Participants Participants were recruited from the Tucson, Arizona, general community and surrounding areas using mass mailings, media advertisements, periodic media coverage, and requests to local physicians for patient referrals. A total of 1,726 people were assessed for eligibility. Eligibility criteria were: age between 35 and 68 years to ensure an early-onset knee OA sample; pain on 4 or more days of the week in one or both knees for at least 4 months during the previous year; less than 5 years’ symptom duration15,16; radiographic status of grade II OA (and no higher) in at least one knee, as defined by the Kellgren and Lawrence classification17; and disability due to knee OA, as assessed with the Western Ontario and McMaster Universities Osteoarthritis (WOMAC) Index.18 Partici358

f

Physical Therapy

Volume 90

pants gave written informed consent and self-reported demographic characteristics (Tab. 1). All participants enrolled in the study met American College of Rheumatology classification criteria for early OA of the knee.19 A CONSORT flowchart describing the progress of participants through the 9-month intervention is presented in Figure 1. Of the 1,726 people who were assessed for eligibility, 293 eligible participants were stratified by sex and randomly assigned via a random number table to 1 of the 3 treatment groups (SM, RT, or RT⫹SM). Concealed allocation was accomplished using envelopes to conceal computer-generated values. Manifest transparency of the treatments rendered blinding unfeasible. Essentially, given our outcome measures, interrater agreement bias was deemed a smaller threat to internal validity than ensuring treatment fidelity, which precluded effective blinding. Of the 293 eligible participants, 39 did not receive any of the allocated intervention and 33 discontinued the intervention prior to 9 months (Fig. 1).

Number 3

Interventions Resistance training. The overall goal of the RT intervention was to encourage participants to maintain a long-term exercise program to increase muscle strength, decrease impairment, maintain and restore function, and protect joint structures from further damage. The RT intervention paralleled programs developed by the ACSM5 and the National Strength Training and Conditioning Association20 and was designed to test expert panel recommendations.8 Sessions targeted improvement in each of 4 core areas: (1) stretching and balance, (2) range of motion (ROM) and flexibility, (3) isotonic muscle strengthening, and (4) aerobics. Participants met with certified physical trainers 3 times per week for 9 months, with a minimum of 1 day of rest between training sessions, to complete a 1-hour exercise regimen that emphasized RT. Supervised, small-group sessions were held to improve adherence. Each session consisted of: (1) 10-minute warm-up on either a bicycle ergometer or treadmill at 50% maximum heart rate, (2) 5 to 10 minutes of stretching and balance exercises, (3) 10 minutes of ROM exercises, (4) 30 March 2010

Effects of Resistance Training on Physical Activity

Figure 1. Flowchart describing the progress of participants through the Knee Study trial. RT⫽resistance training, SM⫽self-management, RT⫹SM⫽resistance training ⫹ self-management.

March 2010

Volume 90

Number 3

Physical Therapy f

359

Effects of Resistance Training on Physical Activity Table 2. Change (⌬) in Weight Lifted From Baseline to 9 Months of Intervention for All Resistance Trainersa Exercise Leg press

RT Group ⌬

RTⴙSM Group ⌬

All (nⴝ114) ⌬

Pb

112.7 (72.1)

91.9 (67.3)

101.6 (69.3)

.19

Incline dumbbell press

11.4 (8.5)

10.7 (8.0)

11.0 (8.2)

.69

Seated row

27.2 (15.8)

20.1 (18.2)

23.3 (17.4)

.08

Leg curl

41.2 (35.5)

33.7 (39.6)

37.3 (37.7)

.40

Calf raise

13.4 (15.1)

10.9 (15.2)

12.2 (15.0)

.60

a

Values are means (SD) for change in load (pounds) from baseline to 9 months of intervention based on 1 to 2 sets of 6 to 8 repetitions. RT⫽resistance training, RT⫹SM⫽resistance training ⫹ self-management. b Independent t test for group difference.

minutes of RT exercises, and (5) 5 minutes of cool-down. Specific RT exercises included leg press, leg curl, hip abduction and adduction, straight leg lift, incline dumbbell press, seated row, and calf raise. The exercises were chosen primarily to directly strengthen the muscles supporting the knee, but also to improve the strength of muscles most involved in activities of daily living. The strength protocol progressed through 2 phases: (1) resistance from body weight and Thera-Band exercise bands† and (2) free weights and machine weights. Participants started with one set per exercise, 6 to 8 repetitions per set, at an intensity of 50% of each individual’s 3-repetition maximum (3-RM). During an orientation session, participants were familiarized with the equipment and instructed by certified trainers on proper lifting techniques for all exercises. All participants began training at a comfortable weight with proper form for each exercise based on standardized protocols developed by the ACSM.5 The program progressed from 1 to 2 sets, along with increases in load when participants were able to complete all repetitions with proper body position and joint alignment for 3 con† The Hygenic Corp, 1245 Home Ave, Akron, OH 44310-2575.

360

f

Physical Therapy

Volume 90

secutive sessions.21 They then progressed to loads between 60% and 75% of their 3-RM and continued to increase loads to maintain vigorous intensity. The ROM exercises were increased for each participant when the exercises could be completed with a Borg scale score of difficulty of ⱕ6.22 Participants completed training logs during all sessions and reported sets, repetitions, and loads for each exercise. Certified physical trainers supervised all RT sessions, monitored progression, and tested participants following standard protocols.5 Throughout the intervention, trainers emphasized good form and encouraged participants to report soreness or pain during and after RT sessions. Changes in load from baseline to 9 months for the RT groups (RT and RT⫹SM) are shown in Table 2. Self-management. The SM intervention was designed to target coping skills, promoting the use of more adaptive strategies and fewer avoidance or passive strategies based on existing self-help programs.23 The intervention also targeted self-efficacy through a variety of educational and behavioral techniques. Self-efficacy skills focused on increasing perceptions of control for physical functioning, pain management, and other ancillary arthritis symptoms. The

Number 3

9-month program began with 12 weekly, 90-minute classroom sessions in which participants completed SM education modules addressing an overview of OA, general exercise principles and PA recommendations, stress management, foot care, pain management, analgesic and anti-inflammatory medications, nutrition for health, coping mechanisms, communication with health care providers, and healthy lifestyle practices. As part the of the exercise module, participants were introduced to the benefits of MVPA and RT for patients with OA of the knee and were given instructions for establishing a regular PA program. They also were provided with PA recommendations implemented by the CDC6 and the ACSM5 and work group PA recommendations from the 2002 EPAC for people with arthritis,7 but no further exercise instruction was given. Classroom sessions were followed by 24 weeks of a structured telephone intervention program that reinforced SM skills. Combined treatment. The combined treatment group (RT⫹SM) engaged in both the RT and SM interventions, with slight alterations to ensure equivalence of contact time across treatment groups. Specifically, participants in the RT⫹SM group were contacted by staff less during the 24 weeks of the telephone intervention program that followed classroom sessions. Anthropometry. Anthropometric measurements were obtained at baseline following standard protocols outlined in the Anthropometric Standardization Reference Manual.24 Total body mass was measured to the nearest 0.1 kg using a calibrated scale (Seca model 770),‡ and height was measured to the nearest 0.1 cm using a portable stadiometer ‡ Seca GMBH and Co KG, Hammer Steindamm 9 25, 20089 Hamburg, Germany.

March 2010

Effects of Resistance Training on Physical Activity (Shorr Height Measuring Board)§ after full inspiration. Pain. Knee pain was assessed using the WOMAC Index, which has been validated in patients with OA of the knee.18 The WOMAC pain subscale comprises 5 items eliciting patient ratings on visual analog scales (0 –100) of pain severity during walking, stair use, lying in bed at night, sitting, and standing. The pain subscale has a maximum summed score of 500, with higher scores reflecting more pain. Physical activity. Baseline PA levels of the Knee Study participants have been published previously.25 Physical activity was measured using the MTI Actigraph accelerometer (model 7164).㛳 The uniaxial Actigraph accelerometer measures vertical-plane accelerations and decelerations and records them as “counts” over a specific time interval (epoch), which provides information regarding the intensity of PA associated with movement.26 The accelerometer assesses accelerations ranging from 0.05 to 2.0g, with a frequency ranging from 0.25 to 2.5 Hz.27 These specifications allow for detection of normal body motion, while filtering out high-frequency vibration movements. Actigraph reliability and validity have been reviewed in detail.28 Counts have been shown to be highly correlated (r⫽.77–.88) with steady-state oxygen consumption during ambulatory activities26,29 and have been shown to be dependent upon movement frequency in a mechanical setup.30 For each assessment, the accelerometer was initialized and downloaded according to the manufacturer’s specifications30 and set to record §

Shorr Productions, 17802 Shotley Bridge Place, Olney, MD 20832. 㛳 Manufacturing Technologies Inc, 70 Ready Ave NW, Fort Walton Beach, FL 32548.

March 2010

data in 60-second epochs. Participants were instructed to wear the accelerometer for 7 contiguous days during all waking hours, except while in water. A previous study31 has shown that when the Actigraph accelerometer is worn for 7 consecutive days, PA can be assessed with 90% reliability. The accelerometer was firmly secured to a belt worn around the waist and positioned on the right hip because this site permits measurement of whole-body movement, does not interfere with daily activities, and is the most frequently used site in epidemiological studies.25 The following measurements were obtained: days worn; registered wear time in hours per day; and average minutes per day spent in moderate-intensity physical activity (MPA, 3– 6 METs), vigorousintensity physical activity (VPA, ⬎6 METs), and MVPA (ⱖ3 METs). A number of studies26,29,32–34 have used criterion methods such as indirect calorimetry and heart rate monitoring to demonstrate the reliability and validity of the Actigraph accelerometer. As described in detail previously,25 we minimized sampling error by averaging the cutoff points reported by calibration studies using the MTI Actigraph model 716426,29,32–34 and applied the resulting cutoff points to differentiate among PA intensities. The applied cutoff points for MPA and VPA intensities were accelerometer recordings of 2,225 to 5,950 and ⬎5,950 counts per minute, respectively. Moderateand vigorous-intensity physical activity was defined as ⱖ2,225 counts per minute. Leisure time PA and exercise habits were assessed at baseline and at 3 and 9 months of intervention using the Aerobics Center Longitudinal Study Physical Activity Questionnaire (ACLS).35 The ACLS elicits selfreports of frequency (sessions per week) and duration (minutes per session) of activities such as walking,

running, treadmill, cycling, swimming, aerobics, yoga, weight lifting, and other sports (eg, golf, tennis, soccer). Data Analysis For the analyses presented here, a valid day of PA was defined as having 10 or more hours of accelerometer wearing, based on previous recommendations from analyses of the National Health and Nutrition Examination Survey (NHANES) accelerometer database.36 Furthermore, 5 to 7 days of valid accelerometer wearing was required for inclusion in the present analysis. We chose a minimum of 5 days of accelerometer wearing because PA levels vary greatly throughout the week and 1 to 4 days of PA may not be representative of habitual PA.28 In order to address the question of whether participants randomly assigned to receive RT would substitute RT for MVPA, the RT groups (RT and RT⫹SM, n⫽114) were collapsed into a single group and compared with the SM group (Tab. 1). Descriptions of the 171 participants randomly assigned to each of the 3 intervention groups (SM, RT, RT⫹SM) and the 114 participants who received the RT intervention (collapsed RT and RT⫹SM group) are shown in Table 1. Means, standard deviations, and 95% confidence intervals were calculated for continuous variables, and frequencies were calculated for categorical variables. Data were checked for missing values and normality prior to analyses. Moderate- and vigorous-intensity physical activity (minutes per day), which was skewed, was natural log transformed for analysis, resulting in a normal distribution. Preliminary tests for baseline differences in descriptive characteristics among the 3 intervention groups were performed using an analysis of variance or the chi-square test for

Volume 90

Number 3

Physical Therapy f

361

Effects of Resistance Training on Physical Activity proportions as appropriate. For all subsequent analyses, the RT groups (RT and RT⫹SM) were collapsed into a single group and compared with the SM group to address the question of whether RT affects MVPA. The RT groups were combined because both groups of participants engaged in the RT intervention and no significant (P⬎.05) group differences using independent t tests were observed for any variables at any time point. Effect sizes representing the magnitude of difference between baseline and 3 and 9 months were calculated using Cohen’s method37 based on adjusted means of MVPA for the 2 groups (SM and collapsed RT). A repeated-measures analysis of covariance was performed with group (SM and collapsed RT) as the between-subjects factor and time (3 and 9 months) as the within-subjects factor. Baseline MPVA, age, BMI, knee OA pain, and sex were entered as covariates to account for baseline between-group differences in MVPA and the putative influence of these variables on MVPA. The specific effect of interest in this analysis was the group ⫻ time interaction. A significant group ⫻ time interaction would indicate that the degree of change in MVPA over time was different for the 2 groups (SM and collapsed RT). Statistical significance was set at P⬍.05 for all tests. Analyses were conducted using the Statistical Package for the Social Sciences, version 17.0.#

Results At baseline, 3 months, and 9 months, 5, 17, and 26 participants, respectively, either did not wear an accelerometer or did not meet the inclusion criteria. Comparisons of baseline descriptive characteristics for noncompleters (n⫽83) versus # SPSS Inc, 233 S Wacker Dr, Chicago, IL 60606.

362

f

Physical Therapy

Volume 90

completers (n⫽171) showed no significant (P⬎.05) differences (Tab. 1). After 9 months of intervention, the numbers of participants who successfully completed all measures at baseline, 3 months, and 9 months were as follows: SM group⫽57, RT group⫽52, RT⫹SM group⫽62 (Fig. 1). The final sample of participants with early onset OA of the knee who successfully adhered to the accelerometer protocol comprised 171 participants (74% women, 26% men; mean age⫽55.1 (SD⫽7.1) years, and mean BMI⫽27.6 (SD⫽4.2) kg/m2). Preliminary baseline analyses of the 3 intervention groups showed no significant (P⬎.05) difference among groups or between completers (n⫽171) and noncompleters (n⫽83) at baseline for any variables (Tabs. 1 and 3). Participants wore the accelerometer, on average, 6.8 (SD⫽0.5) days and 14.2 (SD⫽2.2) hours per day over all 3 PA assessments. There were no significant group differences (P⬎.05) in number of days the accelerometer was worn or in accelerometer wearing time at any time interval. Significant differences were not observed for any of the measured variables (P⬎.05) when participants with 5 to 6 days of accelerometer data were compared with participants with 7 days of data. Unadjusted means, standard deviations, and 95% confidence intervals for MPA, VPA, and MVPA for the collapsed RT group and the SM group at baseline, 3 months, and 9 months are presented in Table 3. Exercise session attendance was 75.9% (SD⫽17.9%) for the collapsed RT group, and SM class attendance was 85.9% (SD⫽14.2%) for the SM group. However, attendance did not significantly (P⫽.48) differ between the collapsed RT group and the SM group. The RT groups significantly (P⬍.001) increased their leg press, leg curl, incline dumbbell press, seated row, and calf raise loads from baseline to 9 months (Tab. 2). Data

Number 3

from the ACLS questionnaire showed that few SM group participants (n⫽11, 19%) reported engaging in any form of resistance exercise throughout the intervention. Despite high exercise session attendance and significant improvements in muscle strength, very little time (minutes per day) was spent in VPA as measured by the accelerometer, and there were no significant (P⬎.05) differences in VPA among the intervention groups at any time interval (Tab. 3). Consequently, average daily MVPA (ⱖ3 METs) was representative of total time spent in health-enhancing PA intensities. The collapsed RT group participants increased their MVPA by 18% at 3 months (P⫽.001, effect size [d]⫽ 0.26) and by 10% at 9 months (P⫽.047, effect size [d]⫽0.15) compared with baseline levels. The SM group participants increased their MVPA by 22% at 3 months (P⫽.023, effect size [d]⫽0.25) and by 2% at 9 months (P⫽.80, effect size [d]⫽0.03) compared with baseline levels. After adjusting for baseline MVPA, age, BMI, sex, and knee OA pain, there was a significant (P⫽.034) group ⫻ time interaction for MVPA, which indicated that longitudinal MVPA decreased at a greater rate in the SM group than in the collapsed RT group (Fig. 2).

Discussion The overall increase in MVPA by the RT groups suggests that patients with early-onset OA of the knee can engage in a structured resistance exercise program without a compensatory decrease in MVPA levels. Compared with baseline, MVPA increased in the collapsed RT group by 18% at 3 months and 10% at 9 months, and the SM group showed a 22% increase in MVPA at 3 months but only a 2% increase at 9 months. These findings indicate that both SM and RT programs are effective for increasing short-term MVPA in paMarch 2010

Effects of Resistance Training on Physical Activity Table 3. Means (SD) and 95% Confidence Intervals for Average Time Spent in the 3 Physical Activity Intensities and for Knee Osteoarthritis Paina SM Group (nⴝ57)

RT Group (nⴝ52)

RTⴙSM Group (nⴝ62)

Collapsed RT and RTⴙSM Group (nⴝ114)

X (SD) [95%CI]

X (SD) [95%CI]

X (SD) [95%CI]

X (SD) [95%CI]

Baseline

23.4 (18.2) [18.9–28.6]

24.6 (17.8) [19.8–29.7]

27.9 (18.3) [23.2–32.6]

26.5 (18.1) [23.1–29.8]

3 mo

29.0 (22.5) [23.1–35.0]

27.9 (19.4) [23.1–32.7]

32.1 (17.1) [27.7–36.4]

30.2 (17.2) [27.0–33.4]

9 mo

24.1 (17.5) [19.4–28.7]

26.1 (17.7) [21.2–31.1]

30.1 (15.0) [26.3–33.9]

28.3 (16.3) [25.3–31.3]

Measure MPAb (min/d)

VPAc (min/d) Baseline

0.8 (2.9) [0.0–1.6]

0.8 (2.6) [0.1–1.6]

1.0 (2.0) [0.5–1.5]

0.9 (2.3) [0.5–1.4]

3 mo

0.7 (1.7) [0.3–1.2]

2.1 (4.9) [0.7–3.5]

1.6 (2.5) [0.9–2.2]

1.8 (3.8) [1.1–2.5]

9 mo

0.7 (2.1) [0.1–1.2]

1.6 (4.2) [0.4–2.8]

1.9 (3.7) [26.3–33.9]

1.8 (4.0) [1.0–2.5]

MVPAd (min/d) Baseline

24.2 (19.3) [19.4–29.7]

25.4 (19.4) [20.2–31.0]

28.6 (19.4) [24.0–33.8]

27.4 (19.4) [23.8–31.0]

3 mo

29.7 (23.1) [23.6–35.9]

30.0 (19.7) [24.5–33.3]

33.6 (18.2) [29.0–38.3]

32.0 (18.9) [28.5–35.5]

9 mo

24.8 (18.7) [19.7–29.6]

27.7 (20.0) [22.2–33.3]

31.8 (16.8) [27.7–36.2]

30.1 (18.3) [26.7–33.5]

Knee OA paine Baseline

82.2 (68.3) [64.1–100.3]

84.3 (70.1) [64.8–103.8]

81.9 (67.3) [64.8–99.0]

83.0 (68.3) [70.3–95.6]

3 mo

72.0 (66.3) [54.4–89.6]

47.6 (50.9) [33.5–61.8]

67.1 (68.8) [49.5–84.7]

58.2 (61.8) [46.6–69.7]

9 mo

62.9 (81.0) [41.4–84.4]

48.6 (61.3) [31.4–65.9]

56.2 (75.3) [36.7–75.6]

52.7 (29.1) [39.7–65.7]

SM⫽self-management, RT⫽resistance training, RT⫹SM⫽resistance training ⫹ self-management, CI⫽confidence interval, OA⫽osteoarthritis. b MPA⫽moderate-intensity physical activity (3– 6 metabolic equivalents [METs]). c VPA⫽vigorous-intensity physical activity (⬎6 METs). d MVPA⫽combined moderate- and vigorous-intensity physical activity (ⱖ3 METs). e Knee OA pain assessed with Western Ontario and McMaster Universities Osteoarthritis (WOMAC) Index pain subscale comprising 5 items eliciting patient ratings on visual analog scale (0 –100) of pain severity during walking, stair use, lying in bed at night, sitting, and standing. The pain subscale has a maximum summed score of 500, with higher scores reflecting more pain. a

tients with early OA of the knee, which is consistent with the findings of previous studies.38 – 40 The greater 3-month increase in MVPA in the SM group compared with the RT group may have resulted from differences in adherence (SM group⫽86%, RT group⫽76%) or because both programs encouraged PA. Although both treatments were effective in increasing short-term MVPA, RT was better than SM for maintaining longterm MVPA levels. Indeed, a significant group ⫻ time interaction indicated that the degree of change in MVPA was different between groups (ie, MVPA in the RT groups regressed between 3 and 9 months at a slower rate than in the SM group). Thus, rather than simply substituting RT for MVPA, the RT groups were March 2010

able to maintain MVPA levels in addition to attaining strength benefits from RT. It is possible that participation in the RT sessions may have contributed to the long-term maintenance of MVPA levels. On the other hand, adherence and MVPA in the SM group may have dropped off because of the intervention content. For example, early sessions focused on developing SM skills, whereas later sessions focused on reinforcing those skills. Thus, those participants who already mastered the skills may have seen less utility in the continued support. Previous studies have shown that when older adults participate in structured RT programs, there is a tendency for a compensatory de-

crease in aerobic MVPA.11,12 For example, Goran and Poehlman11 observed a decrease of ⬎544 kJ per day in free-living PA in elderly adults who were healthy after an 8-week high-intensity (85% of maximal oxygen consumption) training program. It is possible that the high intensity of the exercise program used in that study was too vigorous, thereby fatiguing the participants to the extent that they were no longer able to engage their regular PA throughout the remainder of the day. Meijer et al12 reported that a 12-week, moderateintensity combined aerobic and RT program resulted in improved physical fitness but had no effect on total daily PA (ie, after subtracting the PA of the exercise training sessions, this study showed that training PA was

Volume 90

Number 3

Physical Therapy f

363

Effects of Resistance Training on Physical Activity

Figure 2. Average daily moderate- and vigorous-intensity physical activity (MVPA, ⱖ3 metabolic equivalents) at 3 and 9 months of intervention for the self-management group (SM) and collapsed resistance training (RT) groups (RT and RT⫹SM). Values were adjusted for baseline MVPA, age, body mass index, sex, and knee osteoarthritis pain. There was a significant group ⫻ time interaction (P⫽.034) for MVPA.

compensated for by a decrease in nontraining PA, consistent with the findings of Goran and Poehlman11). In contrast, a 26-week RT program in a study by Hunter et al10 was not associated with a compensatory drop in free-living PA. The findings of Hunter et al are consistent with our findings because participants were able to engage in RT without substituting RT for MVPA. Although accelerometers allow for accurate measurements of daily time spent in various health-enhancing PA intensities (ie, MPA, VPA, and MVPA),26 hip placement may underestimate energy expenditure during certain activities (ie, biking, climbing stairs, and weight lifting) and provide no estimate of PA during water activities (eg, swimming) because accelerometers cannot be worn.28 Given the time spent standing, sitting, and lying during RT, this mode of activity is not measured well by accelerometers.28 This finding was evident in the present study because 364

f

Physical Therapy

Volume 90

despite rather high exercise session attendance (RT group⫽76%) and significant (P⬍.001) increases in load from baseline to 9 months for a number of RT exercises (eg, leg press, leg curl, seated row, incline dumbbell press, calf raise), average daily VPA in the RT group remained extremely low at 3 months (1.8 [SD⫽3.8] minutes per day) and 9 months (1.8 [SD⫽4.0] minutes per day). Thus, based on the observed improvements in strength, the RT groups likely received additional benefits of RT compared with the SM group, such as improved REE, TEE, AEE, lipid oxidation rates, musculoskeletal function, and body composition, which have been observed in response to RT in older adults.9,10,41 Furthermore, the ACLS questionnaire indicated that few SM group participants (n⫽11, 19%) engaged in any form of RT, and thus the majority of participants in this group did not receive additional benefits of RT.

Number 3

In addition to increasing TEE,10 RT has been shown to improve a number of functional limitations that lead to disability in patients with OA of the knee, such as quadriceps muscle weakness,42 neurological deficits,43 and decreased knee ROM.44 Furthermore, RT programs have been shown to improve psychological factors such as mood, self-efficacy, anxiety, and depression.45,46 Aerobic MVPA also can improve some of the same functional limitations in patients with OA of the knee and can reduce risks of obesity-linked health problems, including diabetes and cardiovascular disease,5,6 which often coexist with knee OA. Therefore, in clinical settings, the ability to engage in RT without sacrificing MVPA is important. We acknowledge several limitations of our study. For example, 5 to 7 days of contiguous accelerometer recordings may not be representative of habitual PA, and adipose tissue around the waist might affect the validity of the accelerometer outputs.25 Furthermore, accelerometers may underestimate PA during activities such as biking, climbing stairs, and weight lifting and cannot be worn during water activities (eg, swimming).28 Thus, PA may be underestimated in individuals who engage in these activities on a regular basis. More detailed physiological studies are needed in patients with OA of the knee to measure directly different energy expenditures (ie, REE, TEE, and AEE), which have been shown to be elevated in older adults in response to RT.9,10 Lastly, we cannot establish a causal relationship between increased levels of MVPA and improved physical function because we did not measure functional changes. However, a number of randomized controlled trials8,39,40,47,48 in patients with OA of the knee have shown a strong association between increased levels of MVPA and RT and improved physical March 2010

Effects of Resistance Training on Physical Activity function, which provides support for improved knee function of individuals who undertook the RT intervention.

Conclusion Patients with early-onset knee OA were able to engage in an RT program without a compensatory decrease in their overall MVPA levels. Because RT has been shown to increase energy expenditure in adults9,10,41 and has been shown to improve muscle strength and physical function and to reduce pain in patients with OA of the knee,8 it is a vital component of knee OA therapy. Given the relevant health benefits of RT and aerobic MVPA for management of knee OA, future studies are necessary to improve adherence to both modes of exercise. Dr Going and Dr McKnight provided concept/idea/research design. Mr Farr, Dr McKnight, and Dr Kasle provided writing. Mr Farr, Dr McKnight, Dr Kasle, and Dr Cornett provided data collection. Dr Kaske provided data management. Mr Farr, Dr McKnight, and Ms Cussler provided data analysis. Dr Going, Dr McKnight, and Dr Cornett provided project management. Dr McKnight provided fund procurement. Dr Going and Ms Cussler provided consultation (including review of manuscript before submission). The authors thank the men and women with knee OA who generously volunteered their time, the project coordinators for their oversight of all aspects of the study, and the other members of the Knee Study investigative team. Dr Isidro Villanueva is gratefully acknowledged for his contributions to the Knee Study. This study was approved by the University of Arizona Institutional Review Board and conducted in accordance with the Helsinki Declaration. This project was supported by National Institutes of Health/National Institute of Arthritis and Musculoskeletal and Skin Diseases grant R01-AR-047595. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of Arthritis and Musculoskeletal and Skin Diseases or the National Institutes of Health. NIH Clinical Trials Registry: NCT00586300.

March 2010

This article was received February 10, 2009, and was accepted October 30, 2009. DOI: 10.2522/ptj.20090041

References 1 Mazzuca SA. Is behavioral graded activity effective for the treatment of hip and knee osteoarthritis? Nat Clin Pract Rheumatol. 2007;3:322–323. 2 American College of Rheumatology Subcommittee on Osteoarthritis Guidelines. Recommendations for the medical management of osteoarthritis of the hip and knee: 2000 update. Arthritis Rheum. 2000;43:1905–1915. 3 Jordan KM, Arden NK, Doherty M, et al. EULAR Recommendations 2003: an evidence-based approach to the management of knee osteoarthritis: Report of a Task Force of the Standing Committee for International Clinical Studies Including Therapeutic Trials (ESCISIT). Ann Rheum Dis. 2003;62:1145–1155. 4 Scott DL, Shipley M, Dawson A, et al. The clinical management of rheumatoid arthritis and osteoarthritis: strategies for improving clinical effectiveness. Br J Rheumatol. 1998;37:546 –554. 5 American College of Sports Medicine Position Stand: The recommended quantity and quality of exercise for developing and maintaining cardiorespiratory and muscular fitness, and flexibility in healthy adults. Med Sci Sports Exerc. 1998;30:975–991. 6 US Department of Health and Human Services. Physical Activity and Health: A Report of the Surgeon General. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion; 1996. 7 Minor MA. 2002 Exercise and Physical Activity Conference, St Louis, Missouri: exercise and arthritis “we know a little bit about a lot of things em leader.” Arthritis Rheum. 2003;49:1–2. 8 Roddy E, Zhang W, Doherty M, et al. Evidence-based recommendations for the role of exercise in the management of osteoarthritis of the hip or knee: the MOVE consensus. Rheumatology (Oxford). 2005;44:67–73. 9 Withers RT, Smith DA, Tucker RC, et al. Energy metabolism in sedentary and active 49- to 70-year-old women. J Appl Physiol. 1998;84:1333–1340. 10 Hunter GR, Wetzstein CJ, Fields DA, et al. Resistance training increases total energy expenditure and free-living physical activity in older adults. J Appl Physiol. 2000; 89:977–984. 11 Goran MI, Poehlman ET. Endurance training does not enhance total energy expenditure in healthy elderly persons. Am J Physiol. 1992;263(5 pt 1):E950 –E957. 12 Meijer EP, Westerterp KR, Verstappen FT. Effect of exercise training on total daily physical activity in elderly humans. Eur J Appl Physiol Occup Physiol. 1999;80: 16 –21.

13 Booth FW, Gordon SE, Carlson CJ, Hamilton MT. Waging war on modern chronic diseases: primary prevention through exercise biology. J Appl Physiol. 2000;88: 774 –787. 14 Pate RR, Pratt M, Blair SN, et al. Physical activity and public health: a recommendation from the Centers for Disease Control and Prevention and the American College of Sports Medicine. JAMA. 1995;273: 402– 407. 15 Hart DJ, Spector TD, Brown P, et al. Clinical signs of early osteoarthritis: reproducibility and relation to x-ray changes in 541 women in the general population. Ann Rheum Dis. 1991;50:467– 470. 16 Sullivan T, Allegrante JP, Peterson MG, et al. One-year followup of patients with osteoarthritis of the knee who participated in a program of supervised fitness walking and supportive patient education. Arthritis Care Res. 1998;11:228 –233. 17 Kellgren JH, Lawrence JS. Radiological assessment of osteo-arthrosis. Ann Rheum Dis. 1957;16:494 –502. 18 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. 19 Altman R, Asch E, Bloch D, et al; for the Diagnostic and Therapeutic Criteria Committee of the American Rheumatism Association. Development of criteria for the classification and reporting of osteoarthritis: classification of osteoarthritis of the knee. Arthritis Rheum. 1986;29: 1039 –1049. 20 Pearson D, Faigenbaum A, Conley M, Kraemer WJ. The National Strength and Conditioning Association’s basic guidelines for the resistance training of athletes. Strength and Conditioning Journal. 2000;22:14 –27. 21 Baker KR, Nelson ME, Felson DT, et al. The efficacy of home based progressive strength training in older adults with knee osteoarthritis: a randomized controlled trial. J Rheumatol. 2001;28:1655–1665. 22 Borg GA. Psychophysical bases of perceived exertion. Med Sci Sports Exerc. 1982;14:377–381. 23 Lorig KR, Holman H. Self-management education: history, definition, outcomes, and mechanisms. Ann Behav Med. 2003;26: 1–7. 24 Lohman TG, Roche AF, Martorell R. Anthropometric Standardization Reference Manual. Champaign, IL: Human Kinetics Publishers; 1988. 25 Farr JN, Going SB, Lohman TG, et al. Physical activity levels in patients with early knee osteoarthritis measured by accelerometry. Arthritis Rheum. 2008;59:1229 –1236. 26 Freedson PS, Melanson E, Sirard J. Calibration of the Computer Science and Applications Inc accelerometer. Med Sci Sports Exerc. 1998;30:777–781. 27 Tryon W, Williams R. Fully proportional actigraphy: a new instrument. Behav Res Meth Instrum Comput. 1996;28:392– 403.

Volume 90

Number 3

Physical Therapy f

365

Effects of Resistance Training on Physical Activity 28 Matthew CE. Calibration of accelerometer output for adults. Med Sci Sports Exerc. 2005;37(11 suppl):S512–S522. 29 Hendelman D, Miller K, Baggett C, et al. Validity of accelerometry for the assessment of moderate intensity physical activity in the field. Med Sci Sports Exerc. 2000; 32(9 suppl):S442–S449. 30 Computer Science and Applications Inc Activity Monitor Operator’s Manual, Model 7164, Release 1.04. Shalimar, FL: Computer Science and Applications Inc; 1995. 31 Matthews CE, Ainsworth BE, Thompson RW, Bassett DR Jr. Sources of variance in daily physical activity levels as measured by an accelerometer. Med Sci Sports Exerc. 2002;34:1376 –1381. 32 Brage S, Wedderkopp N, Franks PW, et al. Reexamination of validity and reliability of the CSA monitor in walking and running. Med Sci Sports Exerc. 2003;35:1447–1454. 33 Nichols JF, Morgan CG, Chabot LE, et al. Assessment of physical activity with the Computer Science and Applications Inc accelerometer: laboratory versus field validation. Res Q Exerc Sport. 2000;71: 36 – 43. 34 Yngve A, Nilsson A, Sjostrom M, Ekelund U. Effect of monitor placement and of activity setting on the MTI accelerometer output. Med Sci Sports Exerc. 2003;35: 320 –326.

366

f

Physical Therapy

Volume 90

35 Kohl HW, Blair SN, Paffembarger RS, et al. The Aerobics Center Longitudinal Study Physical Activity Questionnaire: a collection of physical activity questionnaires for health-related research. Med Sci Sports Exerc. 1997;29(suppl):S10 –S14. 36 Troiano RP, Berrigan D, Dodd KW, et al. Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc. 2008;40:181–188. 37 Cohen J. A power primer. Psychol Bull. 1992;112:155–159. 38 Evcik D, Sonel B. Effectiveness of a homebased exercise therapy and walking program on osteoarthritis of the knee. Rheumatol Int. 2002;22:103–106. 39 Kovar PA, Allegrante JP, MacKenzie CR, et al. Supervised fitness walking in patients with osteoarthritis of the knee: a randomized, controlled trial. Ann Intern Med. 1992;116:529 –534. 40 Talbot LA, Gaines JM, Huynh TN, Metter EJ. A home-based pedometer-driven walking program to increase physical activity in older adults with osteoarthritis of the knee: a preliminary study. J Am Geriatr Soc. 2003;51:387–392. 41 Van Etten LM, Westerterp KR, Verstappen FT, et al. Effect of an 18-week weighttraining program on energy expenditure and physical activity. J Appl Physiol. 1997; 82:298 –304. 42 Schilke JM, Johnson GO, Housh TJ, O’Dell JR. Effects of muscle-strength training on the functional status of patients with osteoarthritis of the knee joint. Nurs Res. 1996;45:68 –72.

Number 3

43 Hurley MV, Scott DL. Improvements in quadriceps sensorimotor function and disability of patients with knee osteoarthritis following a clinically practicable exercise regime. Br J Rheumatol. 1998;37: 1181–1187. 44 Badley EM, Wagstaff S, Wood PH. Measures of functional ability (disability) in arthritis in relation to impairment of range of joint movement. Ann Rheum Dis. 1984; 43:563–569. 45 Byrne A, Byrne DG. The effect of exercise on depression, anxiety and other mood states: a review. J Psychosom Res. 1993; 37:565–574. 46 McAuley E, Shaffer S, Rudolph D. Physical activity, aging, and psychological well-being. JAPA. 1995;3:67–96. 47 Ettinger WH Jr, Burns R, Messier SP, et al. A randomized trial comparing aerobic exercise and resistance exercise with a health education program in older adults with knee osteoarthritis: the Fitness Arthritis and Seniors Trial (FAST). JAMA. 1997;277:25–31. 48 Bautch JC, Malone DG, Vailas AC. Effects of exercise on knee joints with osteoarthritis: a pilot study of biologic markers. Arthritis Care Res. 1997;10:48 –55.

March 2010

Research Report Pediatric Endurance and Limb Strengthening (PEDALS) for Children With Cerebral Palsy Using Stationary Cycling: A Randomized Controlled Trial Eileen G. Fowler, Loretta M. Knutson, Sharon K. DeMuth, Kara L. Siebert, Victoria D. Simms, Mia H. Sugi, Richard B. Souza, Roksana Karim, Stanley P. Azen; for the Physical Therapy Clinical Research Network (PTClinResNet)

Background. Effective interventions to improve and maintain strength (forcegenerating capacity) and endurance are needed for children with cerebral palsy (CP). Objective. This study was performed to examine the effects of a stationary cycling intervention on muscle strength, locomotor endurance, preferred walking speed, and gross motor function in children with spastic diplegic CP.

Design. This was a phase I randomized controlled trial with single blinding. Setting. The interventions were performed in community-based outpatient physical therapy clinics. Outcome assessments were performed in university laboratories.

Participants. Sixty-two ambulatory children aged 7 to 18 years with spastic diplegic CP and Gross Motor Function Classification System levels I to III participated in this study. Intervention and Measurements. Participants were randomly assigned to cycling or control (no-intervention) groups. Thirty intervention sessions occurred over 12 weeks. Primary outcomes were peak knee extensor and flexor moments, the 600-Yard Walk-Run Test, the Thirty-Second Walk Test, and the Gross Motor Function Measure sections D and E (GMFM-66). Results. Significant baseline-postintervention improvements were found for the 600-Yard Walk-Run Test, the GMFM-66, peak knee extensor moments at 120°/s, and peak knee flexor moments at 30°/s for the cycling group. Improved peak knee flexor moments at 120°/s were found for the control group only, although not all participants could complete this speed of testing. Significant differences between the cycling and control groups based on change scores were not found for any outcomes.

Limitations. Heterogeneity of the patient population and intrasubject variability were limitations of the study.

Conclusions. Significant improvements in locomotor endurance, gross motor function, and some measures of strength were found for the cycling group but not the control group, providing preliminary support for this intervention. As statistical differences were not found in baseline-postintervention change scores between the 2 groups; the results did not demonstrate that stationary cycling was more effective than no intervention. The results of this phase I study provide guidance for future research. March 2010

Volume 90

E.G. Fowler, PT, PhD, is Associate Professor, Department of Orthopaedic Surgery, Director of Research and Education and Peter William Shapiro Chair, UCLA/Orthopaedic Hospital Center for Cerebral Palsy, University of California at Los Angeles, and Faculty, Tarjan Center at UCLA, 22-70 Rehabilitation Center, 1000 Veteran Ave, Los Angeles, CA 90095-1795 (USA). Address all correspondence to Dr Fowler at: efowler@mednet. ucla.edu. L.M. Knutson, PT, PhD, PCS, is Adjunct Faculty, Krannert School of Physical Therapy, University of Indianapolis, Indianapolis, Indiana. She was Professor, Department of Physical Therapy, Missouri State University, Springfield, Missouri, at the time this study was conducted. S.K. DeMuth, PT, DPT, MS, is Assistant Professor of Clinical Physical Therapy, Division of Biokinesiology and Physical Therapy at the School of Dentistry, University of Southern California, Los Angeles, California. K.L Siebert, PT, DPT, MEd, is Staff Physical Therapist and Clinical Instructor, South Bay Medical Therapy Unit, California Children’s Services of Los Angeles County. She was Postdoctoral Fellow, Department of Orthopaedic Surgery, UCLA/Orthopaedic Hospital Center for Cerebral Palsy, University of California at Los Angeles, at the time this study was conducted. V.D. Simms, PT, MPT, was Research Assistant, Department of Physical Therapy, Missouri State University, at the time this study was conducted. M.H. Sugi, PT, DPT, is Staff Physical Therapist, Hollywood Presbyterian Medical Center, Los Angeles, California, and Staff Physical Therapist, Every Child Achieves, Studio City, California. She was Postdoctoral Fellow, Department of Orthopaedic Surgery, UCLA/ Orthopaedic Hospital Center for Cerebral Palsy, University of California at Los Angeles, at the time this study was conducted. Author information continues on next page.

Number 3

Post a Rapid Response or find The Bottom Line: www.ptjournal.org Physical Therapy f

367

Pediatric Endurance and Limb Strengthening R.B. Souza, PT, PhD, is Postdoctoral Scholar in Radiology and Biomedical Imaging, University of California, San Francisco, California. He was a doctoral student in the Division of Biokinesiology and Physical Therapy at the School of Dentistry, University of Southern California, at the time the study was conducted. R. Karim, MBBS, PhD, is Assistant Professor of Research, Department of Preventive Medicine, Keck School of Medicine, University of Southern California. S.P. Azen, PhD, is Professor, Department of Preventive Medicine, Keck School of Medicine, University of Southern California. Physical Therapy Clinical Research Network (PTClinResNet) (see list of investigators on page 379). [Fowler EG, Knutson LM, DeMuth SK, et al; Physical Therapy Clinical Research Network (PTClinResNet). Pediatric Endurance and Limb Strengthening (PEDALS) for children with cerebral palsy using stationary cycling: a randomized controlled trial. Phys Ther. 2010;90:367–381.] © 2010 American Physical Therapy Association

Available With This Article at ptjournal.apta.org • Discussion Podcast: With Doreen Bartlett and author Eileen Fowler. Carolynn Patten (moderator). • Audio Abstracts Podcast This article was published ahead of print on January 21, 2010, at ptjournal.apta.org.

368

f

Physical Therapy

Volume 90

C

hildren with cerebral palsy (CP) have decreased capacity to participate in play and sports activities at intensities sufficient to develop and maintain adequate levels of muscular strength (force-generating capacity) and cardiorespiratory fitness.1–3 The interaction of these factors can lead to a continuous cycle of deconditioning and decreased functional ability.4 These findings are of concern because they can exacerbate secondary effects associated with CP5 and reduced overall health and well-being.6 Therefore, safe and effective interventions to improve and maintain strength and endurance are needed.5,7 Although the overall level of research evidence supporting exercise interventions for children with CP is low, positive results, particularly for strengthening, have been reported.7,8 The problem of reduced endurance in children with CP has received little attention. In a systematic review of the literature, only 5 studies were found that addressed lowerextremity aerobic exercise for children with CP.7 Interventions included lower-limb cycling,9 –11 12 10,12,13 walking, running, jumping,13 12 10 stepping, swimming, and mat exercises.10 Only 1 study was a randomized controlled trial (RCT).10 The other 4 studies were given the lowest evidence rating level.7 Three of these 5 studies demonstrated statistically significant improvements in aerobic capacity following interventions at frequencies of 2 or 4 times per week.10,11,13 Additionally, interventions that combined strengthening14,15 or anaerobic16 exercises with cardiorespiratory training have been investigated. The largest of these studies16 was an RCT that examined the effect of a school-based program of aerobic and anaerobic exercise for 65 children with spastic CP over an 8-month period. Significant improvement (P⬍.05) was found for aerobic

Number 3

capacity, anaerobic capacity, agility, strength, participation, and quality of life. Outcome measures varied considerably among all of these studies. Some researchers focused on measures of oxygen consumption or heart rate (HR),9 –15 whereas others used field tests,16,17 such as the 10-m shuttle run test or the 600-Yard Walk-Run Test. These field tests require the child to walk or run as fast as possible. Correlations between field tests and laboratory measures of aerobic capacity have been reported for children with disability.17 Results of field tests may convey more about a child’s ability to keep up with his or her peers during school, sports, and play activities. Lower-extremity cycling is a rehabilitation tool used by physical therapists to improve strength and cardiorespiratory fitness and appears wellsuited as a therapeutic intervention for children with CP. Simultaneous strengthening of hip, knee, and ankle musculature may be achieved without the need to perform isolated joint movement out of the basic flexion and extension movement synergies. In contrast to aerobic exercises that require walking or running, cycling is less dependent on balance, coordination, and motor control. Cycling may induce positive speedrelated changes in neuromotor control and muscle physiology by promoting higher speeds of movement than are possible during daily activities of most children with CP. Although cycling has been recommended as an appropriate exercise for individuals with CP,6,18 research is limited. Children with CP exhibiting a wide range of disability were able to improve oxygen uptake, at a given HR, following an intervention emphasizing stationary cycling.9 Six adolescents with mild CP improved their physical endurance during cycling, as evidenced by increased oxygen consumption at the anaerobic March 2010

Pediatric Endurance and Limb Strengthening

A detailed description of the PEDALS RCT protocol has been reported elsewhere.20

tions; (3) ability to walk independently, with or without an assistive device, for short distances (Gross Motor Function Classification System [GMFCS] levels I–III)21; and (4) good or fair selective voluntary motor control for at least one limb. Good selective voluntary motor control was defined as the ability to isolate both knee and ankle movement out of synergy (knee extension with the hip positioned in flexion; ankle dorsiflexion with the knee positioned in extension). Fair selective voluntary motor control was defined as the ability to isolate knee extension but not ankle dorsiflexion. Exclusion criteria were: (1) orthopedic surgery, neurological surgery, or baclofen pump implantation within the preceding 12 months; (2) botulinum toxin injections within the preceding 3 months; (3) serial casting or new orthotic devices within the preceding 3 months; (4) initiation of oral medications that affect the neuromuscular system (eg, baclofen) within the preceding 3 months; (5) initiation of physical therapy, exercise, sports activity, or change in assistive devices for walking within the preceding 3 months; (6) inability or unwillingness to maintain ageappropriate behavior; (7) serious medical conditions such as cardiac disease, diabetes, or uncontrolled seizures; (8) current participation in a fitness program that included a minimum of once-weekly cardiorespiratory endurance exercise; (9) significant hip, knee, or ankle joint contractures preventing passive movement of the lower limbs through the pedaling cycle; and (10) bilateral poor selective voluntary motor control (inability to isolate knee or ankle joint motion out of synergy).

Participants All participants had spastic diplegic CP. Inclusion criteria were: (1) between 7 and 18 years of age; (2) ability to follow simple verbal direc-

Participants were recruited from southern California and southwest Missouri via flyers and brochures placed in clinics and schools, mailed, or posted on disability-related Web

threshold, following the first 3 months of a stationary cycling intervention.11 Benefits also have been reported for children with more severe physical disability. Following a 6-week intervention using an adapted stationary bicycle, 11 nonambulatory adolescents with CP improved their gross motor function (P⫽.01).19 None of these studies reported the effect of cycling on lowerextremity strength, preferred walking speed, or walking and running endurance. Considering the limited research in this area, the Pediatric Endurance Development and Limb Strengthening (PEDALS) project for children with CP was designed as a phase I preliminary investigation to examine the effectiveness of a communitybased stationary cycling intervention. Stationary cycling allows precise definition of exercise intensity, duration, and systematic guidelines for exercise progression—important factors for research. Our goal was to improve strength and walking and running endurance in ambulatory children with spastic diplegic CP. To generalize the results, the intervention was conducted in partnership with pediatric physical therapy clinics under “typical clinical conditions.” Our hypothesis was that children who participated in a 12-week, stationary cycling intervention would improve their preferred walking speed, walking and running endurance, gross motor function, knee extensor and flexor strength, and gait kinematics.

Method

March 2010

sites. A telephone screening was performed for potential participants who contacted the investigators. Children meeting the study criteria received an in-person screening to confirm their diagnosis and assess GMFCS level, selective voluntary motor control, and range of motion. An interpreter translated for parents or guardians who did not speak English. The institutional review board of each institution approved the study protocol and consent procedures. Informed consent was obtained from a parent or guardian and from participants over the age of 14 years. Assent was obtained from each participant under the age of 14 years. If formal physical therapy had been initiated or discontinued recently, we postponed baseline data collection until 3 months had elapsed. For the duration of the study, participants who were receiving physical therapy were asked to maintain their present regimen. Study Design This study was a phase I, multi-site RCT with single blinding. Power analyses determined that a sample size of 58 participants (29 intervention, 29 control) would have 80% power to detect a moderate effect size of 0.7 associated with a 15% strength improvement. This gain was a conservative estimate based on improved peak knee extensor and flexor moments following an isokinetic knee strengthening program.22 Outcome measurements were assessed at baseline and following the 12-week intervention period. Children were randomly assigned to either an intervention (cycling) group or a control (no cycling) group. Randomization was blocked by age group (7–11 years, 12–18 years) and selective voluntary motor control ability (good, fair)20 to minimize the effects of maturation and physical impairment. Participants who demonstrated good selective voluntary motor control bilaterally were

Volume 90

Number 3

Physical Therapy f

369

Pediatric Endurance and Limb Strengthening placed in the “good” selective voluntary motor control category for stratification. Those with fair motor control for at least one limb were placed in the “fair” category. Outcome Measures A conscious effort was made to select outcome measures that differed from the skill practiced during the intervention and that had functional meaning to the families and clinicians. This article reports the results for PEDALS primary outcome measures at the body function and activity levels of the International Classification of Functioning, Disability and Health23: (1) the 600-Yard WalkRun Test,17 (2) the Thirty-Second Walk Test (30sWT),24 (3) the Gross Motor Function Measure sections D and E (GMFM-66),25 and (4) peak knee extensor and flexor isometric and isokinetic moments. In addition, gait analysis results, obtained for a subset of participants, are included. Data collection took place at the University of Southern California and Missouri State University. Evaluators were blinded to participant group assignment and had to pass a rigorous standardization procedure for each outcome measurement protocol by demonstrating 90% competency. Each participant’s height and weight were recorded. Walking and running tests were performed on a circular path at a nearby track or school gymnasium. For the 600-Yard Walk-Run Test, children were directed to walk or run as fast as they could, and the time to complete the distance was recorded. If a participant could not complete the test within the 15-minute time limit, the distance covered and time elapsed were recorded. For the 30sWT, children were asked to walk at their preferred speed. The distance completed in 30 seconds was recorded. The GMFM-66 scores were obtained using sections D (standing) and E (walking, running, and jumping). 370

f

Physical Therapy

Volume 90

We examined peak knee extensor and flexor moments across a range of speeds to capture changes reflective of muscle strength, power, and endurance. A KinCom dynamometer* was used at the Los Angeles site, and a Biodex Multijoint System† was used at the Missouri site. Five repetitions of knee joint extension and flexion at 0, 30, 60, and 120°/s were performed bilaterally. A subgroup of 16 children (8 from each group) underwent gait analysis. Three-dimensional motion analysis was performed using a Vicon motion system‡ at 60 Hz. Calibration markers were placed over specific anatomical landmarks to define lowerextremity segments, and tracking clusters consisting of 3 or 4 markers were placed on the lateral surface of the thigh, leg, and lateral calcaneus. Calibration markers were removed following collection of measurements for a standing trial. Measurements for 3 walking trials at a selfselected speed were collected for each participant. Physical Activity Calendars All participants were provided with physical activity calendars for the 12week intervention period so that differences in physical activity between the 2 groups could be quantified. Participants were instructed to place a sticker on each day of the calendar that corresponded to the following activity levels. Running or jogging, participating in contact sports, hiking, dancing, climbing stairs, or biking for approximately 1 hour per day was considered a high level of activity. Participating in the same activities for 30 minutes per day or in activities such as swimming, skateboarding, scooter riding, or walking * Chattanooga Group, 4717 Adams Rd, Hixson, TN 37343. † Biodex Medical Systems, 20 Ramsay Rd, Shirley, NY 11967-4704. ‡ Oxford Metrics Ltd, 14 Minns Business Park, West Way, Oxford OX2 0JB, United Kingdom.

Number 3

for approximately 1 hour was considered a moderate level of activity. A low activity level indicated a sedentary day that included schoolwork, watching television, or playing computer games. Bed rest indicated that a child was inactive due to illness or injury. Participants assigned to the cycling group were instructed to exclude the cycling intervention when recording their daily activity levels. Cycling Intervention The cycling intervention occurred in community-based pediatric physical therapy clinics. Standardization was ensured by using detailed intervention protocols. Each intervention physical therapist demonstrated 90% competency for the performance of critical components. The stationary bicycle† used for this study was designed for rehabilitation. Features included a semirecumbent design with a wide padded seat, trunk support, foot straps, and a unique “cyclocentric” lower-limb–loading feature to provide resistance.26 The cycling intervention was performed 3 times per week, for a total of 30 sessions, within a 12-week period. A generalized stretching program was performed prior to cycling. Ankle-foot orthoses, if used for walking, were worn during cycling. Resting HR was recorded prior to cycling. If the participant could not cycle independently, manual assistance was provided until independence was achieved. If limb movement was not maintained in the sagittal plane, corrections were made using physical guidance by the therapist, verbal cueing, or adaptations, such as modification of the foot position on the pedal. Each 60minute cycling session was divided into 2 phases: (1) lower-extremity strengthening and (2) cardiorespiratory endurance.

March 2010

Pediatric Endurance and Limb Strengthening Phase 1: lower-extremity strengthening. The cyclocentric strengthening protocol was initiated after independent cycling was achieved. The bicycle seat was unlocked and allowed to slide backward along a linear track. Up to 10 tension cords, each providing 10 lb (1 lb⫽0.4536 kg) of force, acted to pull the seat forward. Lower-limb extension was required to prevent the seat from being pulled forward and to maintain the seat in a range demarcated as the “cyclocentric exercise zone.” Training began with the attachment of one tensioning cord. Resistance was progressed to the next higher cord when 10 revolutions were performed in a smooth pattern while keeping the seat within the desired zone. Subsequent sessions began with a warm-up at previously attained resistance levels prior to progressing to a higher level of resistance. For each session, minimal and maximal resistance and the number of revolutions for each cord were recorded. If a participant could not cycle with the seat unlocked or if the maximum resistance (10 cords) was reached, a “constant power” resistance mode, typical for most stationary bicycles, was used. Phase 2: cardiorespiratory endurance. The goal of this phase was to gradually increase duration and intensity. The seat was locked in a location that positioned the participant’s knee in 15 to 20 degrees of flexion when maximally extended. Heart rate was monitored using a sensor attached to the participant’s ear or the chest. A target HR range of 70% to 80% of maximum heart rate (HRmax) was calculated for every session using the Karvonen formula: target HR ⫽ [(HRmax ⫺ HRrest) ⫻ 0.70 – 0.80] ⫹ HRrest, where HRrest ⫽ resting heart rate and HRmax ⫽ 220 ⫺ age. Typical exercise heart rate (TEHR) was documented at the conclusion of all sessions.

March 2010

Cycling was initiated at the lowest level of constant power mode resistance and was adjusted according to the participant’s ability. Children were verbally encouraged to cycle as fast as they could with a goal of increasing their HR to within their target range. Each participant reported perceived exertion throughout the cycling session using the Children’s Effort Rating Table (CERT), a 1 to 10 scale with verbal descriptions corresponding to each number.27 Resistance and pedaling rate were adjusted based on HR and CERT rating. The exercise duration goal was 15 to 30 minutes. A cool-down period consisted of pedaling without resistance until HR decreased to within 20 bpm above the baseline measurement. Data Analysis The 600-Yard Walk-Run Test data were converted to speed (m/min). The GMFM-66 scores were calculated using the gross motor ability estimator software. Peak joint moments from the left and right limbs were averaged for each speed. If limb movement did not meet the specified speed, a joint moment could not be obtained, thus decreasing the number of limbs included for analysis. Peak hip extension, knee extension, and ankle dorsiflexion angular positions during the stance phase of gait, as well as peak hip and knee flexion angles during the swing phase of gait, were calculated from joint marker data. Statistical tests were conducted using JMP version 6.0 software§ and SAS version 9,§ with significance level set at P⬍.05. Demographics, mobility level, anthropometrics, related medical history, and baseline primary outcome measurements were compared between the cycling and control groups using chi-square tests for comparison of proportions

and one-way analysis of variance for continuous variables. Baselinepostintervention change scores were calculated for each outcome measure. Paired t tests were applied to examine baseline-postintervention differences within the cycling and control groups. Independent t tests were used to examine betweengroup differences in change scores. For data that were not normally distributed, Wilcoxon rank sum tests were used. Role of the Funding Source This study was supported by a grant from the Foundation for Physical Therapy to establish PTClinResNet, a clinical research network to evaluate the efficacy of physical therapist practice.

Results Recruitment and Retention The Figure summarizes our trial profile. Of 129 individuals (88 in California, 41 in Missouri) who responded to the recruitment efforts, 65 individuals were excluded during telephone or in-person screening; specific reasons are reported in Table 1. Of the 64 individuals who were randomly assigned to the cycling and control groups, 6 later withdrew from the study. Two participants withdrew for personal reasons prior to baseline data collection. During the intervention period, an additional 2 participants withdrew for personal reasons, and 2 others did not maintain the criteria necessary for inclusion and were withdrawn by the investigators. One child initiated an intensive sports program, and the other child underwent a medical treatment for vision during the study. A total of 58 participants (29 in the cycling group, 29 in the control group; 37 from the California site, and 21 from the Missouri site) completed the study.

§

SAS Institute Inc, PO Box 8000, Cary, NC 27513

Volume 90

Number 3

Physical Therapy f

371

Pediatric Endurance and Limb Strengthening

Figure. CONSORT diagram illustrating the flow of participants through the trial.

Table 1. Summary of Recruitment Efforts and Ineligibility by Sitea Reason for Exclusion

California

Missouri

88

41

4

4

Total no. of contacts Protocol-specific reasons Age ⬍7 y Age ⬎18 y

1

2

Diagnosis other than CP

1

0

Has CP, but not spastic diplegic CP

4

4

Unable to follow simple verbal directions

3

4

GMFCS levels I to III not met

5

2

Musculoskeletal or neurosurgical surgery or baclofen pump implantation within the past year

5

0

Casted or received braces in past 3 mo

2

0

Health complications

3

1

Legally blind

1

0

1

0

Serious medical conditions

Participant-specific reasons Lack of transportation Personal reasons or failed appointments

3

0

12

3

Total no. of exclusions

45

20

Total no. of participants enrolled

43

21

Family not interested

a

CP⫽cerebral palsy, GMFCS⫽Gross Motor Function Classification System.

372

f

Physical Therapy

Volume 90

Number 3

Demographics, Participant Characteristics, and Baseline Outcome Measures Significant differences were not found for demographic data, participant characteristics, and baseline measures between the cycling and control groups (Tab. 2). A history of visual impairment was the most commonly reported medical problem, and the incidence was higher in the control group, but not significantly different (P⫽.07) from the cycling group. There was considerable racial and ethnic diversity; 49% of participants reported their race as African American, Asian, or “other” and 31% reported Hispanic ethnicity. Although all children were fluent in English, 19% of their parents or guardians were not. Due to difficulty in recruiting participants for the older age category (12–18 years), the constraint for equality between age groups was dropped, resulting in a greater number of participants in the younger age category (7–11 years). Blocking was maintained for selective voluntary motor control categories; therefore, similar numbers of participants were classified as “fair” or “good.” Participants at GMFCS level III had greater representation than those at levels I and II. A history of speech, learning, attention, or behavioral problems and asthma were common. Adherence, Protocol Variations, and Adverse Events The adherence rate for cycling group session attendance was 89.6%. Protocol variations occurred for 3 participants who missed 1, 3, or 4 of the 30 scheduled sessions. Fifty-eight adverse events were reported. Twentyeight mild events, for 18 participants, were potentially related to the study procedures. These events were 6 observed falls; 17 complaints of mild pain, soreness, or muscle cramping; 4 reports of feeling fatigued; and 1 report of skin rash related to the HR sensor. Thirty adMarch 2010

Pediatric Endurance and Limb Strengthening Table 2. Demographics, Characteristics, and Primary Outcome Measures for Cycling and Control Group Participants at Baseline (N⫽62)a Variable

Cycling Group (nⴝ31)

Control Group (nⴝ31)

Pb

Demographics Sex: male

18 (58%)

11 (36%)

.13

Age (y)

11.1 (9.9–12.3)

11.6 (10.6–12.6)

.59

Ethnicity: Hispanic

12 (39%)

7 (23%)

.27 .52

Race African American

5 (16%)

3 (10%)

White

18 (58%)

15 (48%)

Asian

1 (3%)

5 (16%)

Other

7 (23%)

8 (26%)

English

24 (77%)

26 (87%)

Spanish

6 (19%)

4 (13%)

Other

1 (3%)

1 (3%)

Parental language .79

Age categories (y) 7–11

20 (65%)

18 (58%)

12–18

11 (36%)

13 (42%)

Fair

17 (55%)

15 (48%)

Good

14 (45%)

16 (52%)

.80

Selective voluntary motor control .80

Mobility GMFCS level I

11 (35%)

8 (26%)

GMFCS level II

8 (26%)

6 (19%)

GMFCS level III

12 (39%)

17 (55%)

.52

Anthropometrics Height (m) Weight (kg)

1.38 (1.32–1.44) 38.8 (32.9–44.6)

1.38 (1.3–1.4)

.94

37.9 (33.0–43.0)

.83

Related medical history Asthma

11 (36%)

6 (19%)

.25

Attention/behavioral problems

8 (26%)

8 (26%)

⬎.99

Mental retardation

4 (13%)

4 (13%)

⬎.99

Seizure disorder

2 (7%)

4 (13%)

.67

Learning problems

10 (32%)

16 (52%)

.20

Speech problems

11 (36%)

10 (32%)

⬎.99

Vision problems

15 (48%)

23 (74%)

1 (3%)

2 (7%)

Hearing problems

.07 ⬎.99

Primary outcomes at baseline 30sWT speed (m/min)

66.0 (57.9–74.0)

57.7 (50.4–65.1)

.13

600-Yard Walk-Run Test speed (m/min)

87.7 (73.4–102.0)

80.3 (65.9–94.7)

.46

GMFM-66 (maximum score⫽100)

69.0 (64.9–73.1)

68.8 (64.7–72.9)

.96

Peak knee extensor moments (N䡠m/kg) 0°/s (n⫽28/28)

1.23 (1.04–1.42)

1.12 (0.99–1.26)

.34

30°/s (n⫽31/31)

1.03 (0.91–1.17)

1.09 (0.92–1.26)

.63

60°/s (n⫽30/31)

0.86 (0.75–0.97)

0.87 (0.71–1.02)

.92

120°/s (n⫽27/28)

0.65 (0.57–0.74)

0.70 (0.58–0.81)

.55

Peak knee flexor moments (N䡠m/kg) 0°/s (n⫽26/27)

0.44 (0.33–0.54)

0.39 (0.27–0.52)

.61

30°/s (n⫽30/28)

0.29 (0.22–0.35)

0.34 (0.24–0.44)

.39

60°/s (n⫽27/29)

0.28 (0.21–0.34)

0.28 (0.20–0.37)

.93

120°/s (n⫽22/23)

0.21 (0.16–0.30)

0.21 (0.14–0.28)

.98

a

Values are mean (95% confidence intervals) for continuous variables, frequency (%) for categorical variables. Related medical history was obtained from the parent Pediatric Outcomes Data Collection Instrument (PODCI) questionnaire. GMFCS⫽Gross Motor Function Classification System, 30sWT⫽ThirtySecond Walk Test, GMFM-66⫽Gross Motor Function Measure (66 items). b Chi-square test for categorical variables; one-way analysis of variance for continuous variables.

March 2010

Volume 90

Number 3

Physical Therapy f

373

Pediatric Endurance and Limb Strengthening Table 3. Gait Speed and Gross Motor Function Outcomesa Measure

Cycling Group

600-Yard Walk-Run Test speed (m/min)

Control Group

n⫽27

n⫽28

Baseline

85.0 (69.7 to 100.4)

81.6 (65.9 to 97.4)

Postintervention

90.6 (75.4 to 105.7)

84.1 (67.6 to 100.7)

5.6 (1.6 to 9.5)

2.5 (⫺1.1 to 6.0)

Changec

d

P 30sWT speed (m/min)

.008

.16

n⫽29

n⫽29

Baseline

66.9 (58.6 to 75.1)

58.7 (51.0 to 66.5)

Postintervention

68.0 (60.4 to 75.7)

62.1 (54.4 to 69.8)

1.2 (⫺3.9 to 6.2)

3.4 (⫺1.7 to 8.4)

.64

.18

Change P GMFM-66

n⫽29

n⫽29

Baseline

69.6 (65.4 to 73.8)

68.8 (64.5 to 73.0)

Postintervention

70.8 (66.6 to 74.9)

69.3 (65.4 to 73.3)

1.2 (0.5 to 1.8)

0.5 (⫺0.2 to 1.3)

Change P

.002

d

Pb

.24

.

.52

.23

.12

a

Values are mean (95% confidence intervals). 30sWT⫽Thirty-Second Walk Test, GMFM-66⫽Gross Motor Function Measure (66 items). b P value for between-group comparisons using independent t tests. c Postintervention change calculated by subtracting baseline value from postsession value. d P⬍.05 for baseline-postintervention comparison using paired t tests.

verse events not related to the study were illness (colds, flu), tooth loss, headache, stomachache, tonsillectomy, and skin irritation due to orthotic wear. Physical activity calendars indicated that the number of days with high or moderate levels of activity were similar for the cycling and control groups (64.8 and 64.4%, respectively). There was a shift toward high levels of activity for the cycling group (32.3% versus 21.7%). Reports of bed rest were slightly higher for the cycling group (4.9% versus 1.2%). “Flu” and “colds” were most the most commonly recorded comments on these days. Cycling Group Training Intensity The majority of participants were able to perform the strengthening task for phase 1 using the cyclocentric feature of the bicycle. The child with the lowest level of physical function (8 years of age, GMFCS level III, lowest baseline GMFM-66 374

f

Physical Therapy

Volume 90

score⫽47.5) did not develop this ability. In contrast, the child with the highest level of function (17 years of age, GMFCS level I, highest GMFM-66 score⫽100) reached the maximum load capability (100 lb) of the bicycle during the first session. Twelve additional participants reached the maximum load later in the intervention (sessions 9 –30). For these participants, resistance was provided or enhanced using the constant power mode resistance feature described for phase 2. The average maximum load, over the first 3 days of the intervention, was 26.9 lb (SD⫽26.6, 30% of body weight). The maximum load reached by the end of the intervention was 65.5 lb (SD⫽34.2, 74% of body weight). The average gain was 38.6 lb (SD⫽25.7, range⫽0 –90). For phase 2, initial cycling ability of the participants was variable. Some children cycled independently at a

Number 3

high rate with minimal cueing, achieving HRs between 70% and 80% of HRmax within the first session. Other children required considerable verbal cueing, adaptations, and physical assistance to complete a single cycling revolution. Alignment of the limb such that the knee was either medial or lateral to the cycling plane was a common problem requiring correction. Regardless, all children were able to cycle independently by the end of the intervention. The primary strategy used to intensify physical effort was increased cycling cadence, rather than increased resistance, as a majority of children could not maintain cycling speed when resistance was increased. Mean TEHR across all sessions was 147.2 bpm (SD⫽14.4, range⫽117–176), representing a mean of 52.2% of HRmax (SD⫽12.2%, range⫽8%–77%). Mean TEHR exceeded 50% of HRmax for the majority of the participants. Only one child had an average HR below 30% of HRmax. Outcomes Walking and running endurance, preferred walking speed, and GMFM-66 results are presented in Table 3. As we anticipated that some children would not be able to complete the 600-Yard Walk-Run Test, speed for the distance completed, rather than time, was the outcome measure for the test. At baseline, all but 6 of the 62 participants tested were able to complete the 600-Yard Walk-Run Test within the 15-minute time limit. Five children were GMFCS level III, and 1 child was level II. The speed of 1 child, at GMFCS level I, was within normal values for this test28 (baseline speed⫽240.3 m/min). A significant baselinepostintervention improvement of 5.6 m/min (P⫽.008) was found for the 600-Yard Walk-Run Test for the cycling group but not for the control group (Tab. 3). Preferred walking speed did not change significantly March 2010

Pediatric Endurance and Limb Strengthening within either group based on the 30sWT. A significant baselinepostintervention improvement was found for the GMFM-66 within the cycling group but not the control group. Specific test items that demonstrated the most improvement were unilateral standing with arms free, attaining a squat position from standing, stepping over a stick at knee level, running 4.5 m, jumping 30 cm high with both feet simultaneously, and walking up stairs alternating feet. Significant differences were not found between change scores for the cycling and control groups. Higher peak moments were found for the knee extensors than for the knee flexors (Tab. 4). Isometric (0°/s) values were highest. As speed of concentric muscle contraction increased, peak moments decreased. The cycling group showed a significant baseline-postintervention improvement in peak knee flexor moments at 30°/s (P⫽.025) and in knee extensor moments at 120°/s (P⫽ .006). At 120°/s, the number of limbs that could generate recordable knee flexor moments bilaterally decreased substantially. Fifteen participants (8 in the cycling group, 7 in the control group) were not able to generate measurable knee flexor joint moments with either limb at this speed. For the subset of participants who could produce measurable knee flexor joint moments at this speed, a significant increase was found for the control group (P⫽.01) but not the cycling group (P⫽.09). Significant differences were not found between change scores for the cycling and control groups for strength. Significant differences were not found in baseline-postintervention gait kinematics within either group or between groups (P⬎.05, data not shown).

Table 4. Knee Extensor and Flexor Moments Normalized to Body Weighta Variable

Cycling Group

Control Group

n⫽26c

n⫽26c

Pb

Knee extensor moments (N䡠m/kg) 0°/s Baseline

1.24 (1.04 to 1.45)

1.14 (1.0 to 1.28)

Postintervention

1.25 (1.10 to 1.41)

1.19 (1.02 to 1.36)

0.01 (⫺0.11 to 0.12)

0.05 (⫺0.04 to 0.14)

.88

.25

Changed P 30°/s Baseline Postintervention

n⫽29

n⫽29

1.05 (0.91 to 1.19)

1.09 (0.91 to 1.27)

1.09 (0.95 to 1.22)

1.01 (0.83 to 1.19)

0.04 (⫺0.05 to 0.12)

⫺0.08 (⫺0.19 to 0.03)

P

.39

.13

60°/s

n⫽28

n⫽29

Baseline

0.88 (0.76 to 0.99)

0.88 (0.72 to 1.05)

Postintervention

0.89 (0.76 to 1.0)

0.86 (0.69 to 1.04)

0.01(⫺0.06 to 0.09)

⫺0.02 (⫺0.09 to 0.06)

.76

.63

Change

Change P 120°/s

n⫽26

n⫽26

Baseline

0.66 (0.57 to 0.75)

0.72 (0.60 to 0.84)

Postintervention

0.75 (0.64 to 0.85)

0.75 (0.59 to 0.92)

Change

0.09 (0.03 to 0.15)

0.03 (⫺0.05 to 0.12)

P

.006

e

.55

.08

.58

.27

.45

Knee flexor moments (N䡠m/kg) 0°/s Baseline Postintervention

n⫽24c

n⫽25c

0.46 (0.36 to 0.57)

0.40 (0.26 to 0.54)

0.47 (0.36 to 0.58)

0.45 (0.32 to 0.58)

0.01 (⫺0.06 to 0.08)

0.05 (⫺0.01 to 0.11)

P

.69

.11

30°/s

n⫽28

n⫽26

Change

Baseline

0.30 (0.23 to 0.37)

0.34 (0.23 to 0.44)

Postintervention

0.35 (0.27 to 0.42)

0.35 (0.24 to 0.46)

Change

0.05 (0.01 to 0.09)

0.01 (⫺0.04 to 0.07)

.025e

.57

P 60°/s Baseline Postintervention Change

n⫽25

n⫽27

0.29 (0.22 to 0.36)

0.28 (0.19 to 0.36)

0.29 (0.21 to 0.36)

0.27 (0.18 to 0.37)

0.00 (⫺0.06 to 0.06)

⫺0.01(⫺0.04 to 0.04)

P

.95

.94

120°/s

n⫽21

n⫽22

Baseline

0.21 (0.16 to 0.26)

0.20 (0.13 to 0.28)

Postintervention

0.26 (0.19 to 0.32)

0.28 (0.17 to 0.38)

0.04 (⫺0.01 to 0.10)

0.08 (0.02 to 0.12)

.09

.01e

Change P

.41

.31

.99

.43

a

Values are mean (95% confidence intervals). P value for between-group comparisons using independent t tests. 3/29 participants did not undergo isometric testing, as it was added after the start of the study. d Postintervention change calculated by subtracting baseline value from postsession value. e P⬍.05 for baseline-postintervention comparison using paired t tests. b

Discussion We were unable to demonstrate that stationary cycling was better than no March 2010

c

Volume 90

Number 3

Physical Therapy f

375

Pediatric Endurance and Limb Strengthening intervention; however, significant baseline-postintervention improvements within the cycling group provided preliminary support for cycling in this phase I study. Other recent RCTs that examined exercise interventions in children with CP did not find between-group differences for some29,30 or all31,32 outcome measures. Using this design, betweengroup statistical significance can be most easily detected when: (1) intersubject and intrasubject variability is minimal, (2) control group outcomes are stable, and (3) there is a large treatment effect. If these factors are not optimal, the inclusion of larger number of participants may be required to obtain sufficient power to realize between-group effects, increasing the expense and effort required for research. In examining our primary outcome data, there were moderate effect sizes for the 600-Yard Walk-Run (0.33) and the GMFM-66 (0.38). In order to show a statistically significant difference between the 2 groups for these effect sizes, 130 participants in each group (a 2-sided test at the .05 level with 80% power) would be required. Intersubject variability between the 2 groups was addressed by random assignment and blocking by selective motor control. A significant correlation between selective voluntary motor control and GMFCS levels33 has been reported (Spearman r⫽⫺.83, P⬍.001), but other measures of CP severity, such as balance and spasticity (a velocity-dependent hyperexcitability of the muscle stretch reflex), may have been more disproportionate. Intrasubject variability is more difficult to anticipate and was high, as evidenced by the large confidence intervals observed for the change scores. Consistent performance during outcome evaluations may have been challenging for participants with comorbidities, including asthma and intellectual, behavioral, and visual deficits. Additionally, 376

f

Physical Therapy

Volume 90

there is evidence that children with CP have increased sensitivity to environmental alterations during testing. Heat consistent with a warm climate was found to increase metabolic rate and body temperature in children with CP (GMFCS levels I and II) during treadmill exercise but not in children without disability.34 Biological factors such as mood, comfort, and amount of rest can differ between testing sessions. Although it is difficult to control all factors, we informed families about the testing procedures, performed repeat testing at the same time of day, and ensured that each child had adequate nutrition and rest during data collection. Clinical trials for other pediatric populations who exhibit both physical and intellectual disabilities have addressed intrasubject variability by conducting multiple baseline sessions and either averaging measures35 or excluding participants who exhibit excessive performance variability from further testing.36 Future RCTS for children with CP might identify intrasubject variability through the performance of at least 2 baseline evaluations prior to the beginning of the intervention. Once identified, statistical analyses could be performed to determine the effect of these data on the results. The control group in the present study was not exposed to the cycling intervention, yet their mean baseline-postintervention scores increased for walking and running tests, the GMFM-66, and 5 out of 8 isokinetic tests. Mean improvements often were associated with fairly low P values (Tabs. 3 and 4). Statistically significant improvement (P⬍.05) was limited to peak knee flexor moments at 120°/s. That significance was reached for this one outcome measurement is likely a spurious result. The change in the cycling group was not significant (P⫽.09). Possible

Number 3

reasons for the control group’s improvements are: (1) accommodation, (2) practice, (3) physical activity level, (4) attention, and (5) a desire to compete with the first performance. Accommodation occurs when a participant changes his or her performance after becoming comfortable with testing conditions. In the present study, participants were tested by physical therapists previously unknown to them, in unfamiliar environments, and were asked to perform novel, physically challenging activities. Becoming comfortable with the environment and testing procedures may have positively affected postintervention measurements for some participants. Andersson et al37 addressed this issue in their study of Six-Minute Walk Test reliability in adults with CP. As significant improvement was found between test 1 and test 2 but not between tests 2, 3, and 4, a practice test was recommended when using this test for intervention studies. In our test of locomotor endurance, 68% of the control group participants improved in the 600-Yard Walk-Run Test, and 39% improved by more than 10%. The effect of accommodation and practice in the present study could have been examined with 2 baseline assessment sessions. Self-report measures of physical activity were similar between the 2 groups. Activity calendars indicated a similar frequency for sports and play activities, although fewer days at high-level, as opposed to moderate-level, categories were reported for the control group. The use of more quantitative measures, such as accelerometers,1 may better characterize physical activity and prevent the possibility of underreporting or overreporting. Participants agreed to maintain their present level of exercise, sports, and physical therapy during the study. Their attitude toward intensity, howMarch 2010

Pediatric Endurance and Limb Strengthening ever, may have been affected by information detailing the potential benefits of exercise that was provided during the recruitment and consenting process. Finally, considerable positive attention and anticipation of receiving an adapted bicycle may have influenced the effort of control group participants during the postintervention session. A desire to exceed their baseline performance was expressed by several participants during postintervention testing. A large treatment response in the intervention group is desired for RCTs examining exercise interventions in CP. The response to therapeutic interventions in children, however, is complex and likely influenced by factors such as impairment, inherent characteristics of the child, and family dynamics.38 Motivation and comorbidities, particularly intellectual, attentional, and behavioral problems, can limit the child’s ability to fully engage in the intervention and tolerate the feeling of physical effort associated with intense exercise. In pediatrics, parents typically initiate physical therapy, and the child may not be inherently motivated to exercise. Motivation techniques were individualized in this study using music, verbal praise, cheering, or rewards, and, overall, participants appeared to be engaged and motivated. A diagnosis of mental retardation was reported for 2 cycling group participants who elected to withdraw from the study. Two other children with this diagnosis, however, successfully completed the intervention. As there is a high prevalence of intellectual and other comorbidities in children with CP, their effect on choice of optimal treatment interventions warrants further study. A walking and running intervention would have been more specific to our goal of improving walking and March 2010

running endurance; however, balance, coordination, and selective motor control deficits are factors that made cycling more desirable. The cycling intervention was consistent with specificity of training principles due to its effect on the energy systems used.39 Prolonged stimulation of the cardiorespiratory system results in changes in heart, vascular, and blood function. These adaptations can improve performance in all types of endurance activities. Intensity threshold is the level of exercise that must be obtained to discern a training effect,40 but intensity is rarely described for physical therapy interventions in children with CP.5 The exercise intensity for the present study appeared sufficient to improve walking and running endurance, gross motor function, and a subset of strength measurements. The cyclocentric method of strengthening, with the addition of constant resistance features for stronger participants, allowed the progression of intensity throughout the intervention duration, for an average gain in resistance that approximated 54% of body weight. For cardiorespiratory training, an intensity threshold of 70% to 80% of HRmax is recommended for young adults and is at least that high for children.40 Although the average TEHR for the cycling group did not reach this threshold, improved walking and running endurance indicates a training effect occurred. The intensity threshold for children with CP may be below 70% of HRmax due to the reduced peak aerobic capacity reported for this population.41 The TEHR findings were consistent with those of Darrah et al,14 who reported that most adolescents with CP attained HRs above 145 bpm during an aerobic dance intervention. Another consideration is that HRmax may not equal 220 ⫺ age for this population. There is controversy

over the estimation of HRmax using the formula HRmax ⫽ 220 ⫺ age, which was not based on original research.42 Despite the widespread acceptance of this formula, research has revealed a large standard error of estimate (Sxy⫽7–11 bpm). Actual HRmax is the HR that cannot be surpassed despite continued increases in exercise intensity—a challenge for children with CP and, therefore, a limitation of our study. Training duration is another important consideration. Verschuren et al16 demonstrated between-group differences in an RCT with a longer study duration (8 months) than that of our study. Improvements in the exercise group over this time period were accompanied by a decline in most measures by the control group, enhancing the detection of betweengroup differences. Mean anaerobic and aerobic performance increased in the control group over the first 4 months but declined to below baseline values by the end of the study by Verschuren and colleagues. Their findings suggest that interventions exceeding 4 months may be required to discern a treatment effect in children with CP. A large treatment response may occur but not be identified without sensitive outcome measurements. We chose outcome measures that were specific for cardiorespiratory endurance and strength but differed from the intervention task to avoid practice of the test procedures. Two other RCTs29,30 did not find betweengroup differences in their 6-week interventions using outcome measures that differed from the intervention task. In contrast, Verschuren et al16 demonstrated between-group differences in an RCT with training exercises that appeared similar to some of the outcome assessments. Isokinetic strength testing in the present study may not have fully captured the training effects. Although this is

Volume 90

Number 3

Physical Therapy f

377

Pediatric Endurance and Limb Strengthening the standard method for strength measurement throughout a range of speeds, it requires isolated joint movement, which is problematic because children with CP have impaired selective voluntary motor control. Strength assessment of full limb extension and flexion across a range of speeds might be preferable; however, a standardized, objective method was not available. Previous research has shown that children with CP have increasingly greater strength deficits as speed increases from 0 to 120°/s compared with those without impairment.43 In the present study, we found that fewer participants could generate recordable joint moments for one or both limbs at the highest speed, particularly for knee flexors. Other researchers have noted that children with CP have difficulty moving at sufficient speeds for isokinetic testing.44,45 In one study,45 4 of 12 children with CP were considered “too weak” to participate in isokinetic testing. Improvements in peak knee joint moments for the cycling group were consistent with training speeds and the specificity of training principles for skeletal muscle.39 In skeletal muscle, different motor unit types are recruited in response to alteration in intensity and duration of load and stimulus. Adaptation is specific to joint actions, specific muscle groups recruited, and the velocity of contraction. The inclusion of strength testing across a range of speeds proved important, as we did not find changes with isometric testing. Previous research has shown that improvements in peak joint moments are specific to the training velocity.46 College-aged students who trained at 60°/s demonstrated significantly greater peak moments (P⬍.05) than a placebo group at this speed. Highspeed training (300°/s) resulted in a significant effect at 180°/s but not at slower speeds (0 or 60°/s). Cycling 378

f

Physical Therapy

Volume 90

during the PEDALS strengthening phase was performed at relatively slow speeds, most comparable to isokinetic testing at 30°/s. We found selective improvement of the cycling group knee flexors at this speed. The postintervention hamstring muscle to quadriceps muscle peak torque ratio for the cycling group (32%) was well below normative data for children at 30 to 60°/s (⬎60%)47; therefore, increased strength in this typically spastic muscle group was not a concern. Johnston et al48 reported a greater duration of hamstring muscle activity during recumbent stationary cycling for adolescents with CP compared with adolescents without disability. Hamstring muscle recruitment duration and intensity during cycling may have exceeded recruitment demand during daily activities for children with CP, who often have a crouch gait, walk slowly, or do not participate in sports. Cycling speed during the cardiorespiratory phase was most comparable to isokinetic testing at 120°/s. Improved peak knee extensor strength for the cycling group was found at this speed. Higher cycling cadences, typically used during the cardiorespiratory phase, may have challenged the knee extensors beyond speeds typically encountered during normal activities. Participants chose to maintain the lowest level of resistance and increased cycling cadence to elevate their HR. The alternative strategy of increasing resistance proved too strenuous for most children. Research examining children without disability showed that lower loads were optimal for obtaining peak power output during cycling, as higher loads induced fatigue.49 Improved GMFM-66 scores found for the cycling group support greater functional strength. The 1.2-point

Number 3

gain observed for the cycling group was between medium (0.8) and large (1.3) effect sizes, corresponding to minimum clinically important differences for ambulatory children with CP (GMFCS levels I, II, and III).50 Williams and Pountney19 reported a larger increase in mean GMFM-66 of 3 points for 11 children with dyskinetic or spastic CP following a cycling intervention. These participants had GMCFS levels of IV or V and lower gross motor function (mean GMFM-66 score⫽39.2) at baseline. The investigators attributed this substantial gain to a lack of other opportunities for physical activity. Participants in the present study were ambulatory with higher baseline functional ability (mean GMFM-66 scores⫽69.6 and 68.8 for the cycling and control groups, respectively) and, therefore, had more opportunities for physical activity. Lack of improvement in preferred walking speed suggests either that the intervention was not task specific for this parameter or that it is an innate behavior that is somewhat impervious to change in children with CP. Improved preferred walking speeds have not been a consistent finding following progressive resistive exercise programs for individuals with CP; some authors reported significant improvements,51–53 whereas other authors did not.22,54 Our findings are similar to those of Sullivan and colleagues,26 who compared body-weight– supported treadmill training and stationary cycling, both combined with lower-extremity strength training, in adults poststroke. They found that, although all participants had improved walking endurance (on the Six-Minute Walk Test), those participants assigned to the cycling intervention did not have improved walking speed over a short distance (10 m). In the present study, gait improvements were not found for preferred speed or kinematics. March 2010

Pediatric Endurance and Limb Strengthening

Conclusions Stationary cycling is an exercise that can address impairments in both strength and endurance. It can be incorporated into physical therapy programs for children with CP and transitioned into independent physical activity at home or school or in the community. Although access to adaptive physical education, sports, and recreation varies, most children with CP are followed during their childhood by a physical therapist, who can include fitness exercise in therapy sessions and assist in the design of independent programs. The PEDALS intervention took place in community pediatric physical therapy settings. A specialized stationary bicycle was used in order to precisely document increases in load; however, alternative methods of increasing and quantifying resistance are standard features of most stationary bicycles. Training children to increase their HR, progressively increase resistance, and rate their feelings of exertion are skills that can empower them to become selfsufficient in lifelong fitness programs. At the conclusion of this study, all participants received adapted overground or stationary bicycles. It is hoped that long-term exercise will promote general health and prevent secondary conditions as children with CP age and mature. Questionnaires addressing healthrelated quality of life and participation were administered to all participants in this study and may provide additional information about factors that affect outcomes for this population. These results will be reported in a separate publication. There has been a recent growth in research examining strengthening and cardiorespiratory fitness interventions for children with CP. The current level of evidence supporting exercise programs is low, and results have been inconsistent, particularly March 2010

for RCTs. This phase I study provides information to help guide future research. As considerable diversity of personal and environmental factors exists within this patient population, research must be carefully designed. Identification of participants who are inconsistent in their performance, consideration of accommodation and practice effects, sufficiently intense interventions, and selection of highly sensitive outcome measures may improve the detection of between-group differences in future RCTs. The results of this study stress the importance of including a control group to examine potential improvements that are not due to the intervention. Otherwise, the evidence supporting interventions may be overstated. We failed to prove that the intervention was better than no intervention using an RCT design, but the results for the cycling group were promising and offer guidance for future research. The benefits of exercise for health and well-being are well established in the general population. Individuals with CP should be provided with exercise protocols that maximize health, promote functional improvement, and minimize secondary conditions. Dr Fowler, Dr Knutson, Dr DeMuth, and Dr Azen provided concept/idea/research design. All authors provided writing. Dr Fowler, Dr Knutson, Dr DeMuth, Dr Siebert, Ms Simms, Dr Sugi, and Dr Souza provided data collection. Dr Fowler, Dr Knutson, Dr DeMuth, Dr Souza, Dr Karim, and Dr Azen provided data analysis. Dr Fowler, Dr Knutson, Dr DeMuth, and Dr Siebert provided project management. Dr Fowler, Dr Knutson, and Dr DeMuth provided fund procurement. Dr Fowler and Dr Knutson provided participants and institutional liaisons. Dr Fowler, Dr Knutson, Dr DeMuth, and Dr Souza provided facilities/equipment. Dr Azen provided consultation (including review of manuscript before submission). The authors acknowledge the following individuals, facilities, foundations, and corporations for their contributions: Intervention facilities: Meyer Center at CoxHealth, Orthopaedic Hospital, Pediatric Therapy Network,

Phelps County Community Center, St. John’s Lebanon, The Children’s Therapy Center, and Therapy West. Intervention therapists: Vasti Blake, PTA, Tanjay Castro, PT, DPT, Kara Crockett, PT, Nancy Egizii, PT, MPT, Noel Marie Enriquez, PT, PCS, David Europongpan, PT, DPT, Amanda Glendenning, PT, Dianne E. Jones, PT, MEd, Betsy King, PT, DPT, Jean Knapp, PT, PCS, Rennie T. Lee, PT, Barbara Lopetinsky, PT, Nisha Pagan, MPT, NCS, Tracy Phenix, PT, Deborah Rothman, PT, MSPT, Susan Rouleau, PT, Christy Skura, PT, DPT, Josie Stickles, PT, DPT, Margaretha Van Gool, PT, PCS, and Julie Yang, PT, DPT. Outcome data collection personnel and facilities: George Salem, PhD, Matt Sandusky, Rich Souza, PT, MPT, PhD, Albert Vallejo, PhD, and Francisco Bravo Medical Magnet High School. Data personnel: Cariza Alvarez, Frances Chien, Carolyn Ervin, PhD, Kyle Fink, Evan Goldberg, MS, Chris Hahn, Michelle Hudson, PT, MPT, Karina Kunder, Serge Modoyan, Sarah Mohajeri, Terence Padden BS, Rakhista Satyarthi, and Wenli Wang, MS. Interpreters: Nena Becerra, Karla Cordova, PT, DPT, Lidia Corte´s, Carmen Diaz, Wil Diaz, PT, DPT, Minchul Jung, Linda Kang, Kevin Lee, PT, DPT, Raul Lona, PT, DPT, Irene Morado, Susumu Ota, PT, Janelle Rodriquez, Pietro Scaglioni-Solano, MS, and Jooeun Song, MS. Recruitment: Los Angeles and Orange County California Children’s Services (Eric Lingren, PT, MPT, Lisa Mena, PT, MPT). Corporate donations or discounts: Biodex Inc, Freedom Concepts, Helen’s Cycles, Santa Monica, National AMBUCS Inc, and Sam’s Club. Volunteers and foundations: Caitlin Fowler, Ernie Meadows, Sidney Stern Memorial Trust, Steinmetz Foundation, Sykes Family Foundation, and United Cerebral Palsy Research and Education Foundation. Physical Therapy Clinical Research Network (PTClinResNet): Network Principal Investigator is Carolee J. Winstein, PT, PhD, FAPTA, and Co-Principal Investigator is James Gordon, PT, EdD, FAPTA (both at University of Southern California). Project Principal and Co-Principal Investigators are: David A. Brown, PT, PhD (Northwestern University); Sara Mulroy, PT, PhD, and Bryan Kemp, PhD (Rancho Los Amigos National Rehabilitation Center); Loretta M. Knutson, PT, PhD, PCS (University of Indianapolis); Eileen G. Fowler, PT, PhD (University of California at Los Angeles); and Sharon K. DeMuth, PT, DPT, Kornelia Kulig, PT, PhD, and Katherine J. Sullivan, PT, PhD (University Southern California). The Data Management Center is located at the University of Southern California and is directed by Stanley P. Azen, PhD. The 4 members of the Data Safety and Monitoring Committee are: Nancy Byl, PT, PhD, FAPTA, Chair (University of California at San

Volume 90

Number 3

Physical Therapy f

379

Pediatric Endurance and Limb Strengthening Francisco), Hugh G. Watts, MD (Shriners’ Hospital for Children–LA Unit, Los Angeles, California), June Isaacson Kailes, MSW (Western University, Los Angeles, California), and Anny Xiang, PhD (University of Southern California). This research was presented at the Margaret Jones Annual Conference on Cerebral Palsy; May 5, 2007; University of California at Los Angeles, Los Angeles, California; the Combined Sections Meeting of the American Physical Therapy Association; February 9 –12, 2009; Las Vegas, Nevada; and the 62nd Annual Meeting of the American Academy for Cerebral Palsy and Developmental Medicine; September 17–20, 2009; Atlanta, Georgia. Other presentations focused on protocol development and design. The presentations currently listed included the results of the primary outcome measures, the focus of this article. The institutional review boards of the University of Southern California, the University of California at Los Angeles, the State of California–Health and Human Services Agency, and Missouri State University approved the study protocol and consent procedures. This study was supported by a grant from the Foundation for Physical Therapy to establish PTClinResNet, clinical research network to evaluate the efficacy of physical therapist practice. Clinical Trials.gov Identifier: NCT00401154. This article was received November 14, 2008, and was accepted September 11, 2009. DOI: 10.2522/ptj.20080364

380

f

Physical Therapy

Volume 90

References 1 Bjornson KF, Belza B, Kartin D, et al. Ambulatory physical activity performance in youth with cerebral palsy and youth who are developing typically. Phys Ther. 2007; 87:248 –257. 2 Pirpiris M, Graham HK. Uptime in children with cerebral palsy. J Pediatr Orthop. 2004;24:521–528. 3 van den Berg-Emons HJ, Saris WH, de Barbanson DC, et al. Daily physical activity of schoolchildren with spastic diplegia and of healthy control subjects. J Pediatr. 1995;127:578 –584. 4 Durstine JL, Painter P, Franklin BA, et al. Physical activity for the chronically ill and disabled. Sports Med. 2000;30:207–219. 5 Fowler EG, Kolobe TH, Damiano DL, et al. Promotion of physical fitness and prevention of secondary conditions for children with cerebral palsy: Section on Pediatrics Research Summit Proceedings. Phys Ther. 2007;87:1495–1510. 6 Rimmer JH. Physical fitness levels of persons with cerebral palsy. Dev Med Child Neurol. 2001;43:208 –212. 7 Verschuren O, Ketelaar M, Takken T, et al. Exercise programs for children with cerebral palsy: a systematic review of the literature. Am J Phys Med Rehabil. 2008;87: 404 – 417. 8 Dodd KJ, Taylor NF, Damiano DL. A systematic review of the effectiveness of strength-training programs for people with cerebral palsy. Arch Phys Med Rehabil. 2002;83:1157–1164. 9 Berg K. Effect of physical training of school children with cerebral palsy. Acta Paediatr Scand Suppl. 1970;204:27–33. 10 van den Berg-Emons RJ, Van Baak MA, Speth L, Saris WH. Physical training of school children with spastic cerebral palsy: effects on daily activity, fat mass and fitness. Int J Rehabil Res. 1998;21:179 – 194. 11 Shinohara TA, Suzuki N, Oba M, et al. Effect of exercise at the AT point for children with cerebral palsy. Bull Hosp Jt Dis. 2002;61:63– 67. 12 Schlough K, Nawoczenski D, Case LE, et al. The effects of aerobic exercise on endurance, strength, function and selfperception in adolescents with spastic cerebral palsy: a report of three case studies. Pediatr Phys Ther. 2005;17:234 –250. 13 Lundberg A, Ovenfors CO, Saltin B. Effect of physical training on school-children with cerebral palsy. Acta Paediatr Scand. 1967;56:182–188. 14 Darrah J, Wessel J, Nearingburg P, O’Connor M. Evaluation of a community fitness program for adolescents with cerebral palsy. Pediatr Phys Ther. 1999; 11:18 –23. 15 Unnithan VB, Katsimanis G, Evangelinou C, et al. Effect of strength and aerobic training in children with cerebral palsy. Med Sci Sports Exerc. 2007;39:1902–1909.

Number 3

16 Verschuren O, Ketelaar M, Gorter JW, et al. Exercise training program in children and adolescents with cerebral palsy: a randomized controlled trial. Arch Pediatr Adolesc Med. 2007;161:1075–1081. 17 Fernhall B, Pitetti KH, Vukovich MD, et al. Validation of cardiovascular fitness field tests in children with mental retardation. Am J Ment Retard. 1998;102:602– 612. 18 United Cerebral Palsy Research and Education Foundation. Exercise principles and guidelines for persons with cerebral palsy and neuromuscular disorders, 1999. Available at: http://www.ucp.org/ ucp_channeldoc.cfm/1/15/11500/11500– 115– 00/639. Accessed October 5, 2008. 19 Williams H, Pountney T. Effects of a static bicycling programme on the functional ability of young people with cerebral palsy who are non-ambulant. Dev Med Child Neurol. 2007;49:522–527. 20 Fowler EG, Knutson LM, DeMuth SK, et al. Pediatric endurance and limb strengthening for children with cerebral palsy (PEDALS): a randomized controlled trial protocol for a stationary cycling intervention. BMC Pediatr. 2007;7:14. 21 Palisano R, Rosenbaum P, Walter S, et al. Development and reliability of a system to classify gross motor function in children with cerebral palsy. Dev Med Child Neurol. 1997;39:214 –223. 22 MacPhail HE, Kramer JF. Effect of isokinetic strength-training on functional ability and walking efficiency in adolescents with cerebral palsy. Dev Med Child Neurol. 1995;37:763–775. 23 World Health Organization. International Classification of Functioning, Disability and Health, Report by the Secretariat, Fifty-fourth World Health Assembly, Provisional Agenda Item 13.9, April 9, 2001. Geneva, Switzerland, World Health Organization; 2001. 24 Knutson LM, Schimmel PA, Ruff A. Standard task measurement for mobility: thirtysecond walk test. Pediatr Phys Ther. 1999;11:183–190. 25 Russell DJ, Rosenbaum PL, Avery LM, Lane M. Gross Motor Function Measure (GMFM-66 & GMFM-88) User’s Manual. London, United Kingdom: Mac Keith Press; 2002. 26 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. 27 Williams JG, Eston R, Furlong B. CERT: a perceived exertion scale for young children. Percept Mot Skills. 1994;79:1451– 1458. 28 Vodak PA, Wilmore JH. Validity of the 6-minute jog-walk and the 600-yard runwalk in estimating endurance capacity in boys, 9 –12 years of age. Res Q. 1975;46: 230 –234. 29 Dodd KJ, Taylor NF, Graham HK. A randomized clinical trial of strength training in young people with cerebral palsy. Dev Med Child Neurol. 2003;45:652– 657.

March 2010

Pediatric Endurance and Limb Strengthening 30 Liao HF, Liu YC, Liu WY, Lin YT. Effectiveness of loaded sit-to-stand resistance exercise for children with mild spastic diplegia: a randomized clinical trial. Arch Phys Med Rehabil. 2007;88:25–31. 31 Patikas D, Wolf SI, Armbrust P, et al. Effects of a postoperative resistive exercise program on the knee extension and flexion torque in children with cerebral palsy: a randomized clinical trial. Arch Phys Med Rehabil. 2006;87:1161–1169. 32 Patikas D, Wolf SI, Mund K, et al. Effects of a postoperative strength-training program on the walking ability of children with cerebral palsy: a randomized controlled trial. Arch Phys Med Rehabil. 2006;87: 619 – 626. 33 Fowler EG, Staudt LA, Greenberg MB, Oppenheim WL. Selective Control Assessment of the Lower Extremity (SCALE): development, validation, and interrater reliability of a clinical tool for patients with cerebral palsy. Dev Med Child Neurol. 2009;51:607– 614. 34 Maltais D, Wilk B, Unnithan V, Bar-Or O. Responses of children with cerebral palsy to treadmill walking exercise in the heat. Med Sci Sports Exerc. 2004;36: 1674 –1681. 35 Skura CL, Fowler EG, Wetzel GT, et al. Albuterol increases lean body mass in ambulatory boys with Duchenne or Becker muscular dystrophy. Neurology. 2008;70: 137–143. 36 Escolar DM, Buyse G, Henricson E, et al. CINRG randomized controlled trial of creatine and glutamine in Duchenne muscular dystrophy. Ann Neurol. 2005;58: 151–155. 37 Andersson C, Asztalos L, Mattsson E. Sixminute walk test in adults with cerebral palsy: a study of reliability. Clin Rehabil. 2006;20:488 – 495.

March 2010

38 Bartlett DJ, Palisano RJ. A multivariate model of determinants of motor change for children with cerebral palsy. Phys Ther. 2000;80:598 – 614. 39 Wenger HA, McFayden PF, McFayden RA. Physiological principles of conditioning. In: Zachazewski JE, Magee DJ, Quillen WS, eds. Athletic Injuries and Rehabilitation. Philadelphia, Pa: WB Saunders Co; 1996: 189 –205. 40 Bar-Or O, Rowland, TW. Pediatric Exercise Medicine: From Physiologic Principles to Health Care Applications. Champaign, IL: Human Kinetics; 2004. 41 Hoofwijk M, Unnithan V, Bar-Or O. Maximal treadmill performance in children with cerebral palsy. Pediatr Exerc Sci. 1995;7:305–313. 42 Robergs RA, Landwehr R. The surprising history of the “HRmax⫽220-age” equation. Journal of Exercise Physiology Online. 2002;5:1–10. 43 Damiano DL, Martellotta TL, Quinlivan JM, Abel MF. Deficits in eccentric versus concentric torque in children with spastic cerebral palsy. Med Sci Sports Exerc. 2001; 33:117–122. 44 Ross SA, Engsberg JR. Relation between spasticity and strength in individuals with spastic diplegic cerebral palsy. Dev Med Child Neurol. 2002;44:148 –157. 45 van den Berg-Emons RJ, van Baak MA, de Barbanson DC, et al. Reliability of tests to determine peak aerobic power, anaerobic power and isokinetic muscle strength in children with spastic cerebral palsy. Dev Med Child Neurol. 1996;38:1117–1125. 46 Coyle EF, Feiring DC, Rotkis TC, et al. Specificity of power improvements through slow and fast isokinetic training. J Appl Physiol. 1981;51:1437–1442.

47 De Ste Croix MB, Armstrong N, Welsman JR, Sharpe P. Longitudinal changes in isokinetic leg strength in 10 –14-year-olds. Ann Hum Biol. 2002;29:5– 62. 48 Johnston TE, Barr AE, Lee SCK. Biomechanics of submaximal recument cycling in adolescents with and without cerebral palsy. Phys Ther. 2007;87:572–585. 49 Dore E, Bedu M, Franca NM, et al. Testing peak cycling performance: effects of braking force during growth. Med Sci Sports Exerc. 2000;32:493– 498. 50 Oeffinger D, Bagley A, Rogers S, et al. Outcome tools used for ambulatory children with cerebral palsy: responsiveness and minimum clinically important differences. Dev Med Child Neurol. 2008;50:918 –925. 51 Damiano DL, Abel MF. Functional outcomes of strength training in spastic cerebral palsy. Arch Phys Med Rehabil. 1998; 79:119 –125. 52 Eagleton M, Iams A, McDowell J, et al. The effects of strength training on gait in adolescents with cerebral palsy. Pediatr Phys Ther. 2004;16:22–30. 53 Thorpe DE, Reilly MA, Case LE. The effects of an aquatic resistive exercise program on leg strength, balance, energy expenditure, functional mobility and selfperception in children and young adults with cerebral palsy. Aquatic Phys Ther. 2005;13:21–34. 54 Unger M, Faure M, Frieg A. Strength training in adolescent learners with cerebral palsy: a randomized controlled trial. Clin Rehabil. 2006;20:469 – 477.

Volume 90

Number 3

Physical Therapy f

381

Research Report Effect of Treadmill Training and Supramalleolar Orthosis Use on Motor Skill Development in Infants With Down Syndrome: A Randomized Clinical Trial Julia Looper, Dale A. Ulrich J. Looper, PT, PhD, is Assistant Professor, Physical Therapy Program, University of Puget Sound, 1500 N Warner St #1070, Tacoma, WA 98416-1070 (USA). Address all correspondence to Dr Looper at: [email protected]. D.A. Ulrich, PhD, is Professor, Center for Physical Activity and Pediatric Disabilities, School of Kinesiology, University of Michigan, Ann Arbor, Michigan. [Looper J, Ulrich A. Effect of treadmill training and supramalleolar orthosis use on motor skill development in infants with Down syndrome: a randomized controlled trial. Phys Ther. 2010;90: 382–390.] © 2010 American Physical Therapy Association

Background. Children with Down syndrome (DS) often display delayed onset of independent walking. Treadmill training is an effective intervention that leads to an earlier walking onset. In addition, orthoses often are provided to infants with DS to increase stability and promote earlier independent walking. However, this early use of orthoses has not been scientifically verified in infants with DS. Objective. The purpose of this study was to provide insight into the developmental outcomes of early orthosis use in combination with treadmill training in infants with DS compared with treadmill training alone.

Design. This study was a randomized controlled trial. Setting. This study was conducted in participants’ homes and in the motor development laboratory.

Participants and Intervention. Seventeen infants with DS entered the study when they could pull themselves to a standing position. They were randomly assigned to either a control group (which received treadmill training) or an experimental group (which received treadmill training and orthoses). During monthly visits to the infants’ homes, 3 minutes of treadmill stepping was recorded and each child’s motor development skills were tested. The treadmill training ended once the child took 3 independent steps. One month following walking onset, developmental tests were readministered. Measurements. The Gross Motor Function Measure (GMFM) was used to test motor skill development. Results. The average (SD) time in the study was 268 (88) days for the control group and 206 (109) days for the experimental group. All infants showed significantly increased GMFM scores over time. At 1 month of walking experience, the control group had higher GMFM scores than the experimental group, with higher standing and walking, running, and jumping subscale scores.

Limitations. Limitations of this study included a small sample of convenience, a statistical model that may have reduced validity at the tail end, and a lack of blinding in the GMFM scorer. Post a Rapid Response or find The Bottom Line: www.ptjournal.org 382

f

Physical Therapy

Conclusions. Orthoses may have a detrimental effect on overall gross motor skill development.

Volume 90

Number 3

March 2010

Motor Skill Development in Infants With Down Syndrome

D

own syndrome (DS) is a genetic disorder that occurs in 13.65 out of every 10,000 live births.1 Though DS is most commonly known for its effects on cognitive ability, children with DS also have delayed and atypical motor development. On average, children with DS sit independently at 11 months, pull themselves to a standing position at 17 months, and walk independently at 24 months.2 These delays, separately and in combination, lead to difficulties with function and social interaction.

an increase in walking speed during the stance phase of gait. Martin6 studied orthosis use in 3- to 8-yearold children with DS. She found that the children showed immediate and long-term improvements in postural stability, based on standardized test scores, including Gross Motor Function Measure (GMFM) scores, when they used a flexible SMO. These studies showed that orthoses are an effective intervention for children with DS; however, little is known about the effects of orthoses on prewalkers with DS.

Difficulties with functional skills and social interaction also arise from atypical movement patterns. Infants with DS display atypical antigravity control in the legs and neck3; more specifically, they have atypical kicking patterns.4 Although these delayed and atypical movements are the focus of physical therapy interventions in infants with DS, there is little evidence of the effectiveness of specific interventions in this population.

Walking is an important milestone. Like other motor skills, walking in children with DS is not only delayed, but also abnormal.7–11 Although infants with DS walk, on average, 1 year later than infants who are developing typically, they are able to step when supported on a treadmill at a much earlier age.12,13 About 13 months before they walk, infants with DS respond to the treadmill with alternating steps.14 This preference for alternating steps while being supported on a treadmill begins to emerge by 11 months, about the time when infants with DS begin to sit independently.15

One specific intervention that has been studied in children with DS is orthosis use. Lower-extremity orthoses, specifically foot orthoses and supramalleolar orthoses (SMOs), often are provided to young children with DS to improve functional gait. These orthoses are external devices that stabilize the subtalar joint, thus maintaining the calcaneus in an upright position. An upright calcaneus improves the bony alignment of the foot and ankle and influences postural and gait characteristics. SelbySilverstein and colleagues5 examined orthotic use in 3- to 6-year-old children with DS. They found that foot orthoses led to decreased ankle eversion while the children remained in a static standing position. In addition, there were a decrease in foot progression angle and variability in this angle, a change in the initial contact site from flat foot to heel-strike, and March 2010

Ulrich and colleagues13 trained this stepping behavior in infants with DS before they were able to walk independently. Children who received treadmill training as a supplement to physical therapy walked independently 101 days earlier than a control group of children with DS who did not receive the treadmill intervention but continued to receive physical therapy. Although there were no group differences in age at study entry, the average chronological age at walking onset was 23.9 months in the control group and 19.9 months in the intervention group. A subsequent treadmill intervention study showed that a higher-intensity, individualized training protocol led to improved outcomes. There were

improvements in gait parameters and a decrease in gait variability when compared with the original treadmill paradigm and with toddlers with DS who did not receive treadmill training.8,16 The higherintensity, individualized training also has been reported to have a positive impact on adaptive gait and physical activity.17,18 This study is an extension of the previous treadmill training research involving infants with DS. The intent was to incorporate orthosis use into the lower-intensity treadmill training protocol of Ulrich and colleagues13 to provide improved biomechanical alignment during stepping practice. Because orthoses only work when children are able to attain an upright standing position, this study did not begin until the infants could pull themselves to a standing position. By intervening with a combination of orthosis use and treadmill training, we expected to see larger improvements in activity level skills than with treadmill training alone, as well as an improvement in the onset of independent walking.

Method Participants Twenty-two infants with DS were enrolled in this study. The study was

Available With This Article at ptjournal.apta.org • eFigure 1: GMFM Crawling and Kneeling Scale Scores Over Time • eFigure 2: GMFM Standing Scale Scores Over Time • eFigure 3: GMFM Walking, Running, and Jumping Scale Scores Over Time • Audio Abstracts Podcast This article was published ahead of print on January 14, 2010, at ptjournal.apta.org.

Volume 90

Number 3

Physical Therapy f

383

Motor Skill Development in Infants With Down Syndrome ceived them 1 to 2 weeks later. They wore the SMOs for 8 hours a day, 5 days per week, for 2 to 3 weeks prior to the final developmental measurement in order to accommodate to the orthoses. Data Collection Infants participated in the study from the time they could pull themselves to a standing position until they had 1 month of independent walking experience. Once an infant could take 3 independent steps overground without support, the treadmill training stopped. One month later, each child came to the Motor Development Laboratory at the University of Michigan to participate in follow-up measurements. At the first visit to the family’s home, parents reviewed and signed an informed consent document. Next, the researcher (J.L.) taught parents how to hold the infants on the treadmill in a safe manner and how to operate the treadmill (Fig. 2). At each monthly visit, 3 minutes of treadmill training were videotaped, anthropometric measurements were taken, and developmental tests were administered.

Figure 1. Participant flowchart.

limited to infants with trisomy 21. Infants had to be able to pull themselves to a standing position independently but were excluded if they were able to cruise. Infants also were excluded if they had a history of other developmental disabilities, uncorrected visual or hearing impairment, previous use of orthoses, or an intolerance to the orthoses. The included infants were randomly assigned to either a control group (n⫽12) or an experimental group (n⫽10). Over time, 5 children in the control group dropped out of the study (Fig. 1).

Use of Orthoses The infants in the experimental group were measured for SMOs* on their first visit and received them before their second visit. They wore the SMOs for 8 hours a day, 5 days a week, for the duration of the study, including the period between walking onset and the follow-up testing. The infants in the control group were measured for SMOs when they could take 3 independent steps without support, which was when the treadmill training ended, and re-

* Surestep, 17530 Dugdale Dr, South Bend, IN 46635.

384

f

Physical Therapy

Volume 90

Number 3

Treadmill Training Infants in both groups received treadmill training that began when the infants entered the study and continued until they took 3 independent steps overground without support. Each family was provided with a small, motorized treadmill† on the first visit. The parents were instructed to turn the treadmill on to a speed of 0.2 m/s and hold their child under the arms on the treadmill for 8 minutes a day, 5 days a week, allowing the child to support as much of his or her weight as possible. Initially, the treadmill training was in 1-minute increments with a brief † Carlin’s Creations, 27366 Oak Dr, Sturgis, MI 49091.

March 2010

Motor Skill Development in Infants With Down Syndrome rest. Parents were encouraged to increase the training interval gradually as the infant demonstrated the ability to take more steps. Parents also were instructed to have their infant wear shoes (for the control group) or shoes and orthoses (for the experimental group) during the training. The 8 minutes of training did not have to be continuous. Parents recorded the amount of daily treadmill training in a logbook provided by the researcher. Developmental Test Data Developmental progress was measured 1 time per month in each child’s home while the children were receiving treadmill training and in the laboratory when they had 1 month of independent walking experience. During this testing, the infants were not wearing SMOs. All developmental tests were all administered by 1 researcher (J.L.) who was aware of the children’s group assignments. The GMFM was used to measure gross motor skill acquisition.19 The test was developed for children with cerebral palsy but has since been validated for children with DS.20 It measures gross motor skill level in 5 subsections: (1) lying and rolling, (2) sitting, (3) crawling and kneeling, (4) standing, and (5) walking, running, and jumping. Each item within the subsections is scored on a scale from 0 (“does not initiate”) to 4 (“completes”). Data Analysis Data were analyzed using SPSS, version 13.‡ Between-group comparisons at 1 time point were analyzed with a t test. Longitudinal data were analyzed using mixed linear models. In addition, effect size, using the Cohen D statistic, was calculated for between-group comparisons.21 An effect size of 0.5 indicates that the mean scores in each group are 0.5 ‡ SPSS Inc, 233 S Wacker Dr, Chicago, IL 60606.

March 2010

Figure 2. Infant and parent during treadmill training.

standard deviation apart. Effect sizes of 0.2, 0.5, and 0.8 are considered small, moderate, and large, respectively.

and walking, running, and jumping. The initial differences between the groups were controlled for in the statistical analysis by including them as covariates in the statistical models.

Results Study Sample Characteristics The participants’ physical characteristics at study onset are reported in Table 1. The only a priori group difference was birth length, but by study onset, height was no longer significantly different. At study onset, there were no group differences in age, anthropometric measures, or the amount of time the infants received physical therapy per week. There also were no differences between the groups in parental age or education levels, birth order, or number of siblings. In terms of developmental test scores, there were group differences at study entry (Tab. 2). The experimental group had a higher overall GMFM score that was made up of higher scores on the GMFM scales for crawling and kneeling; standing;

Although the stated treadmill training protocol was 8 minutes per day, 5 days a week, the participants in both groups performed the training an average (SD) of 6 (2) minutes per day, 5 days a week. In addition, the experimental group wore the SMOs an average (SD) of 6.25 (4) hours per day. Walking Onset The average (SD) time in study was 268 (88) days (8.9 [2.9] months) for the control group and 206 (109) days (6.9 [3.6] months) for the experimental group. We performed t tests to determine whether there were differences between the groups in age at walking onset and time in study. There was no significant difference between groups for time in study (P⫽.23, power⫽0.21). There was a moderate effect (Cohen D⫽0.63) in

Volume 90

Number 3

Physical Therapy f

385

Motor Skill Development in Infants With Down Syndrome Table 1. Characteristics of the Study Sample at Entry Control Group Characteristic

Mean (SD)

Corrected age at entry (d)

578 (188)

P .41

9.41 (1.39)

10.26 (0.61)

.11

Height (cm)

75.81 (7.93)

78.67 (2.74)

.30

Shank length (cm)

13.93 (1.50)

14.77 (1.43)

.26

Shank circumference (cm)

18.27 (0.95)

17.69 (1.39)

.35

Thigh length (cm)

15.00 (1.50)

15.88 (1.58)

.27

Thigh circumference (cm)

26.04 (1.95)

25.28 (1.62)

.39

Physical therapy (min/wk)

60 (85)

63 (75)

.94

Birth weight (kg)

2.79 (0.47)

3.18 (0.41)

.10

Birth height (in)a

47.09 (2.29)

50.37 (2.49)

.02b

2 (2)

Birth order Maternal age (y)

b

Mean (SD) 642 (121)

Weight (kg)

Number of siblings

a

Experimental Group

1 (1)

.23

2 (2)

2 (1)

.20

32 (5)

34 (6)

.43

Paternal age (y)

34 (5)

33 (7)

.87

Maternal education (y)

15 (3)

16 (1)

.36

Paternal education (y)

14 (3)

15 (2)

.65

Income (⫻$1,000)

80–100 (40)

60–80 (20)

.64

1 in⫽2.54 cm. Significant at P⬍.05.

favor of the experimental group for time in the study. Given the power for this test and the effect size, a larger sample size may produce significant results.21 Developmental Testing During Treadmill Training Linear mixed models were used to test for differences between groups and over time in the overall GMFM scores, as well as in scores on the

sitting; crawling and kneeling; standing; and walking, running, and jumping scales. The models included a term for the linear effects of time, the quadratic effects of time, group effect, and both possible group ⫻ time (linear or quadratic) interactions. All models except the one for the sitting scale also included a term for initial score to account for differences between the groups at study onset. The models also allowed time

to vary randomly by infant. Predicted GMFM scores, based on the estimates of the fixed effects (Fig. 3), and the raw data are shown in eFigures 1, 2, and 3 (available at ptjournal. apta.org). The estimates are shown in Table 3. These figures show an overall upward developmental trajectory in GMFM scores over time on the crawling and kneeling; standing; and walking, running, and jumping sections. However, on the crawling

Table 2. Developmental Level at Study Entry Control Group Measure GMFMa total

a b

Experimental Group

Mean (SD)

Mean (SD)

P

130.89 (7.10)

148.40 (9.08)

.01b

GMFM sitting

52.00 (4.24)

54.70 (4.06)

.20

GMFM crawling and kneeling

22.43 (5.29)

27.20 (4.02)

.05b

GMFM standing

4.43 (2.44)

9.80 (3.19)

.01b

GMFM walking, running, and jumping

1.00 (1.15)

5.70 (3.77)

.01b

GMFM⫽Gross Motor Function Measure. Significant at P⬍.05.

386

f

Physical Therapy

Volume 90

Number 3

March 2010

Motor Skill Development in Infants With Down Syndrome and kneeling scale and the standing scales, the developmental trajectory was different for the control and experimental groups. All children in the study improved their overall GMFM scores over time (P⬍.001). There also was a significant quadratic effect of time in the total GMFM score (P⬍.001), indicating that the developmental trajectory of gross motor skill acquisition has a significant nonlinear component. Although there was no significant group difference on the overall GMFM score, there was a significant linear time ⫻ group interaction for the overall GMFM score (P⫽.04), indicating that the 2 groups followed a different developmental trajectory. All children in the study improved their GMFM scores over time (sitting, P⬍.001; crawling and kneeling, P⬍.001; standing, P⬍.001; and walking, running, and jumping, P⬍.001). There was a quadratic component to this improvement for all scales (sitting, P⬍.001; crawling and kneeling, P⬍.001; standing, P⫽.03; and walking, running, and jumping, P⫽.01), indicating a nonlinear developmental process across all infants.

Figure 3. Predicted Gross Motor Function Measure subscale scores over time. The predicted scores are fitted values representing estimated fixed effects of initial score, group, time (linear), time (quadratic), and their significant group ⫻ time interactions.

Table 3. Estimates of Fixed Effects With 95% Confidence Intervals (CI)

Measure GMFMa total 95% CI for estimates GMFM crawling and kneeling 95% CI for estimates GMFM standing 95% CI for estimates GMFM walking, running, and jumping 95% CI for estimates a

Intercept

Onset Score

Birth Length

Time (Linear)

Time (Quadratic)

Control Group

⫺0.33

⫺1.82

Time (Linear) ⴛ Group

Time (Quadratic) ⴛ Group

26.83

0.86

⫺0.63

6.97

⫾39.64

⫾0.24

⫾2.09

⫾1.91

⫾0.15

⫾6.99

11.98

0.41

0.13

1.66

⫺0.08

⫺4.45

⫾16.94

⫾0.18

⫾0.86

⫾0.08

⫾3.12

⫾1.22

⫺2.98

0.95

0.16

0.89

⫺0.03

⫺0.35

Not Significant

Not Significant

⫾12.14

⫾0.20

⫾0.62

⫾0.61

⫾0.05

⫾2.27

⫺1.55

0.84

0.03

2.03

⫺0.09

Not Significant

Not Significant

⫾0.29

⫾0.71

⫾1.01

⫾0.12

⫾13.583

⫾0.9

0.003

Not Significant

1.99

Not Significant

⫺0.12 ⫾0.10

⫾2.24

GMFM⫽Gross Motor Function Measure.

March 2010

Volume 90

Number 3

Physical Therapy f

387

Motor Skill Development in Infants With Down Syndrome Table 4. Developmental Level at 1 Month of Walking Experience Control Group

Experimental Group

Mean (SD)

Mean (SD)

P

Effect Size

Power

195.67 (8.12)

183.78 (7.22)

.01b

1.55

0.14

GMFM crawling and kneeling

35.67 (0.82)

35.89 (1.17)

.69

0.22

0.07

GMFM standing

28.67 (4.50)

21.00 (5.12)

.01b

1.59

0.79

15.89 (2.80)

b

1.36

0.69

Variable GMFMa total

GMFM walking, running, and jumping a b

20.33 (3.72)

.02

GMFM⫽Gross Motor Function Measure. Significant at P⬍.05.

There also were significant linear time ⫻ group interactions (crawling and kneeling, P⫽.03; standing, P⫽.01) and quadratic time ⫻ group interactions (crawling and kneeling, P⫽.05; standing, P⫽.01). These significant interactions indicate that the 2 groups followed a different developmental progression from study entry through independent walking without support. Developmental Testing in New Walkers The developmental test scores at 1 month after walking onset are shown in Table 4. We performed t tests to determine whether the scores differed by group. Effect sizes also were calculated. The control group scored significantly higher than the experimental group on the GMFM total score; the standing scale score; and the walking, running, and jumping scale score (P⫽.01, P⫽.01, and P⫽.02, respectively). In addition, there was a large effect in favor of the control group in GMFM total score; standing scale score; and walking, running, and jumping scale score (Cohen D⫽1.55, 1.59, and 1.36, respectively). That is, the average infant in the control group had a GMFM total score that was 1.55 standard deviations higher than that of the average infant in the experimental group.

Discussion The purpose of this study was to determine whether the addition of 388

f

Physical Therapy

Volume 90

SMOs to a treadmill training protocol for children with DS would lead to improved developmental outcomes. These outcomes included raw developmental test scores and time in study until each infant could take 3 steps independently without support. The hypothesis was that the addition of SMOs would lead to improved developmental test scores and decreased time in study. The results, however, were not so straightforward.

however, the moderate effect size and the low power of this statistical analysis suggest that a study with a larger sample is needed to determine whether this is indeed the case. The fact that the treadmill training is consistently successful may make small improvements difficult to detect statistically without having a large sample size. This difficulty is due to the high level of variability in most developmental measures in the DS population.

“Traditional” treadmill training, as described by Ulrich and colleagues,13,22 is an extremely effective intervention for most infants with DS. In studies by Ulrich and colleagues,13,16 traditional treadmill training led to a large decrease in time to independent walking onset and to improved gait at walking onset compared with no treadmill training. In the current study, infants with DS who received traditional treadmill training were compared with a group of children with DS who received traditional treadmill training as well as early orthotic use. Although there was not a significant difference between the groups in terms of time to walking onset, the statistical effect size (Cohen D⫽0.63) indicates that there may have been a moderate treatment effect in favor of the group that received SMOs. This finding suggests that the SMOs may positively affect the rate of walking development;

Although the major difference between this study and previous treadmill studies on infants with DS is the addition of orthoses, another large difference in this study is the time at which treadmill training was initiated. Previous studies began the treadmill training protocol when children could sit independently13 or take 6 to 10 supported steps on the treadmill.22 Because the current study focused on SMO use, the intervention did not begin until the children were able to pull themselves to a standing position and bear weight on their feet. This method corresponds to that of Ulrich et al,15 who found that children with DS began to prefer alternating stepping patterns on the treadmill when they could pull themselves to a standing position and make forward progress in a prone position. On average, the children in this study pulled themselves to a standing position at 20.5 months, or about 2 weeks after the

Number 3

March 2010

Motor Skill Development in Infants With Down Syndrome children who received treadmill training in the original study began to walk.

group showed a much more linear improvement over time, as indicated by the group ⫻ time interactions.

Although the cohorts from these 2 studies were different and may have varied in factors affecting gross motor development, this large difference in developmental level at the age of 20 months points to the importance of early implementation of the treadmill intervention. Perhaps a better experimental combination of treadmill training and orthoses would include treadmill training, beginning at 10 months of age, and use of SMOs when the children can pull themselves to a standing position independently. This combination would allow the infants to derive the maximal benefits from the treadmill training while still introducing the orthoses at a developmentally appropriate point.

The crawling and kneeling scale contains items such as crawling up and down 4 steps, kneeling, and halfkneeling. Orthoses that limit the mobility around the ankle may make these activities difficult, leading to a slower rate of progression on this scale for the children with orthoses. The standing scale contains items that test balance, such as standing independently for 3 seconds or 20 seconds and lifting one leg for 3 seconds. Although the orthoses provided external stability to the foot and ankle, which helps to maintain balance in older children with DS,6 the children in this study have not yet learned how to balance. The use of orthoses while learning these skills may have limited the available solutions to solving the problem by limiting the amount of movement at the foot and ankle. In turn, the initial rate of increase in the standing scale scores was not as large in the experimental group as it was in the control group.

The effect of orthoses on GMFM scores is complex. As expected, all children in the study showed improvement in their gross motor skills during the course of the intervention. In addition, there was no group difference over the course of the intervention in the overall GMFM score. This finding was expected because all of the children entered the study at the same gross motor level (ie, ability to pull themselves to a standing position) and ended the intervention at the same gross motor level (ie, ability to take 3 independent steps without support). However, the control group showed larger improvements on the crawling and kneeling scale of the GMFM compared with the experimental group. In addition, the predicted developmental trajectories for the crawling and kneeling scale and the standing scale differed by group, leading to significant group ⫻ time interactions. For both of these scales, the control group displayed a rapid increase in scores followed by a leveling out, whereas the experimental March 2010

At 1 month of walking experience, differences between the groups in the GMFM scores persisted. The control group scored significantly higher on the overall GMFM, as well as on the crawling and kneeling; standing; and walking, running, and jumping scales. This finding suggests that the children who learned to walk without the orthoses had an advantage in terms of 4-point mobility, balance, and upright mobility. During the development of a skill, infants experiment with and explore multiple solutions to solving movement tasks.23 Through this process, they learn how to perform a skill and how to adapt that skill to new or differing circumstances. Perhaps the SMOs externally impose limits in ankle and foot alignment and range of motion during this important developmental

period that detract from the variability of practice and thus the adaptability of the learned skills. Although the infants in the experimental group were limited by the SMOs during the development of walking, the infants in the control group had the opportunity to engage in this process of exploration before they received the orthoses. They were able to use the skills that they developed during the attainment of walking in addition to the stability gained from the SMOs to improve their gross motor skills. The imposed limitations of the orthoses during a critical period in development could account for the decreased motor skill level in the children who learned to walk wearing orthoses compared with those who did not receive orthoses until after they were able to walk. Limitations There are some limitations that should be kept in mind when considering the results of this study. This study compared a group of children with DS who received treadmill training with another group of children with DS who received treadmill training and orthoses. We cannot predict the differences between a group that received orthoses and a group that did not receive treadmill intervention. Future studies should contain an orthoses-only group. This study did not consider individual differences in terms of orthosis type. Less restrictive orthoses may be more beneficial to some children with DS. Future studies are needed to determine how to choose orthosis type and whether other forms of orthoses may be more beneficial. In addition, we did not examine lowerlimb kinematics. Future studies are needed to show how orthoses affect lower-limb kinematics. The sample size in this study was smaller than originally planned due to the loss of 5 infants, mostly due to recommendations that infants in the control group begin SMO use before the end

Volume 90

Number 3

Physical Therapy f

389

Motor Skill Development in Infants With Down Syndrome of the study, requiring them to drop out of the study. The small sample size negatively affected statistical power. Further studies should include a larger sample size. In addition, the statistical models represent only children with slower development at the later visit dates. This limitation may reduce the models’ validity at the tail end. Finally, the GMFM scorer was not blinded to group assignment. This lack of blinding may have led to a bias in scoring. Future studies should have a blinded clinician score the data.

Use of SMOs appears to have a detrimental effect on overall motor skill development in infants and new walkers who have learned to walk while wearing the orthoses. Based on this information, health care professionals may want to postpone the use of SMOs in children with DS until they have learned to walk independently. Dr Looper provided concept/idea/research design, writing, data collection and analysis, and project management. Dr Ulrich contributed to the study design and provided consultation. The authors thank the children who participated in this study and their families. This study was completed in partial fulfillment of the requirements for Dr Looper’s doctoral dissertation. This study was reviewed and approved by the University of Michigan Health Sciences Institutional Review Board. Some of the data presented here have been presented in abstract form. An abstract containing data on the developmental trends of 10 infants in this study was presented as a poster at the Combined Sections Meeting of the American Physical Therapy Association; February 6 –9, 2008; Nashville, Tennessee. Another abstract containing data on the developmental test scores of 17 infants at walk-

f

Physical Therapy

This research was supported by Foundation for Physical Therapy PODS II awards to Dr Looper, a grant from the Michigan Physical Therapy Association, and a grant from the Rackham Graduate School, University of Michigan. Clinical trial registration number: NCT00825175 (ClinicalTrials.gov). This article was received January 21, 2009, and was accepted October 25, 2009. DOI: 10.2522/ptj.20090021

Conclusions

390

ing onset was presented as a poster at the 62nd annual meeting of the American Academy of Cerebral Palsy and Developmental Medicine; September 17–20, 2008; Atlanta, Georgia.

Volume 90

References 1 Centers for Disease Control and Prevention. Improved national prevalence estimates for 18 selected major birth defects—United States, 1999 –2001. MMWR. 2006;54:1301–1305. 2 Henderson SE. Motor skill development. In: Lane D, Stratford B, eds. Current Approaches to Down’s Syndrome. London, United Kingdom: Holt, Rinehart & Winston; 1985:187–218. 3 Rast MM, Harris SR. Motor control in infants with Down syndrome. Dev Med Child Neurol. 1985;27:682– 685. 4 Ulrich BD, Ulrich DA. Spontaneous leg movements of infants with Down syndrome and nondisabled infants. Child Dev. 1995;66:1844 –1855. 5 Selby-Silverstein L, Hillstrom HJ, Palisano RJ. The effect of foot orthoses on standing foot posture and gait of young children with Down syndrome. NeuroRehabilitation. 2001;16:183–193. 6 Martin K. Effects of supramalleolar orthoses on postural stability in children with Down syndrome. Dev Med Child Neurol. 2004;46:406 – 411. 7 Cioni M, Cocilovo A, Rossi F, et al. Analysis of ankle kinetics during walking in individuals with Down syndrome. Am J Ment Retard. 2001;106:470 – 478. 8 Looper J, Wu J, Angulo Barroso R, et al. Changes in step variability of new walkers with typical development and with Down syndrome. J Mot Behav. 2006;38: 367–372. 9 Ulrich BD, Haehl V, Buzzi UH, et al. Modeling dynamic resource utilization in populations with unique constraints: preadolescents with and without Down syndrome. Hum Mov Sci. 2004;23: 133–156.

Number 3

10 Parker AW, Bronks R. Gait of Children with Down Syndrome. Arch Phys Med Rehabil. 1980;61:345–351. 11 Parker AW, Bronks R, Snyder CW Jr. Walking patterns in Down’s syndrome. J Ment Defic Res. 1986;30:317–330. 12 Palisano RJ, Walter SD, Russell DJ, et al. Gross motor function of children with Down syndrome: creation of motor growth curves. Arch Phys Med Rehabil. 2001;82:494 –500. 13 Ulrich DA, Ulrich BD, Angulo-Kinzler RM, Yun J. Treadmill training of infants with Down syndrome: evidence-based developmental outcomes. Pediatrics. 2001; 108:E84. 14 Ulrich BD, Ulrich DA, Collier DH, Cole EL. Developmental shifts in the ability of infants with Down syndrome to produce treadmill steps. Phys Ther. 1995;75:14 – 23. 15 Ulrich BD, Ulrich DA, Collier DH. Alternating stepping patterns: hidden abilities of 11-month-old infants with Down syndrome. Dev Med Child Neurol. 1992;34: 233–239. 16 Wu J, Looper J, Ulrich BD, et al. Exploring effects of different treadmill interventions on walking onset and gait patterns in infants with Down syndrome. Dev Med Child Neurol. 2007;49:839 –945. 17 Wu J, Ulrich DA, Looper J, et al. Strategy adoption and locomotor adjustment in obstacle clearance of newly walking toddlers with Down syndrome after different treadmill interventions. Exp Brain Res. 2008; 186:261–272. 18 Angulo-Barroso R, Burghardt AR, Lloyd M, Ulrich DA. Physical activity in infants with Down syndrome receiving a treadmill intervention. Infant Behav Dev. 2008;31: 255–269. 19 Russell DJ, Rosenbaum PL, Avery LM, Lane M. Gross Motor Function Measure (GMFM-66 and GMFM-88) User’s Manual. Cambridge, MA: Mac Keith Press; 2002. 20 Russell DJ, Palisano R, Walter S, et al. Evaluating motor function in children with Down syndrome: validity of the GMFM. Dev Med Child Neurol. 1998;40:693–701. 21 Cohen J. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale, NJ: Lawrence Erlbaum; 1988. 22 Ulrich DA, Lloyd MC, Tiernan CW, et al. Effects of intensity of treadmill training on developmental outcomes and stepping in infants with Down syndrome: a randomized trial. Phys Ther. 2008;88:114 –122. 23 Thelen E, Corbetta D. Exploration and selection in the early acquisition of skill. Int Rev Neurobiol. 1994;37:75–102.

March 2010

Research Report Comprehensive Facial Rehabilitation Improves Function in People With Facial Paralysis: A 5-Year Experience at the Massachusetts Eye and Ear Infirmary Robin W. Lindsay, Mara Robinson, Tessa A. Hadlock

Background. The Facial Grading Scale (FGS) is a quantitative instrument used to evaluate facial function after facial nerve injury. However, quantitative improvements in function after facial rehabilitation in people with chronic facial paralysis have not been shown. Objective. The objectives of this study were to use the FGS in a large series of consecutive subjects with facial paralysis to quantitatively evaluate improvements in facial function after facial nerve rehabilitation and to describe the management of chronic facial paralysis.

Design. The study was a retrospective review. Methods. A total of 303 individuals with facial paralysis were evaluated by 1 physical therapist at a tertiary care facial nerve center during a 5-year period. Facial rehabilitation included education, neuromuscular training, massage, meditationrelaxation, and an individualized home program. After 2 months of home exercises, the participants were re-evaluated, and the home program was tailored as necessary. All participants were evaluated with the FGS before the initiation of facial rehabilitation, and 160 participants were re-evaluated after receiving treatment. All participants underwent the initial evaluation at least 4 months after the onset of facial paralysis; for 49 participants, the evaluation took place more than 3 years after onset.

Results. Statistically significant increases in FGS scores were seen after treatment (P⬍.001, t test). The average initial score was 56 (SD⫽21, range⫽13–98), and the average score after treatment was 70 (SD⫽18, range⫽25–100).

Limitations. A limitation of this study was that evaluations were performed by only 1 therapist.

Conclusions. For 160 patients with facial paralysis, statistically significant improvements after facial rehabilitation were shown; the improvements appeared to be long lasting with continued treatment. The improvements in the FGS scores indicated that patients can successfully manage symptoms with rehabilitation and underscored the importance of specialized therapy in the management of facial paralysis.

R.W. Lindsay, MD LCDR/MC/ USN, is Facial Plastic and Reconstructive Surgeon, Department of Otolaryngology Head and Neck Surgery, National Naval Medical Center, 8901 Wisconsin Ave, Bethesda, MD 20889 (USA); Associate Program Director, National Capitol Consortium Otolaryngology Head and Neck Surgery Residency Program; and Assistant Professor of Surgery, Uniformed Services University of the Health Sciences, Bethesda, Maryland. Address all correspondence to Dr Lindsay at: robin_lindsay@meei. harvard.edu. M. Robinson, PT, MS, NCS, is Facial Nerve Physical Therapist, Department of Otolaryngology/ Head and Neck Surgery, Harvard Medical School, and Division of Facial Plastic and Reconstructive Surgery, Massachusetts Eye and Ear Infirmary. T.A. Hadlock, MD, is Facial Plastic and Reconstructive Surgeon, Department of Otolaryngology/ Head and Neck Surgery, Harvard Medical School, and Division of Facial Plastic and Reconstructive Surgery, Massachusetts Eye and Ear Infirmary. [Lindsay RW, Robinson M, Hadlock TA. Comprehensive facial rehabilitation improves function in people with facial paralysis: a 5-year experience at the Massachusetts Eye and Ear Infirmary. Phys Ther. 2010;90:391–397.] © 2010 American Physical Therapy Association Post a Rapid Response or find The Bottom Line: www.ptjournal.org

March 2010

Volume 90

Number 3

Physical Therapy f

391

Rehabilitation Improves Function in Facial Paralysis

P

eople with long-standing facial paresis experience facial disfigurement and psychological difficulties and may be unable to convey emotions through facial expressions.1 These people often have been told by health care providers that nothing further can be done to improve their facial movements. Many treatments have been used to manage the evolving clinical presentation of facial paralysis as it progresses from the flaccid face to a hypotonic or hypertonic state. Electrical stimulation and regimens of facial movement exercises to be performed with maximal effort are common recommendations; however, both approaches are thought by some clinicians to be harmful, perhaps leading to increased synkinesis.2,3 More recently, facial neuromuscular retraining designed to address the synkinesis associated with partial recovery from facial paralysis2,4 –7 has been shown to benefit people with facial nerve disorders. However, a validated instrument has not been used to quantitatively measure results in a large series of patients.

The Facial Grading Scale (FGS) is a quantitative instrument used by clinicians to evaluate and monitor facial function after facial nerve insult. The FGS is a reliable and validated tool that measures facial dysfunction by scoring resting symmetry, active motion, and synkinesis, the phenomenon of aberrant regeneration.2,8,9 Despite anecdotal evidence to suggest

Available With This Article at ptjournal.apta.org • Video: “Small Smile” • Audio Abstracts Podcast This article was published ahead of print on January 21, 2010, at ptjournal.apta.org.

392

f

Physical Therapy

Volume 90

the benefit of facial rehabilitation, to date there is no quantitative evidence to support the use of comprehensive therapies, such as neuromuscular reeducation, soft-tissue mobilization of the facial muscles, and meditation-relaxation strategies.10 Our objectives were to evaluate a large series of subjects with prolonged facial dysfunction in a facial nerve center setting, to quantitatively evaluate (with FGS scores) improvements after comprehensive facial retraining, and to describe the physical therapy techniques used for these patients.

Method A retrospective chart review was performed for all patients who were referred to a multidisciplinary facial nerve center (Massachusetts Eye and Ear Infirmary Facial Nerve Center) from October 2003 through October 2008 and who received evaluation and treatment by 1 facial nerve physical therapist (M.R.). All patients were evaluated at the facial nerve center by a facial plastic surgeon; the evaluation included a thorough history, physical examination, still photography, and video analysis. Patients who continued to have poor recovery of facial nerve function at least 4 months after the insult were referred for physical therapy evaluation and measurement with the FGS.9 Pretreatment and posttreatment FGS scores were analyzed with paired 1-tailed t tests. Of the 990 patients evaluated at the Massachusetts Eye and Ear Infirmary Facial Nerve Center during the 5-year period, 303 were evaluated by 1 facial nerve physical therapist. These 303 patients were evaluated with the FGS before the initiation of physical therapy, and 160 were re-evaluated after receiving treatment; all underwent the initial evaluation at least 4 months after the onset of facial paralysis. Of the 160 patients for whom

Number 3

pretreatment and posttreatment FGS scores were available, 23% were male (n⫽37) and 77% were female (n⫽123). The average age of the participants was 46 years (SD⫽17, range⫽6 – 81). Facial Rehabilitation Interventions The initial therapy session for all participants included education on the anatomy of the facial nerve and musculature and synkinesis. Instructions on eye protection were provided; these instructions included an eyelid stretch of the levator palpebrae superioris muscle. In addition, participants’ expectations for recovery were discussed. Next, the FGS score was determined by the facial nerve physical therapist. The FGS was used to categorize the face into 5 main regions— brow, ocular region, oral commissure, nasal region, and lips—and to identify the areas of asymmetry and dysfunction. The facial regions were further classified as being flaccid, as having active motion without synkinesis, or as having active motion with synkinesis and hypertonicity. This process allowed appropriate placement into 1 of 4 treatment classification categories: initiation, facilitation, movement control, and relaxation11,12 (Fig. 1; see video, available at ptjournal.apta.org). Initiation treatment category. Participants who had moderate to severe facial asymmetry at rest, who had flaccid facial regions, and who were unable to initiate movement on the affected side were placed in the initiation treatment category.11 Participants in this category were given instructions on gentle superficial massage and active assistive movement exercises and were advised to avoid mass movement patterns (ie, avoid overuse of the uninvolved side, such as by forming a wide smile or chewing gum).

March 2010

Rehabilitation Improves Function in Facial Paralysis

Figure 1. Treatment algorithm for subjects with facial paralysis. EMG⫽electromyographic biofeedback.

Facilitation treatment category. Participants who had mild to moderate facial asymmetry at rest, who were able to initiate slight movement (scores of ⬎2 on the voluntary movement section of the FGS) in any or all regions of the face, and who did not have synkinesis were placed in the facilitation treatment category.11 For these participants, the physical therapy approach had 2 parts: more aggressive soft tissue mobilization of the facial muscles and platysma muscle and neuromuscular reeducation in front of a mirror. First, the participants were given instructions on an individualized soft tissue stretch of the involved musculature. Most participants required attention to the midface, specifically the zygomaticus major and minor muscles. At follow-up appointments, the soft tissue stretching regimen was broadened to involve the upper and lower facial regions. Second, March 2010

neuromuscular re-education exercises were prescribed on the basis of a participant’s impairments of facial motor control, with emphasis on small movements to gain symmetry between the affected and unaffected sides of the face. The strategy used in neuromuscular re-education for this category was to instruct the participants to perform slow, controlled, graded facial expressions to generate symmetry between the sides of the face. It was imperative that the participants initially performed these small-movement exercises with a mirror for visual feedback. Proprioceptive input regarding movement is lacking in facial muscles because they have few, if any, spindle fibers. Thus, visual feedback from the mirror regarding facial movements was necessary for the participants to appreciate the simultaneous motor control of the small facial muscles. Electromyographic biofeedback also was used to guide the appropriate

muscle action to produce symmetry.13 Participants were informed that some typical abnormal movement patterns (synkinesis) might develop with increasing movement. The recognition of any synkinetic movement warranted prompt return to the center so that the participants could be taught how to avoid unwanted movement patterns. Movement control category. Participants who had mild to moderate facial asymmetry at rest and were able to initiate at least slight movement (scores of ⬎2 on the voluntary movement section of the FGS) in any or all regions of the face, but who had developed synkinesis, were placed in the movement control category. For these participants, the physical therapy approach had 3 parts: aggressive deep soft tissue mobilization of the facial muscles and neck, neuromuscular re-education in front of a mirror, and the initiation of

Volume 90

Number 3

Physical Therapy f

393

Rehabilitation Improves Function in Facial Paralysis

Figure 2. Chart reflecting various etiologies of facial paralysis in participants undergoing facial nerve physical therapy.

meditation-relaxation strategies. The primary cause of facial asymmetry in the movement control group was synkinesis, not weakness, as in the facilitation group. Nevertheless, the physical problems were similar, arising largely from abnormal movement patterns. Therefore, the strategy used in neuromuscular re-education for this category was to instruct the participants to perform slow, controlled, graded facial expressions to generate symmetry between the sides of the face while simultaneously controlling synkinetic movements in other regions of the face. For example, when participants complained of ocular synkinesis with smiling or puckering their lips for eating, drinking, or talking, smallmovement exercises were taught. They were trained to form a small, symmetric smile while controlling synkinesis of the obicularis oris muscle by widening the orbital region. For midfacial synkinesis, the participants were instructed to gently close their eyes while releasing the synkinesis in the midface. When participants developed synkinesis of the platysma muscle, soft tissue massage of the platysma muscle and active practice in minimizing the synkinesis were used to prevent the overactive depressor function of the platysma muscle from overcoming the effort of the impaired zygomaticus muscle. 394

f

Physical Therapy

Volume 90

At follow-up visits, the neuromuscular re-education exercises were advanced to teach the participants how to control other zones of the face while still minimizing synkinesis. Relaxation category. Participants with severe pan-facial tightness attributable to synkinesis and hypertonicity were placed in the relaxation category. For these individuals, the physical therapy approach had 3 parts: aggressive deep soft tissue mobilization of the facial muscles and neck, neuromuscular re-education in front of a mirror, and a strong focus on meditation-relaxation strategies. These participants had limited movement because of tightness rather than weakness. Therefore, the emphasis of treatment for these participants was on relaxation-meditation exercises. Meditation with guided visual imagery was focused on relieving the tension in the synkinetic musculature. Verbal cues to help minimize synkinesis included “drain the tension around the eye” (ie, for ocular synkinesis) and “deflate the fully inflated balloon in your cheek” (ie, for midfacial synkinesis felt during closing of the eyes). In addition, our clinic provided participants with a relaxation audio CD to facilitate integration of the relaxation strategies at home.

Number 3

The typical guideline for neuromuscular re-education exercises was 20 to 40 repetitions 2 to 4 times per day. However, more frequent repetition was encouraged, the notion being that repetition improves motor learning.3 The guideline for soft tissue massage was 10 repetitions 1 or 2 times per day. The frequency of relaxation was 1 or 2 times per day. Monthly or bimonthly appointments were scheduled well ahead of time to tailor and further advance a participant’s home program and to reevaluate impairments and functional goals. Frequently, a participant’s treatment category changed as movement returned or synkinesis developed, requiring changes in the treatment regimen. Chemodenervation All participants who developed synkinesis were considered to be eligible to receive chemodenervation for the further management of synkinesis; this technique has been shown to improve the quality of life in subjects with facial paralysis.14 Botulinum toxin is a potent neurotoxin that causes temporary paralysis of the hyperkinetic musculature. The philosophy at the Massachusetts Eye and Ear Infirmary Facial Nerve Center is for patients to participate in facial muscle retraining for a minimum of 16 weeks before the initiation of botulinum toxin injections.15 The premise underlying this philosophy is that patients gain an appreciation of the role of each muscle in the face and an understanding of how synkinesis affects their facial expressions.

Results Figure 2 summarizes the etiologies of facial nerve paralysis. The average initial FGS score was 56 (SD⫽21, range⫽13–98), and the average score after treatment was 70 (SD⫽18, range⫽25–100). Statistically significant increases in FGS scores were seen after treatment (P⬍.005, paired 1-tailed t test) March 2010

Rehabilitation Improves Function in Facial Paralysis (Fig. 3). Forty-nine participants were evaluated more than 3 years after the onset of facial paralysis. For this subset of subjects, the average initial FGS score was 55 (SD⫽20, range⫽32–90), and the average score after treatment was 67 (SD⫽17, range⫽41–94); these values indicated significant improvements after therapy (P⬍.001, paired 1-tailed t test). Eighty-seven participants had Bell palsy. For this subset of participants, the average initial FGS score was 60 (SD⫽21, range⫽13–98), and the average score after treatment was 75 (SD⫽17, range⫽25–100). Statistically significant increases in FGS scores were seen after treatment (P⬍.001, paired 1-tailed t test). Twenty-three participants were evaluated after acoustic neuroma removal. For these participants, the average initial FGS score was 50 (SD⫽22, range⫽20 –93), and the average score after treatment was 65 (SD⫽17, range⫽35–95). Eleven participants had Ramsay Hunt syndrome. For this subset of participants, the average initial FGS score was 54 (SD⫽19, range⫽46 – 88), and the average score after treatment was 66 (SD⫽19, range⫽46.5–98.5). Statistically significant increases in FGS scores were seen after treatment (P⬍.001, paired 1-tailed t test) for both participants with acoustic neuroma and participants with Ramsay Hunt syndrome. In addition, 7 participants diagnosed with Lyme disease showed statistically significant improvements after therapy (P⬍.001, paired 1-tailed t test). For this subset of participants, the average initial FGS score was 63 (SD⫽20, range⫽30.5– 82.5), and the average score after treatment was 76 (SD⫽12, range⫽61.0 – 89.5). Figure 4 shows the appearance of a typical participant before treatment and after treatment (physical therapy and chemodenervation).

March 2010

Figure 3. Chart reflecting improvements in Facial Grading Scale (FGS) scores after physical therapy. The average initial FGS score was 55.9 (SD⫽21, range⫽13–98), and the average score after treatment was 69.9 (SD⫽18, range⫽25–100). Statistically significant increases in FGS scores were seen after treatment (P⬍.005, t test). Error bars represent standard errors (initial⫽5.3, posttreatment⫽5.5).

Discussion Several investigators have attempted to examine the effects of physical therapy on Bell palsy; however, recent reviews have not established convincing evidence to support integrating physical therapy into clinical practice.9,16 One randomized controlled trial did show a statistically significant improvement in facial symmetry with neuromuscular re-education in participants with Bell palsy; however, the control group was treated with electrostimulation, which might have impaired recovery.3,5 In addition, the initial FGS scores were obtained at the onset of facial paralysis, and the posttreatment scores were obtained after 3 months; up to 95% of participants would be expected to recover during this time period without physical therapy if treated with prednisone and valacyclovir.17 In the present study, a subset of participants with poor recovery of function after Bell palsy were evaluated

and treated by a physical therapist at least 4 months after the onset of paralysis. Ten percent of people with Bell palsy never regain normal function, and 5% to 15% experience severe sequelae, including residual paralysis, facial contracture, synkinesis, or spasm.1,17,18 Using FGS scores, we showed a quantitative benefit of comprehensive facial rehabilitation for participants who failed to recover from either Bell palsy or other facial nerve insults during an adequate observation period (4 months). Incomplete or aberrant recovery affects some people with delayed recovery after facial paralysis. Our study established the benefit of standardized, well-documented facial rehabilitation for both participants with Bell palsy and participants with chronic facial paralysis of other etiologies. On the basis of our observations, participants with minimal clinical signs of recovery after 16 weeks tended to lag significantly behind those with earlier recovery. We have repeatedly noted delayed recovery with a high degree

Volume 90

Number 3

Physical Therapy f

395

Rehabilitation Improves Function in Facial Paralysis

Figure 4. Woman with a history of Bell palsy 18 years earlier and with synkinesis before treatment (A–C) and typical results after treatment (physical therapy and chemodenervation) (D–F). (A and D) Mentalis dimpling and synkinesis of the platysma muscle with brow elevation before treatment (A) resolved with treatment (D). (B) Mentalis dimpling, synkinesis of the platysma muscle, and midfacial synkinesis with closing of the eyes. The mentalis dimpling and synkinesis of the platysma muscle resolved with treatment. However, the midfacial synkinesis persisted (E); this problem is difficult to successfully address with chemodenervation without compromising upper-lip function. (C) Before treatment, the woman had an asymmetric smile with narrowing of the palpebral fissure width. (F) During neuromuscular reeducation, the woman was taught to relax the muscles around the eye while smiling to widen the palpebral fissure and to decrease the oral commissure excursion on the unaffected side to gain a more symmetric smile.

396

f

Physical Therapy

Volume 90

Number 3

March 2010

Rehabilitation Improves Function in Facial Paralysis of synkinesis in people who have skull base tumors and in whom the facial nerve is left anatomically intact but does not stimulate movement at the conclusion of extirpation, as in people with Ramsay Hunt syndrome and Lyme disease. Even in people with a poor prognosis for recovery of facial function, physical therapy was found to increase FGS scores. We also noted that many participants benefited from physical therapy even when their facial nerve insults had occurred more than 3 years before the initiation of physical therapy. In fact, of the 160 participants for whom FGS scores before and after therapy were available, 49 participants were beyond the 3-year time frame (range⫽3–30 years), suggesting that there is no impairment beyond which a window for nerve recovery does not exist. The significant quantitative improvements in FGS scores that we found with facial rehabilitation support the integration of physical therapy into the treatment strategy for people with poor recovery after facial nerve insults. Training therapists in subspecialties to adequately treat this underserved patient population would likely provide significant benefits. Facial rehabilitation with neuromuscular re-education, soft tissue mobilization of the facial muscles, meditation-relaxation strategies, and chemodenervation, when needed, resulted in statistically significant improvements in participants with chronic facial paralysis; the improvements appeared to be long lasting with continued treatment. Improvements in FGS scores indicate that people can successfully manage symptoms with comprehensive interventions and highlight the importance of specialized physical therapy in the manage-

March 2010

ment of facial paralysis and teaching people self-management strategies. Thus, facial rehabilitation should be offered to people with chronic facial dysfunction to optimize their facial movements, their social functioning, and their facial expression of emotions. Facial rehabilitation should be delivered by a provider skilled in neuromuscular reeducation to manage the complex and changing needs of people with facial paralysis. All authors provided concept/idea/research design, writing, data collection, data analysis, and consultation (including review of manuscript before submission). Dr Lindsay and Ms Robinson provided project management. Ms Robinson and Dr Hadlock provided subjects. Dr Hadlock provided facilities/equipment. This study was approved by the Institutional Review Board at the Massachusetts Eye and Ear Infirmary. This work was presented at the XI International Facial Nerve Symposium; April 25–28, 2009; Rome, Italy. This article was received May 29, 2009, and was accepted September 21, 2009. DOI: 10.2522/ptj.20090176

References 1 Peitersen E. Bell’s palsy: the spontaneous course of 2,500 peripheral facial nerve palsies of different etiologies. Acta Otolaryngol Suppl. 2002;(549):4 –30. 2 Brach JS, VanSwearingen JM. Physical therapy for facial paralysis: a tailored treatment approach. Phys Ther. 1999;79: 397– 404. 3 Diels HJ. Facial paralysis: is there a role for a therapist? Facial Plast Surg. 2000;16: 361–364. 4 Brach JS, VanSwearingen JM, Lenert J, Johnson PC. Facial neuromuscular retraining for oral synkinesis. Plast Reconstr Surg. 1997;99:1922–1931; discussion 1932–1933. 5 Manikandan N. Effect of facial neuromuscular reeducation on facial symmetry in patients with Bell’s palsy: a randomized controlled trial. Clin Rehabil. 2007;21: 338 –343.

6 Cronin GW, Steenerson RL. The effectiveness of neuromuscular facial retraining combined with electromyography in facial paralysis rehabilitation. Otolaryngol Head Neck Surg. 2003;128:534 –538. 7 Nakamura K, Toda N, Sakamaki K, et al. Biofeedback rehabilitation for prevention of synkinesis after facial palsy. Otolaryngol Head Neck Surg. 2003;128:539 –543. 8 Coulson SE, Croxson GR, Adams RD, O’Dwyer NJ. Reliability of the “Sydney,” “Sunnybrook,” and “House Brackmann” facial grading systems to assess voluntary movement and synkinesis after facial nerve paralysis. Otolaryngol Head Neck Surg. 2005;132:543–549. 9 Ross BG, Fradet G, Nedzelski JM. Development of a sensitive clinical facial grading system. Otolaryngol Head Neck Surg. 1996;114:380 –386. 10 Cardoso JR, Teixeira EC, Moreira MD, et al. Effects of exercises on Bell’s palsy: systematic review of randomized controlled trials. Otol Neurotol. 2008;29:557– 560. 11 VanSwearingen J. Facial rehabilitation: a neuromuscular reeducation, patientcentered approach. Facial Plast Surg. 2008;24:250 –259. 12 VanSwearingen JM, Brach JS. Validation of a treatment-based classification system for individuals with facial neuromotor disorders. Phys Ther. 1998;78:678 – 689. 13 Ross B, Nedzelski JM, McLean JA. Efficacy of feedback training in long-standing facial nerve paresis. Laryngoscope. 1991;101: 744 –750. 14 Mehta RP, Hadlock TA. Botulinum toxin and quality of life in patients with facial paralysis. Arch Facial Plast Surg. 2008;10: 84 – 87. 15 Hadlock TA, Greenfield LJ, WernickRobinson M, Cheney ML. Multimodality approach to management of the paralyzed face. Laryngoscope. 2006;116: 1385–1389. 16 Teixeira LJ, Soares BG, Vieira VP, Prado GF. Physical therapy for Bell’s palsy (idiopathic facial paralysis). Cochrane Database Syst Rev. 2008;(3):CD006283. 17 Hato N, Yamada H, Kohno H, et al. Valacyclovir and prednisolone treatment for Bell’s palsy: a multicenter, randomized, placebo-controlled study. Otol Neurotol. 2007;28:408 – 413. 18 Holland NJ, Weiner GM. Recent developments in Bell’s palsy. BMJ. 2004;329: 553–557.

Volume 90

Number 3

Physical Therapy f

397

Research Report

Interhemispheric Modulation Induced by Cortical Stimulation and Motor Training Julie A. Williams, Alvaro Pascual-Leone, Felipe Fregni J.A. Williams, MA, is Research Assistant, Department of Neurology, Harvard Medical School, Boston, Massachusetts. A. Pascual-Leone, MD, PhD, is Associate Professor, Department of Neurology, Harvard Medical School. F. Fregni, MD, PhD, MPH, is Assistant Professor, Department of Neurology, Harvard Medical School, 330 Brookline Ave, Boston, MA 02215 (USA). Dr Fregni also is Director, Laboratory of Neuromodulation, Spaulding Rehabilitation Hospital, Harvard Medical School. Address all correspondence to Dr Fregni at: [email protected]. [Williams JA, Pascual-Leone A, Fregni F. Interhemispheric modulation induced by cortical stimulation and motor training. Phys Ther. 2010;90:398 – 410.] © 2010 American Physical Therapy Association

Background. Interhemispheric inhibition might be a beneficial cortico-cortical interaction, but also might be maladaptive in people with neurological disorders. One recently revisited technique that has been shown to be effective in improving motor function in people with stroke using interhemispheric modulation is transcranial direct current stimulation (tDCS). Objective. The aim of this study was to investigate the effects of tDCS combined with unilateral motor training with contralateral hand restraint on interhemispheric inhibition between the dominant and nondominant hemispheres of the brain and on motor performance in participants who were healthy.

Design. This was a double-blind, prospective, single-center study with participants who were healthy.

Methods. Twenty participants who were healthy were randomly assigned to receive either active or sham tDCS of the primary motor cortex (M1) bilaterally combined with unilateral motor training and contralateral hand restraint. A blinded rater assessed motor function and cortical excitability, including assessment of transcallosal inhibition (TCI).

Results. There was a larger increase in motor performance in the nondominant hand for the active tDCS group compared with the sham tDCS group. In addition, a decrease in cortical excitability in the dominant hemisphere and a decrease in TCI from the dominant to nondominant hemisphere were observed for the active tDCS group only. The TCI decrease in the active tDCS group was correlated with motor performance improvement for the nondominant hand.

Limitations. Limitations of this study included missing the effect of intracortical inhibition due to a floor effect, not using the optimal tDCS montage, and not being able to assess the effects of other variables such as gender due to the small sample size.

Post a Rapid Response or find The Bottom Line: www.ptjournal.org 398

f

Physical Therapy

Conclusions. The results indicate that tDCS enhances the effects of unilateral motor training and contralateral hand restraint on motor function, and this benefit is associated with a different mechanism of action characterized by bihemispheric modulation in which TCI from the dominant to the nondominant hemisphere is decreased. Transcranial direct current stimulation might be a useful tool to enhance the motor effects of constraint-induced movement therapy.

Volume 90

Number 3

March 2010

Interhemispheric Modulation Induced by Cortical Stimulation and Motor Training

T

he concept of interhemispheric inhibition has been studied for many years, beginning in 1940 with Curtis, who was among the first researchers to stimulate one hemisphere of the brain and measure an evoked response in the opposite hemisphere.1 Many researchers have observed the behavioral and neurophysiological effects of interhemispheric interactions.2– 4 Although this neurophysiological parameter has been intensively described for the motor system,5– 8 interhemispheric inhibition also is an important physiological mechanism of interhemispheric interactions for attention,2,4,9 memory,10,11 and mood.12,13 Interhemispheric inhibition might be beneficial, such as during the execution of unimanual movements14 in people who are healthy, or it might be maladaptive in people with certain neurological diseases such as stroke. In people with stroke, for instance, there is an increase in the excitability of the unaffected hemisphere, presumably due to reduced transcallosal inhibition (TCI) from the damaged hemisphere and increased use of the intact hemisphere.15 Although this mechanism might be beneficial during acute phases of stroke, once the injury is stable, excitatory input to the perilesional area would seem to be best to maximize the capability of the preserved neurons in the injured tissue to drive behavioral output. For this reason, increased TCI from the unaffected hemisphere seems undesirable in people with stroke. Indeed, modulation of interhemispheric interactions with transcranial magnetic stimulation (TMS) to decrease TCI has been used successfully to promote speech and motor recovery in people with stroke.16,17 One recently revisited technique that has been shown to be effective in improving motor function in peoMarch 2010

ple with stroke using interhemispheric modulation is transcranial direct current stimulation (tDCS).18,19 Transcranial direct current stimulation has some advantages over other techniques of noninvasive brain stimulation. First, the small size of the electrodes and stimulator allow for greater portability during motor training. Second, tDCS is potentially able to stimulate both hemispheres simultaneously, thereby increasing activity in one hemisphere and decreasing it in the opposite hemisphere, as shown by a recent modeling study.20 This technique, however, has not been tested in humans. Last, the effects of tDCS are long lasting; 13 minutes of stimulation can change brain excitability for up to 90 minutes.21 One important aspect of tDCS is that this technique is truly neuromodulatory, as it changes the membrane resting potential, thereby rendering the stimulated cortical area ready to be modulated by external stimuli, via behavioral intervention.18,19,22,23 Therefore, we aimed to explore whether the effects of a behavioral intervention (ie, constraint-induced movement therapy [CIMT]) that is associated with the idea of decreasing activity of one hemisphere to release activity of the opposite hemisphere can be enhanced by active tDCS of the motor cortices. Our hypothesis was that active tDCS combined with unilateral motor training with contralateral hand restraint induces a larger motor function improvement and differential changes in cortical excitability compared with sham tDCS with the same motor training therapy. We decided to study people who were healthy because the performance differences between the nondominant and dominant hands can be used to explore interhemispheric differences and their consequences on motor function. In a previous

study, Boggio et al24 demonstrated that anodal tDCS applied to the primary motor cortex of the nondominant hand resulted in increased motor function, as measured by the Jebsen-Taylor Hand Function Test (JTHF). Finally, although previous studies have shown changes in cortical activity with the application of CIMT25,26 and tDCS,24 to our knowledge, no one has explored the neurophysiological correlation using a combination of motor training and tDCS, particularly its effects on interhemispheric inhibition. Therefore, we explored this concept in people who were healthy, assessing: (1) whether unilateral motor training with contralateral hand restraint changes TCI and (2) whether tDCS can modulate activity and, when combined with unilateral motor training with contralateral hand restraint, enhance the behavioral effects due to paradoxical facilitation. Because of the differential activity between the nondominant and dominant motor cortices, we elected to perform this study using participants who were healthy as an initial step. Participants poststroke have a shifted balance of interhemispheric inhibition that depends on lesion du-

Available With This Article at ptjournal.apta.org • eTable 1: Shaping Tasks Used in the Study During Motor Training • eTable 2: Intracortical Inhibition and Facilitation • eTable 3: Interhemispheric Inhibition • eTable 4: Correlation Values for the Active Transcranial Direct Current Stimulation (tDCS) Group • Audio Abstracts Podcast This article was published ahead of print on January 28, 2010, at ptjournal.apta.org.

Volume 90

Number 3

Physical Therapy f

399

Interhemispheric Modulation Induced by Cortical Stimulation and Motor Training

Figure 1. Diagram of study flow, showing time in hours. CE⫽cortical excitability, motor function⫽assessment of motor function with behavioral testing, motor training⫽ unilateral motor training using contralateral hand restraint, tDCS⫽transcranial direct current stimulation, T1⫽measurement taken of nondominant hand after tDCS treatment ended, T2⫽measurement taken of nondominant hand 2 hours into motor training, and T3⫽measurements taken of both dominant and nondominant hands after completion of motor training.

ration and location, thus adding a significant source of variability. Therefore, studying people who are healthy is important to understand the core features of interhemispheric interactions and ultimately guide therapeutic interventions in patients poststroke.

Method Participants Twenty right-handed participants (14 female, 6 male) who were healthy were tested. Their mean age was 20.4 years (SD⫽1.7, range⫽19 – 24). All participants were right handed, as assessed by the Edinburgh Handedness Inventory (participants had to score above 80% on this scale [range⫽80%–100%]); therefore, we refer to them here as right-handed, left-hemisphere– dominant participants. All participants were college students and thus shared the same level of education. The study was performed in accordance with the Declaration of Helsinki (1964) at the Berenson-Allen Center for Noninvasive Brain Stimulation, Boston, Massachusetts. Written informed consent was obtained 400

f

Physical Therapy

Volume 90

from all participants before inclusion in the study, which was approved by the Human Subjects Review Committee of Harvard Medical School. Study Design A diagram of the study flow is presented in Figure 1. Baseline measurements for each participant were obtained for the following measures: (1) motor function, as indexed by the JTHF; (2) cortical excitability, as measured by motor-evoked potentials (MEPs) from the left and right hemispheres; (3) intracortical inhibition (ICI) and intracortical facilitation (ICF), using the paired-pulse technique; and (4) TCI, using an ISI of 10 milliseconds. All of the cortical excitability measurements were based on the motor threshold for the first dorsal interosseous muscle in both the right and left hands. After baseline measurements were obtained, the participants were randomly assigned to receive either active tDCS and unilateral motor training with contralateral hand restraint (active tDCS group) or sham tDCS and unilateral motor training with

Number 3

contralateral hand restraint (sham tDCS group) using the strategy of block randomization and with equal distribution (1:1). One important aspect here is that this trial was truly double blind, as neither the coinvestigator performing the treatment and evaluations nor the participant was aware of which treatment was being used. It was possible to blind the investigator administering the treatment, as we used a device that allows the use of a code to determine whether the treatment was active or sham. Only one investigator, who was not involved with treatment or evaluation, knew the code. After baseline measurements were obtained, unilateral motor training with contralateral hand restraint was performed for 3 hours with 2 breaks in order to perform intermittent evaluations. During the first 40 minutes of motor training, tDCS was applied. All participant evaluations (JTHF, cortical excitability, TCI) were performed at 4 intervals throughout the study. First, initial (baseline) measurements were taken prior to the start of tDCS and unilateral motor training with contralateral hand restraint. After the tDCS treatment ended, a second measurement was taken of only the nondominant hand (time point 1 [T1]). Two hours into the unilateral motor training, a third measurement was taken using only the nondominant hand (time point 2 [T2]). Finally, after the unilateral motor training was completed, the last evaluations were done, taking measurements from both the dominant and nondominant hands (time point 3 [T3]). Assessments were not made with the dominant hand at T1 and T2 due to the need for restricting the dominant hand during unilateral motor training. Participants were trained on the JTHF (they performed 10 training trials in order to reach a stable plateau). Procedures for this test were folMarch 2010

Interhemispheric Modulation Induced by Cortical Stimulation and Motor Training lowed according to the original description by Jebsen et al.27 Participants performed practice trials, baseline trials, 2 abbreviated trials (which recorded only the nondominant hand during unilateral motor training with contralateral hand restraint), and a final evaluation for each hand. Study Intervention Procedures Transcranial direct current stimulation. Direct current stimulation was applied using a pair of salinesoaked sponge electrodes (size 35 cm2) and delivered by a specially developed, battery-driven, constant current stimulator* with a maximum output of 10 mA. The anode electrode was placed over the right (nondominant) primary motor cortex (M1), and the cathode electrode was placed over the left (dominant) primary motor cortex (M1). We used the 10/20 international electroencephalographic system for electrode placement, in which M1 corresponds to C3 or C4. The rationale of this montage is to increase excitability (with anodal tDCS) in the hemisphere of the hand that is being trained (nondominant hand) and decrease it in the hemisphere of the dominant hand (with cathodal tDCS) to decrease TCI. This montage has been shown to be effective, according to our modeling study.20 For the active conditions, participants received 1-mA tDCS for 40 minutes (with 10 seconds of rampup and ramp-down) concurrent with the beginning of unilateral motor training. The same procedure was used for sham stimulation, but current was applied just for the first 30 seconds. This procedure is reliable to blinded participants for the respective stimulation condition.28

* NeuroConn GmbH, Eldith, Ilmenau, Germany.

March 2010

One participant complained of nausea at the start of tDCS during the unilateral motor training. At that time, the participant was evaluated by the study physician and was released from further participation. Other mild adverse effects such as mild headache and scalp tingling were observed in a small number of participants but did not interfere with their participation in this study. Unilateral motor training with contralateral hand restraint. At the same time as the tDCS was set up, a hand mitt (Skil-Care Rigid Palm Padded Mitt†) was placed on the participants’ dominant hand. This mitt remained on throughout the entire 3-hour period when unilateral motor training was being performed with the nondominant hand. Participants were not allowed to remove the mitt, and a piece of tape was placed over the edge to monitor whether they attempted to remove it. Participants were supervised by study staff who helped to direct them throughout the entire period the hand mitt was worn. Nine different shaping tasks were used during this 3-hour period, which included buttoning a shirt, pouring water, and folding towels (see the complete list in eTab. 1, available at ptjournal. apta.org). All 9 tasks were performed using the left (nondominant) hand only. During the 3-hour period, the participants were able to complete, on average, 3 repetitions of each task (given the time to complete the tasks and the small interval between them), and all participants completed a minimum of 2 sets of all 9 tasks during the 3-hour period the hand mitt was worn. Study Assessment Procedures We used single-pulse TMS to measure corticospinal excitability. Focal TMS was performed using a commer†

AliMed Inc, 297 High St, Dedham, MA 02026.

cially available figure-of-eight coil (outside diameter of each wing⫽7 cm) and 2 Magstim 200 stimulators that were coordinated using a Bistim device.‡ Initially, we ensured that the muscle was completely relaxed by online monitoring with surface electromyographic activity at high gain (10 –50 ␮V). Optimal scalp position for induction of MEPs was determined following published guidelines.29 The coil was held tangentially to the skull with the handle pointing occipitally with an angle of 45 degrees to the midline of the participant’s head. To determine the threshold intensity, stimulation was initiated at 65% (or higher, if needed) of the maximum stimulator output and decreased in 5% increments, or 2% of the maximum stimulator output, when near the threshold level. The motor threshold was defined as the lowest stimulation intensity to produce at least 5 MEPs with peakto-peak amplitudes of 50 ␮V from 10 consecutive stimuli. The mean values for motor threshold in the right and left hemispheres, respectively, were 45% and 43% stimulator output intensity. Optimal scalp position and motor threshold were determined at baseline. We then recorded MEPs by adjusting the TMS intensity to achieve an MEP in the first dorsal interosseous muscle of about 1-mV peak-to-peak amplitude, and intensity was maintained constant throughout the experiment. We recorded 10 MEPs for each time point. The MEP gives a measure of global corticospinal excitability.30 To measure ICI and ICF, we used the paired-pulse technique, in which a subthreshold stimulus delivered using TMS can modulate a subsequent test MEP delivered a few millisec‡ Magstim Company Ltd, Spring Gardens, Whitland, Carmarthenshire, Wales, United Kingdom SA34 0HR.

Volume 90

Number 3

Physical Therapy f

401

Interhemispheric Modulation Induced by Cortical Stimulation and Motor Training onds after the conditioning stimulus.30 The mechanisms involved with this modulation are based on interneuronal activity; therefore, it can measure ICI and ICF, depending on the interval between pulses. Maximum inhibitory effects are found when the ISIs are between 1 and 4 milliseconds; when the interval is longer (ie, 9 –15 milliseconds), facilitation can be seen.30 Therefore, for the paired-pulse measurement, a first (subthreshold) conditioning stimulus (using an intensity of 90% of motor threshold) was applied, followed at a variable ISI by a second (suprathreshold) test stimulus (using an intensity of 130% of motor threshold) applied to the same location at which the first stimulus was applied.31 We used the following ISIs: 2, 3, 6, 10, and 12 milliseconds. Various ISIs were randomized and intermixed with control trials (test stimulus alone). For each ISI, 5 trials were averaged, and the resulting MEP amplitude was converted to a percentage of the control MEP amplitude. Paired-pulse parameters were expressed as the amount of inhibition (for ICI, we used ISIs of 2 and 3 milliseconds) and facilitation (for ICF, we used ISIs of 10 and 12 milliseconds). Paired-pulse measurement was performed only at baseline and at the end of unilateral motor training with contralateral hand restraint (T3). In order to assess interhemispheric interactions, we used the technique of bi-hemispheric stimulation, which also used a paired-pulse sequence (but in this case, stimulation was delivered to both motor cortices). The notion here is that inhibitory and excitatory interactions are constantly transferred between hemispheres. Therefore, if a suprathreshold stimulus is applied to the motor cortex of one hemisphere (using an intensity of 120% of motor threshold) and 10 milliseconds later a suprathreshold stimulus is applied to the contralat402

f

Physical Therapy

Volume 90

eral motor cortex (using an intensity of 130% of motor threshold), this later MEP will be modulated by the first pulse. The period of inhibition is between 10 and 15 milliseconds after the minimum corticospinal conduction time to the recorded hand muscle cortical area.30 This inhibition is mediated via transcallosal pathways and originates within the motor cortex. We calculated the percentage of inhibition for each stimulus before and after treatment was performed. Transcranial magnetic stimulation output intensities were maintained from baseline measurements through final evaluations in order to allow for comparison of results from the beginning of the experiment to the final measurements. The electromyographic activity was amplified with a band-pass filter between 10 and 2,000 Hz, and the signal was digitized at a frequency of 5 kHz using a PowerLab 4/25T device§ and stored on a computer for offline analysis using the Scope software (version 4.0.8).§ Data Analysis The main outcome measures in this study were: (1) motor function change, as indexed by the JTHF, and (2) TCI, as assessed using TMS. For motor function, the dependent variable was performance on the JTHF, and the independent variables were time of stimulation and treatment group. We, therefore, performed a mixed analysis-of-variance (ANOVA) model, including as covariates the fixed-effect variables of treatment group, time of stimulation, and group⫻time interaction and the random-effect variable of participant ID. We performed a similar model to assess TCI, but in this analysis we used percentage of MEP inhibition as a dependent variable. When appro§

ADInstruments Inc, 2205 Executive Cir, Colorado Springs, CO 80906.

Number 3

priate, post hoc analyses were conducted using Bonferroni correction for the pair-wise comparisons. We assessed other parameters of cortical excitability (MEP changes and TCI) as an exploratory analysis and, therefore, not correcting for multiple comparisons. Finally, we performed pair-wise correlations between changes in parameters of cortical excitability and motor function changes, as indexed by the JTHF, as an exploratory analysis (without correction for multiple comparisons) using the Pearson correlation test.

Results Motor Function Assessment We initially compared baseline motor function, as indexed by the JTHF, between the active and sham tDCS groups and found no significant differences (P⬎.05). However, there was a differential improvement in motor function, as indexed by JTHF, between the active and sham tDCS groups (Fig. 2). The group ⫻ time interaction was significant (F3,54⫽ 3.02, P⫽.037) for the left (nondominant) hand. The overall results showed that there was a larger improvement in the active tDCS group compared with the sham tDCS group: mean (SD) motor function change between baseline and T3 was 8.9%⫾7.8% in the active tDCS group and 3.9%⫾4.9% in the sham tDCS group (t18⫽1.7, P⫽.05). Both groups had a differential improvement across time points. For the active tDCS group, there was a tendency for a differential effect between baseline and T1 (immediately after tDCS) (t9⫽1.7, P⫽.06) and a significant effect between T2 and T3 (t9⫽3.7; P⫽.008, corrected P value), but there was no significant difference between T1 and T2 (t9⫽0.49, P⫽.3). For the sham tDCS group, the only significant difference was between T1 and T2 (t9⫽4.1; P⫽.004, corrected P value); there was no difference between baseline and T1 (t9⫽1.2, P⫽.14) or between T2 and March 2010

Interhemispheric Modulation Induced by Cortical Stimulation and Motor Training T3 (t9⫽0.85, P⫽.8) (Fig. 2). We performed a similar analysis for the right (dominant) hand. The group ⫻ time interaction was not significant (F1,18⫽0.1, P⫽.75), suggesting that tDCS was not associated with motor function changes in the dominant hand. Cortical Excitability Assessments Motor-evoked potential analysis. We initially measured MEPs before and after training and stimulation in both hemispheres (Fig. 3). We performed 2 ANOVA models, as the time point assessments were different for the left and right hemispheres. The analysis for the left and right hemispheres is shown below. A mixed ANOVA model showed a significant group ⫻ time interaction (F1,18⫽12.59, P⫽.0023) for the left (dominant) hemisphere (corresponding to the restricted right hand), suggesting a differential change in cortical excitability when comparing active and sham tDCS. Post hoc analysis revealed that active tDCS was associated with a significant mean (SD) reduction in MEP amplitude of 19.8%⫾21.4% in the dominant hemisphere (t9⫽3.9; P⫽.004, corrected P value), but there was no significant change in the sham tDCS group (mean [SD] increase of 0.8%⫾8.3%, t9⫽0.45, P⫽.65) (Fig. 2). For the right (nondominant) hemisphere (corresponding to the nondominant trained hand), we initially performed an ANOVA with all data points and found only a tendency for a significant interaction (F3,54⫽2.00, P⫽.1) and a tendency for a significant time effect (F3,54⫽2.61, P⫽.06). Because these results suggested that there was a change in cortical excitability in the same direction in both groups, we performed an exploratory analysis (as this was decided post hoc) considering only 2 time points (baseline and T1). This analyMarch 2010

Figure 2. Changes in Jebsen-Taylor Hand Function Test (JTHF) scores from the beginning to the end of the experiment. A decrease in the amount of time necessary to complete the test was observed in both the active transcranial direct current stimulation (tDCS) group and the sham tDCS group. The active tDCS group improved performance by the first evaluation after baseline and had greater improvement compared with the sham tDCS group. Error bars represent the standard error or the mean. T1⫽measurement taken of nondominant hand after tDCS treatment ended, T2⫽measurement taken of nondominant hand 2 hours into motor training, and T3⫽measurements taken of both dominant and nondominant hands after completion of motor training.

sis showed a significant group ⫻ time interaction (F1,18⫽6.50, P⫽.02). There was a significant increase in MEP amplitude after active tDCS (mean [SD] increase of 13.8%⫾ 14.2%, P⫽.013) and no significant change after sham tDCS (mean [SD] decrease of 5.6%⫾23.4%, P⫽.46). Interestingly, the active tDCS group had a rapid increase of MEP amplitude after stimulation and then had only a small increase, and the sham tDCS group had a delayed increase in MEP amplitude. At the end of the study, although there was a larger increase in MEP amplitude from the nondominant hemisphere in the active tDCS group compared with the sham tDCS group (20.1%⫾25.8% versus 10.0%⫾19.2%), this difference was not significant (between-group comparison, t9⫽.98, P⫽.17) (Fig. 3). Intracortical inhibition and facilitation analysis. The analysis of intracortical excitability changes for the left and right hemispheres did not show any significant changes for either the left (dominant) hemi-

sphere or the right (nondominant) hemisphere. For ICI analysis, the group ⫻ time interaction was not significant for the left hemisphere (F1,18⫽0.06, P⫽.8) or for the right hemisphere (F1,18⫽0.16, P⫽.69). Similar results were obtained for the ICF analysis: no significant group ⫻ time interaction was found for the left hemisphere (F1,18⫽0.06, P⫽.8) or the right hemisphere (F1,18⫽0.11, P⫽.74). In addition, the term effect was not significant in all of these models, suggesting that ICI and ICF did not change over time regardless of treatment (F⬍1 for all models) (eTab. 2, available at ptjournal.apta. org). Transcallosal inhibition analysis. For TCI, we initially analyzed inhibition from the left (dominant) hemisphere to the right (nondominant) hemisphere (Fig. 4). The mixed ANOVA showed a significant group ⫻ time interaction for left to right inhibition (F1,18⫽6.89, P⫽.017), and the post hoc results showed that TCI significantly decreased by 44.2%⫾

Volume 90

Number 3

Physical Therapy f

403

Interhemispheric Modulation Induced by Cortical Stimulation and Motor Training

Figure 3. Motor-evoked potential (MEP) changes for the active transcranial direct current stimulation (tDCS) group and the sham tDCS group: (A) Change in MEP amplitude from the beginning to the end of the experiment in the dominant hemisphere. The sham tDCS group showed no change in MEP amplitude, whereas the active tDCS group showed a significant decrease in MEP amplitude in the dominant hemisphere. (B) Change in MEP from the beginning to the end of the experiment in the motor cortex (M1) of the nondominant hemisphere. Both groups showed an increase in MEP amplitude; however, the active tDCS group showed a significant increase by the first evaluation and had a greater overall increase compared with the sham tDCS group. Error bars represent the standard error of the mean. Note that the dominant hand (corresponding to the dominant hemisphere of the brain) had the mitt on for the duration of the experiment; therefore, MEP could be assessed only 2 times for the dominant hemisphere. T1⫽measurement taken of nondominant hand after tDCS treatment ended, T2⫽measurement taken of nondominant hand 2 hours into motor training, and T3⫽measurements taken of both dominant and nondominant hands after completion of motor training.

30.5% after active tDCS (t9⫽3.1; P⫽.012, corrected P value). Although TCI (left to right) increased after sham tDCS treatment by 16.5%⫾46.3%, this change was not statistically significant (t9⫽1.4, P⫽ .18). For TCI from the right hemisphere to the left hemisphere, the mixed ANOVA showed that the 404

f

Physical Therapy

Volume 90

group ⫻ time interaction was not significant for either the sham tDCS group or the active tDCS group (F1,18⫽3.42, P⫽.08) (Fig. 4). Finally, because the participants’ sex might have an effect on TCI, we added this variable in the model, and the results showed that this variable was not significant for either model (left to

Number 3

right and right to left TCI, P⬎.05 for both models) (eTab. 3, available at ptjournal.apta.org). Correlations We performed correlations between motor function changes (as indexed by the JTHF) and changes in cortical excitability (changes in TCI from left March 2010

Interhemispheric Modulation Induced by Cortical Stimulation and Motor Training to right hemisphere and from right to left hemisphere and changes in MEP in the left and right hands) for the active and sham tDCS groups (eTab. 4 [available at ptjournal. apta.org] and Fig. 5). For the active tDCS group, there was a significant correlation between changes in motor function (baseline versus T2) and changes in TCI from left to right hemisphere (r⫽⫺.72, P⫽.018), indicating that the changes in motor function in the nondominant hand were associated with a decrease in TCI (Fig. 5) from the dominant hemisphere to the nondominant hemisphere. A significant correlation also was found between changes in MEP in the left hand and motor function improvement in this hand (r⫽.72, P⫽.018). For the sham tDCS group, there was no significant correlation between any of these variables (eTab. 4, available at ptjournal. apta.org). Finally, in order to test whether TCI changes, regardless of the treatment, are associated with motor function changes, we performed similar correlations including both groups together. This analysis showed no significant correlation (r⫽.2, P⫽.39).

Figure 4. Changes in transcallosal inhibition from left to right hemisphere and from right to left hemisphere measured as a percentage of baseline for the active transcranial direct current stimulation (tDCS) group and the sham tDCS group. The only significant change was a decrease in transcallosal inhibition from the left to right hemisphere after active tDCS. Error bars represent the standard error of the mean.

amplitude in the dominant hand. In summary, for both JTHF and MEP assessments, the tDCS active group showed a statistically significant enhancement after the first evaluation,

and the sham tDCS group showed gradual increases over the course of the experiment for the nondominant hand. Transcallosal inhibition from right to left hemisphere in the active

Discussion Our results show that both the sham and active tDCS groups had significant improvements in motor function in the nondominant hand as indexed by the JTHF after 1 day of testing, during which there was 3 hours of motor training. The active tDCS group showed a greater improvement after the first evaluation, whereas the sham tDCS group showed gradual improvement over the course of the experiment. The MEPs recorded from the nondominant hand in both groups also increased similarly in amplitude over the course of this experiment; however, in the active tDCS group, there was a significant decrease in MEP

March 2010

Figure 5. Correlation of the change in Jebsen-Taylor Hand Function Test (JTHF) scores and change in transcallosal inhibition (TCI) from left (dominant) to right (nondominant) hemisphere for the active transcranial direct current stimulation (tDCS) group. As the time necessary to complete the JTHF decreased, the TCI from the left to right hemisphere also decreased.

Volume 90

Number 3

Physical Therapy f

405

Interhemispheric Modulation Induced by Cortical Stimulation and Motor Training and sham tDCS groups and from left to right hemisphere in the sham tDCS group did not result in any statistically significant changes. The active tDCS group, however, showed a statistically significant decrease in TCI from left to right hemisphere. This decrease in TCI, along with the increased amplitude of the MEP in the nondominant hemisphere of the active tDCS group, correlated with increased performance of the JTHF in the nondominant hand. We discuss these findings with the emphasis on the following points: (1) the effects of unilateral motor training with contralateral hand restraint on motor function in participants who were healthy; (2) enhancement of unilateral motor training effects with tDCS; (3) differential mechanisms of action of unilateral motor training with contralateral hand restraint and unilateral motor training with contralateral hand restraint and tDCS, based on the neurophysiological findings of this study; and (4) use of these findings for future research aiming at clinical gains. Our trial testing whether tDCS enhances unilateral motor training with contralateral hand restraint is based on the idea of applying these results in participants poststroke receiving CIMT; therefore, we will discuss the use of tDCS as a potential tool to enhance CIMT in people poststroke. Constraint-induced movement therapy was first studied by Taub et al in primates.32 They observed that after unilateral deafferentation by dorsal rhizotomy, these primates never recovered spontaneous use of the affected upper extremity. Research following this observation showed that the deafferented arm could gain reuse through restriction of the unaffected arm or by retraining the affected extremity.33 The first research using this technique in humans was

406

f

Physical Therapy

Volume 90

performed in 1993 by Taub et al.34 In that experiment, the participants had their unaffected hand restricted using a mitt for 90% of their waking hours. These participants also received training of the affected limb for 6 hours a day over a period of 2 weeks.34 Since this first trial, many other studies have investigated the efficacy of CIMT for improving motor function in the affected limb. In our study, we used the same concept in participants who were healthy. Although participants who are healthy do not have lesions in the corticospinal tract, the increased use of the dominant hand is associated with less dexterity in the nondominant hand and less activity in the nondominant motor cortex35; thus, it might be considered a limited functional lesion. In addition, we conducted only 3 hours of unilateral motor training with contralateral hand restraint (which can be viewed as modified CIMT), as we were interested in acute effects only. The improvement of motor function in both groups that received our unilateral motor training procedure suggested that it was an effective intervention to enhance motor performance in the nondominant hand in participants who were healthy. A meta-analysis36 reported that 14 randomized clinical trials have been performed looking at the use of CIMT to improve upper-extremity function. This meta-analysis concluded that CIMT may improve upper-extremity function better than alternative treatments. In addition, previous research indicates that an increase in motor function is correlated with an increase in cortical excitability in the affected hemisphere.37,38 Similar results were shown in our study, as we observed an increase in MEP amplitude in the nondominant hemisphere in both groups, including the group that received unilateral motor training with contralateral hand re-

Number 3

straint and sham tDCS; that was correlated with improvement in motor function in the active tDCS group. Therefore, the effects of unilateral motor training with contralateral hand restraint in improving motor function seems to be correlated with an increase in the local motor cortex activity, as indexed by an increase in MEP amplitudes. One important finding of our study is that we showed that active tDCS enhances the effects of motor training with contralateral hand restraint on motor function. Participants who received active tDCS had a larger and faster improvement in motor function. Transcranial direct current stimulation is a noninvasive type of brain stimulation that uses a small current to modify neural thresholds in brain regions located directly underneath the electrodes. The anode is associated with an increase in cortical activity, and the cathode is associated with a decrease in cortical activity. In this study, the anode was placed over M1 of the nondominant hemisphere, and the cathode was placed over M1 of the dominant hemisphere; thus, our goal was to increase activity in the nondominant hemisphere and decrease it in the dominant hemisphere. This might be viewed as a form of “central CIMT”—that is, of forcing an increase in excitability in the less used hemisphere (in the case of this study, in the nondominant hemisphere) using tDCS as opposed to the traditional method of CIMT, which induces changes in cortical excitability via peripheral modulation. We have shown previously that this strategy improved motor function in the nondominant hand in participants who were healthy24; however, in that study, no neurophysiological parameter was measured.

March 2010

Interhemispheric Modulation Induced by Cortical Stimulation and Motor Training Furthermore, tDCS seems to be a helpful tool to enhance behavioral changes induced by behavioral training such as motor training. The learning of new skills is linked to changes in neuronal activity and cortical excitability.39 As observed in previous experiments, tDCS is a powerful technique to modulate cortical excitability. Nitsche and Paulus21,40 showed that anodal tDCS applied to the motor cortex in humans produces a prolonged increase in cortical excitability, most likely through subthreshold neuronal membrane depolarization41; thus, it can enhance cortical effects induced by behavioral training, resulting in enhancement of behavioral function. Another advantage is that tDCS modifies spontaneous neuronal activity and, therefore, can increase cortical activity in a more physiological manner, thus priming a given area for a subsequent behavioral intervention such as motor training. Several studies have demonstrated such an effect, for instance, that tDCS improved motor learning in participants who were healthy24 and in participants poststroke.18,19 All of these characteristics support the premise that tDCS may be an attractive tool to enhance motor training. Another important finding of our study is that improvement with unilateral motor training with contralateral hand restraint and sham tDCS is different from that of the same motor training with active tDCS, suggesting different mechanisms of action for both interventions in modulating motor function. When active tDCS is combined with our motor training strategy, improved motor function may be the result of decreased TCI from dominant (left) to nondominant (right) hemispheres, further increasing activity in the nondominant hemisphere, as we showed that TCI was reduced only in the active tDCS group and correlated with motor function changes. A reduction of TCI March 2010

Figure 6. Potential mechanism by which constraint-induced movement therapy affects motor function. The results showed that the direct pathway mechanism of action for training of the nondominant hand might be the only mechanism of motor function improvement.

has been shown previously in participants poststroke after inhibitory, low-frequency, repetitive TMS of the unaffected motor cortex.42 To our knowledge, however, this is the first report of tDCS reducing TCI during training with our modified CIMT procedure (unilateral motor training with contralateral hand restraint) and being associated with motor function changes. In the participants who received sham tDCS, there was no change in TCI; the only change was an increase in amplitude of MEP. This finding suggests that the main mechanism of action of our modified CIMT procedure in participants who were healthy was through direct modulation of the trained hemisphere (affected hemisphere in the case of stroke) and not to the reduction of TCI due to the restriction of the contralateral hand. The only benefit of CIMT, in this context, therefore, seems to be correlated to a forced use of the unrestricted hand.

In summary, there are 2 possible mechanisms through which motor function might be improved: (1) a direct effect (direct pathway) (ie, an increase in cortical excitability in the nondominant hemisphere through increased use of the nondominant hand that results in improved motor function) (Fig. 6) and (2) an indirect effect (ie, an increase in the excitability of the nondominant hemisphere through a decrease in TCI from the dominant hemisphere to the nondominant hemisphere) (Fig. 7). Our study showed that our modified CIMT procedure with sham tDCS was associated only with the first mechanism and, in contrast, that our modified CIMT procedure with active tDCS was associated with both direct and indirect mechanisms (Figs. 6 and 7). This experiment is an important step in the planning of new treatments for survivors of stroke. A meta-

Volume 90

Number 3

Physical Therapy f

407

Interhemispheric Modulation Induced by Cortical Stimulation and Motor Training dress this issue, additional testing would be necessary to calculate the optimal conditioning stimulus for ICI. Increasing testing time would greatly extend the duration and potentially overburden participants. The length of time to complete the experiments was already substantial: approximately 7 to 8 hours per participant. Additionally, there were no tendencies for changes in ICI.

Figure 7. With active transcranial direct current stimulation (tDCS), another mechanism of action takes place: the indirect pathway (ie, decreased use of the dominant hand due to the mitt decreases cortical activity, thus decreasing transcallosal inhibition from dominant to nondominant hemisphere and contributing to the increase in local activity and motor function in the nondominant hand).

analysis assessing studies that used CIMT treatment after stroke concluded that CIMT may provide better treatment as opposed to standard therapies.36 However, the beneficial effects of CIMT are still limited. Transcranial direct current stimulation combined with CIMT may increase improvement in motor function through the combination of: (1) a direct effect on motor cortex excitability in the affected hemisphere induced by both tDCS and CIMT and (2) decreased TCI from the unaffected hemisphere to the affected hemisphere through the use of tDCS. The combination of these 2 techniques also may decrease the amount of time necessary to complete treatment, making it more feasible for patients and therapists. Previous studies and analysis have shown that both therapists and patients had con-

408

f

Physical Therapy

Volume 90

cerns about the length of time for treatment in CIMT.36,43,44 This research has demonstrated differences in the mechanism of action by which unilateral motor training with contralateral hand restraint alone and this same procedure combined with tDCS affect motor improvement in the nondominant hand of participants who are healthy. The increased improvement that was observed during the first evaluation, and throughout the remaining time points of this study, suggests that tDCS combined with CIMT might be a beneficial strategy, and we, therefore, encourage further studies in patients poststroke. Limitations The effects of ICI may have been missed due to a floor effect. To ad-

Number 3

Another potential limitation is the montage we chose, in which both motor cortices are stimulated simultaneously. This montage was not shown to be effective in the study by Nitsche and Paulus.40 However, we have shown in our modeling study20 that this montage induces a significant current in the motor cortex, and the current study demonstrates that active tDCS compared with sham tDCS results in significant changes in motor function and corticospinal excitability. The potential discrepancy here might be due to the duration of stimulation. In the study conducted by Nitsche and Paulus,40 the authors used 1 mA for up to 10 minutes. In our study, we used 1 mA for 40 minutes. Given that duration of stimulation is critical to induce changes in cortical excitability, this difference in duration of stimulation might explain the difference in the results of these 2 studies. Finally, although we found no effects of the participants’ sex for the TCI analysis, this result might have been due to the lack of power of this analysis because of the small sample size and asymmetric sex distribution. A previous TMS study showed that women have a higher TCI compared with men, suggesting that interhemispheric connectivity in the anterior half of the trunk of the corpus callosum might be different between men and women.45

March 2010

Interhemispheric Modulation Induced by Cortical Stimulation and Motor Training Dr Pascual-Leone and Dr Fregni provided concept/idea/research design, fund procurement, and facilities/equipment. All authors provided writing and project management. Ms Williams provided data collection and clerical support. Ms Williams and Dr Fregni provided data analysis. Dr Pascual-Leone provided institutional liaisons. The study was approved by the Human Subjects Review Committee of Harvard Medical School and was performed in accordance with the ethical standards of the 1964 Declaration of Helsinki. This article was received March 3, 2009, and was accepted October 31, 2009. DOI: 10.2522/ptj.20090075

References 1 Curtis HJ. Intercortical connections of corpus callosum as indicated by evoked potentials. J Neurophysiol. 1940;3:407– 413. 2 Hilgetag CC, Theoret H, Pascual-Leone A. Enhanced visual spatial attention ipsilateral to rTMS-induced “virtual lesions” of human parietal cortex. Nat Neurosci. 2001;4:953–957. 3 Chen R. Interactions between inhibitory and excitatory circuits in the human motor cortex. Exp Brain Res. 2004;154:1–10. 4 Oliveri M, Rossini PM, Traversa R, et al. Left frontal transcranial magnetic stimulation reduces contralesional extinction in patients with unilateral right brain damage. Brain. 1999;122(pt 9):1731–1739. 5 Ettlinger G, Morton HB. Callosal section: its effect on performance of a bimanual skill. Science. 1963;139:485– 486. 6 Lo YL, Fook-Chong S. Ipsilateral and contralateral motor inhibitory control in musical and vocalization tasks. Exp Brain Res. 2004;159:258 –262. 7 Kobayashi M, Hutchinson S, Schlaug G, Pascual-Leone A. Ipsilateral motor cortex activation on functional magnetic resonance imaging during unilateral hand movements is related to interhemispheric interactions. Neuroimage. 2003;20: 2259 – 2270. 8 Lee DJ, Chen Y, Schlaug G. Corpus callosum: musician and gender effects. NeuroReport. 2003;14:205–209. 9 Kinsbourne M. The cerebral basis of lateral asymmetries in attention. Acta Psychol (Amst). 1970;33:193–201. 10 Christman SD, Propper RE, Dion A. Increased interhemispheric interaction is associated with decreased false memories in a verbal converging semantic associates paradigm. Brain Cogn. 2004;56:313–319. 11 Kahn I, Pascual-Leone A, Theoret H, et al. Transient disruption of ventrolateral prefrontal cortex during verbal encoding affects subsequent memory performance. J Neurophysiol. 2005;94:688 – 698.

March 2010

12 Bajwa S, Bermpohl F, Rigonatti S, et al. Impaired interhemispheric interactions in patients with major depression. J Nerv Ment Dis. 2008;196:671– 677. 13 Mottaghy FM, Pascual-Leone A, Kemna LJ, et al. Modulation of a brain-behavior relationship in verbal working memory by rTMS. Brain Res Cogn Brain Res. 2003; 15:241–249. 14 Cincotta M, Ziemann U. Neurophysiology of unimanual motor control and mirror movements. Clin Neurophysiol. 2008;119: 744 –762. 15 Murase N, Duque J, Mazzocchio R, Cohen LG. Influence of interhemispheric interactions on motor function in chronic stroke. Ann Neurol. 2004;55:400 – 409. 16 Naeser MA, Martin PI, Nicholas M, et al. Improved picture naming in chronic aphasia after TMS to part of right Broca’s area: an open-protocol study. Brain Lang. 2005;93:95–105. 17 Fregni F, Boggio PS, Valle AC, et al. A sham-controlled trial of a 5-day course of repetitive transcranial magnetic stimulation of the unaffected hemisphere in stroke patients. Stroke. 2006;37: 2115– 2122. 18 Hummel F, Celnik P, Giraux P, et al. Effects of non-invasive cortical stimulation on skilled motor function in chronic stroke. Brain. 2005;128:490 – 499. 19 Fregni F, Boggio PS, Mansur CG, et al. Transcranial direct current stimulation of the unaffected hemisphere in stroke patients. NeuroReport. 2005;16:1551–1555. 20 Wagner T, Fregni F, Eden U, et al. Transcranial magnetic stimulation and stroke: a computer-based human model study. NeuroImage. 2006;30:857– 870. 21 Nitsche MA, Paulus W. Sustained excitability elevations induced by transcranial DC motor cortex stimulation in humans. Neurology. 2001;57:1899 –1901. 22 Floel A, Rosser N, Michka O, et al. Noninvasive brain stimulation improves language learning. J Cogn Neurosci. 2008;20: 1415–1422. 23 Boggio PS, Ferrucci R, Rigonatti SP, et al. Effects of transcranial direct current stimulation on working memory in patients with Parkinson’s disease. J Neurol Sci. 2006;249:31–38. 24 Boggio PS, Castro LO, Savagim EA, et al. Enhancement of non-dominant hand motor function by anodal transcranial direct current stimulation. Neurosci Lett. 2006; 404:232–236. 25 Levy CE, Nichols DS, Schmalbrock PM, et al. Functional MRI evidence of cortical reorganization in upper-limb stroke hemiplegia treated with constraint-induced movement therapy. Am J Phys Med Rehabil. 2001;80:4 –12. 26 Liepert J, Miltner WH, Bauder H, et al. Motor cortex plasticity during constraintinduced movement therapy in stroke patients. Neurosci Lett. 1998;250:5– 8.

27 Jebsen RH, Taylor N, Trieschmann RB, et al. An objective and standardized test of hand function. Arch Phys Med Rehabil. 1969;50:311–319. 28 Gandiga PC, Hummel FC, Cohen LG. Transcranial DC stimulation (tDCS): a tool for double-blind sham-controlled clinical studies in brain stimulation. Clin Neurophysiol. 2006;117:845– 850. 29 Rossini PM, Barker AT, Berardelli A, et al. Non-invasive electrical and magnetic stimulation of the brain, spinal cord and roots: basic principles and procedures for routine clinical application: report of an IFCN committee. Electroencephalogr Clin Neurophysiol. 1994;91:79 –92. 30 Kobayashi M, Pascual-Leone A. Transcranial magnetic stimulation in neurology. Lancet Neurol. 2003;2:145–156. 31 Takano B, Drzezga A, Peller M, et al. Shortterm modulation of regional excitability and blood flow in human motor cortex following rapid-rate transcranial magnetic stimulation. NeuroImage. 2004;23:849 – 859. 32 Taub E. Movement in nonhuman primates deprived of somatosensory feedback. Exerc Sport Sci Rev. 1976;4:335–374. 33 Taub E. Somatosensory deafferentation research with monkeys: implications for rehabilitation medicine. In: Ince LP, ed. Behavioral Psychology in Rehabilitation Medicine: Clinical Applications. New York, NY: Williams & Wilkins; 1980:371– 401. 34 Taub E, Miller NE, Novack TA, et al. Technique to improve chronic motor deficit after stroke. Arch Phys Med Rehabil. 1993;74:347–354. 35 De Gennaro L, Cristiani R, Bertini M, et al. Handedness is mainly associated with an asymmetry of corticospinal excitability and not of transcallosal inhibition. Clin Neurophysiol. 2004;115:1305–1312. 36 Hakkennes S, Keating JL. Constraintinduced movement therapy following stroke: a systematic review of randomised controlled trials. Aust J Physiother. 2005; 51:221–231. 37 Tarkka IM, Kononen M, Pitkanen K, et al. Alterations in cortical excitability in chronic stroke after constraint-induced movement therapy. Neurol Res. 2008;30: 504 –510. 38 Boake C, Noser EA, Ro T, et al. Constraintinduced movement therapy during early stroke rehabilitation. Neurorehabil Neural Repair. 2007;21:14 –24. 39 Facchini S, Muellbacher W, Battaglia F, et al. Focal enhancement of motor cortex excitability during motor imagery: a transcranial magnetic stimulation study. Acta Neurol Scand. 2002;105:146 –151. 40 Nitsche MA, Paulus W. Excitability changes induced in the human motor cortex by weak transcranial direct current stimulation. J Physiol. 2000;527(pt 3): 633– 639.

Volume 90

Number 3

Physical Therapy f

409

Interhemispheric Modulation Induced by Cortical Stimulation and Motor Training 41 Purpura DP, McMurtry JG. Intracellular activities and evoked potential changes during polarization of motor cortex. J Neurophysiol. 1965;28:166 –185. 42 Takeuchi N, Chuma T, Matsuo Y, et al. Repetitive transcranial magnetic stimulation of contralesional primary motor cortex improves hand function after stroke. Stroke. 2005;36:2681–2686.

410

f

Physical Therapy

Volume 90

43 Page SJ, Levine P, Sisto S, et al. Stroke patients’ and therapists’ opinions of constraint-induced movement therapy. Clin Rehabil. 2002;16:55– 60. 44 Wolf SL, Newton H, Maddy D, 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.

Number 3

45 De Gennaro L, Bertini M, Pauri F, et al. Callosal effects of transcranial magnetic stimulation (TMS): the influence of gender and stimulus parameters. Neurosci Res. 2004;48:129 –137.

March 2010

Technical Report McConnell Taping Shifts the Patella Inferiorly in Patients With Patellofemoral Pain: A Dynamic Magnetic Resonance Imaging Study

A. Derasari, is Howard Hughes Medical Fellow, Department of Orthopaedics, University of Miami, Miami, Florida.

Aditya Derasari, Timothy J. Brindle, Katharine E. Alter, Frances T. Sheehan

Background. Patellar taping is widely used clinically to treat patients with patellofemoral pain syndrome (PFPS). Although patellar taping has been demonstrated to reduce patellofemoral pain in patients with PFPS, the kinematic source for this pain reduction has not been identified. Objective. The purpose of this study was to quantify the changes in the 6-degreesof-freedom patellofemoral kinematics due to taping in patients with PFPS.

Design. A within-subject design and a sample of convenience were used. Participants. Fourteen volunteers (19 knees) who were diagnosed with patellofemoral pain that was present for a year or longer were included. Each knee had to meet at least 1 of the following inclusion criteria: Q-angle of ⱖ15 degrees, a positive apprehension test, patellar lateral hypermobility (ⱖ10 mm), or a positive “J sign.” Methods. Each knee underwent 2 randomly ordered testing conditions (untaped and taped). A full fast-phase contrast (PC) magnetic resonance image set was acquired for each condition while the participants volitionally extended and flexed their knee. Three-dimensional displacements and rotations were calculated through integration of the fast-PC velocity data. Statistical comparisons between baseline patellofemoral kinematics and the change in kinematics due to taping were performed using a 2-tailed paired Student t test. Correlations between baseline patellofemoral kinematics and the change in kinematics due to taping also were quantified.

Results. Patellar taping resulted in a significant patellofemoral inferior shift. The strongest correlation existed between the change in lateral-medial displacement with taping and baseline (r⫽⫺.60).

Conclusions. The inferior shift in patellar displacement with taping partially explains the previously documented decrease in pain due to increases in contact area. The lack of alteration in 5 of the 6 kinematic variables with taping may have been due to the fact that post-taping kinematic alterations are sensitive to the baseline kinematic values.

T.J. Brindle, PT, PhD, ATC, is Postdoctoral Candidate, Functional and Applied Biomechanics Section, Rehabilitation Medicine Department, National Institutes of Health, Bethesda, Maryland, a collaboration between the National Institute of Child Health and Human Development and the Clinical Center, National Institutes of Health. K.E. Alter, MD, is Senior Clinician, Functional and Applied Biomechanics Section, Rehabilitation Medicine Department, National Institute of Child Health and Human Development, National Institutes of Health, and Medical Director, Rehabilitation Programs, Mt Washington Pediatric Hospital, Baltimore, Maryland, an affiliate of Johns Hopkins Health System Corp and Maryland Medical System Corp. F.T. Sheehan, PhD, is Staff Scientist, Functional and Applied Biomechanics Section, Rehabilitation Medicine Department, National Institutes of Health, Building 10, CRC Room 1-1469, 10 Center Dr, MSC 1604, Bethesda, MD 208921604 (USA). Address all correspondence to Dr Sheehan at: [email protected]. [Derasari A, Brindle TJ, Alter KE, Sheehan FT. McConnell taping shifts the patella inferiorly in patients with patellofemoral pain: a dynamic magnetic resonance imaging study. Phys Ther. 2010;90:411– 419.] © 2010 American Physical Therapy Association Post a Rapid Response or find The Bottom Line: www.ptjournal.org

March 2010

Volume 90

Number 3

Physical Therapy f

411

Patellar Taping in Patients With Patellofemoral Pain

P

atellofemoral pain syndrome (PFPS) is one of the most common causes of orthopedicrelated physician visits each year, especially among women.1,2 Patellofemoral pain syndrome can be caused by direct trauma to the knee, or the cause can be insidious in nature. Altered lower-extremity biomechanics, such as poor hip rotation control,3 excessive foot pronation,4 femoral anteversion, tibial torsion, bone configuration, or tight muscles are thought to contribute to PFPS by initiating alterations in patellofemoral kinematics.5,6 Vastus medialis oblique muscle dysfunction also has been proposed as a contributor to altered patellofemoral kinematics.7–10 Interventions for PFPS include patellar taping, patellar bracing, selective strengthening of the vastus medalis muscle, iliotibial band stretching, ankle-foot orthotics, or a combination of these interventions.11–15 Patellar taping is widely practiced among clinicians to treat patients with PFPS. This intervention involves pushing the patella medially and securing it in this position with tape on the skin. Originally, the McConnell taping technique was developed to correct altered patellofemoral kinematics and permit participation in normal daily activity.16 Today, there exist several variations of McConnell taping techniques, where attempts are made to alter patellar tilt, glide, or rotation.17 Clinical application of these techniques

Available With This Article at ptjournal.apta.org • Audio Abstracts Podcast This article was published ahead of print on January 28, 2010, at ptjournal.apta.org.

412

f

Physical Therapy

Volume 90

is based on a physical examination of patellar position, orientation, and mobility.18,19 The underlying concept is that most patients with PFPS would benefit from medialization of the patella, which theoretically would off-load the compressive forces at the lateral patellofemoral joint.16,20 Although patellar taping has been demonstrated to reduce patellofemoral pain in patients with PFPS,21–26 the kinematic source of this pain reduction has not been identified. Likely, this failure to identify the kinematic source of the pain reduction is due to the fact that the complete 6 degrees of freedom patellofemoral kinematics (post-taping) during active extension has not been quantified in a group of patients with PFPS. Instead, patellofemoral axial alignment changes (patellar translation and tilt) due to taping in patients with PFPS have been quantified in a small number of studies.21,27–30 The majority of studies focused on static evaluation of knee posture without muscle activity. Isometric quadriceps muscle activity has been incorporated into a few of these studies.21,28 In these studies, no differences in lateral patellar translation or tilt were found with taping, with or without muscle contraction. Thus, the purpose of this study was to quantify the changes in the 6-degrees-of-freedom patellofemoral kinematics due to McConnell taping in people with PFPS during dynamic knee flexion and extension using fast-phase contrast (PC) magnetic resonance imaging (MRI). In doing so, 2 null hypotheses were tested: (1) McConnell taping does not alter the patellofemoral kinematics in all 6 degrees of freedom, and (2) no correlations exist between the baseline patellofemoral kinematics (untaped condition) and the change in kinematics due to taping.

Number 3

Method Participants Fourteen volunteers diagnosed with PFPS gave informed consent prior to participating in this study. Participants were recruited from local orthopedic clinics and ongoing National Institutes of Health studies. This recruitment was primarily Webbased (http://clinicalresearch.nih. gov), conducted through the Clinical Center Patient Recruitment and Public Liaison Office. Participants were excluded if they had any contraindications to having an MRI scan. Both knees were evaluated in the initial screening examination. To be included in the study, each knee had to be clinically diagnosed with PFPS, with symptoms present for a year or longer. An in-house physiatrist, using both a physical examination and the patient’s history, excluded any knee that had: (1) previous surgery (including arthroscopy); (2) ligament, meniscus, iliotibial band, or cartilage damage; (3) other lower-leg pathology or injury; or (4) traumatic onset of PFPS. In addition, a physical therapist (T.J.B.) examined each knee for evidence of altered patellofemoral kinematics: (1) Q-angle of ⱖ15 degrees,18,31,32 (2) a positive apprehension test, (3) patellar lateral hypermobility of ⱖ10 mm,18 or (4) a positive J-sign.18,32,33 Any knee that did not exhibit 1 or more of these 4 signs was excluded. These measures were used solely as inclusion and exclusion criteria and were not included in the kinematic data analysis arm of the study. Both the Q-angle and lateral hypermobility were measured with the participants in a supine position with the knee in full extension and muscles relaxed. In this position, patellar lateral hypermobility was defined as the ability of the examiner to move the patella laterally by 10 mm or more. To confirm or deny the presence of a J-sign, participants were seated at the end of March 2010

Patellar Taping in Patients With Patellofemoral Pain the table and the examiner evaluated patellar motion as the participants raised their leg into full extension. The pain in each knee was evaluated using a visual analog scale (VAS) pain score and a Kujula score.30 In total, 19 knees from 14 participants were included in this study (Tab. 1). Procedure Each knee underwent 2 testing conditions in the MRI (untaped and taped), which were randomly ordered. For each condition, participants were placed supine in a 1.5Tesla magnetic resonance imager (CV-9.1M4 or LX-9.1M4).*,32 After acquiring a set of static alignment scans, the participants were allowed to practice the extension-flexion exercise until they could repeatedly extend and flex their knee to and from maximum attainable flexion and full extension at 35 cycles per minute to the beat of an auditory metronome. Participants were instructed to generate knee extension without hip movement, and a loose strap was placed over the thighs as a reminder, but minor femoral motion was allowed to enable a natural movement. Next, a full fast-PC MRI image set (x,y,z velocity and anatomic images over 24 time frames) was acquired while the participants volitionally extended and flexed their knee to the beat of the auditory metronome, as practiced. In addition, fastcard images (anatomic only) were acquired in 3 axial planes during the cyclic movement. An in-house musculoskeletal radiologist reviewed all magnetic resonance images to rule out ligament, meniscus, iliotibial band, or cartilage damage. During tape application, participants were positioned supine with their quadriceps muscle relaxed. They were taped by a physical therapist (T.J.B.) with more than 15 years of * GE Medical Systems, 4855 W Electric Ave, Milwaukee, WI 53219-1628.

March 2010

Table 1. Demographics and Clinical Intake Parametersa Measure

Values

No. of knees by sex

16 female, 3 male

Age (y)

28.7⫾8.3

Height (cm)

168.5⫾8.2

Weight (kg)

65.4⫾13.5

Visual analog score

36.4⫾27.9

Kujula score

78.8⫾9.2

Q-angle (°)

16.0⫾3.5

Lateral hypermobility (mm)

8.3⫾4.0

J-sign

9 yes, 10 no

a

In total, 19 knees from 14 participants were included in the study. Where appropriate, values are provided as mean⫾standard deviation. The visual analog score and the Kujula score were based on a scale of 100.

experience in orthopedics and sports medicine. Initially, Hypafix† was applied to the skin covering the knee. Next, a medial glide of the patella was obtained by manually pushing the patella medially to its end range of motion. Rigid strapping tape then was used to maintain the medial glide of the patella by pulling the skin and patella medially. Participants were required to walk for 5 minutes at a self-selected, comfortable pace after taping and prior to the MRI experiment.27,29 Three-dimensional rigid-body rotations and displacements of the femur, tibia, and patella were quantified through integration of the fast-PC velocity data. It is important to note that although the fast-PC acquisition was based on a single imaging plane, the 3-dimensional velocity data allowed the orientation and displacement of all bones to be accurately tracked 3-dimensionally throughout the movement.34 Kinematics were defined based on an anatomical coordinate system. Unlike earlier cine MRI experiments,35,36 this identification was completed for a single time frame only, and the fast-PC data were used to track the † Smith & Nephew DonJoy, 2777 Loker Ave, Carlsbad, CA 92010.

bones’ kinematics through all time frames. Thus, the need to visually identify anatomical landmarks at multiple time points35,36 was eliminated, which minimized errors due to inconsistency in image plane location and orientation37,38 and provided excellent accuracy (⬍0.5 mm39) and precision (⬍1.16°40). The kinematics of the patella relative to the femur were defined using 6 variables: 3 displacements—lateralmedial (LM), inferior-superior (IS), and posterior-anterior (PA)—and 3 rotations—lateral-medial tilt (LM tilt), extension-flexion (EF), valgusvarus rotation (VV). A significant increase in the superior position of the patellar origin relative to the femoral origin was defined as patella alta. From the patellar-tibial kinematic data, patellar tendon length was calculated as in a previous study.41 All data were presented for the extension portion of the movement only. Due to individual variations in range of motion within the magnetic resonance scanner, not all participants were represented at the extreme ranges of knee flexion and extension. Thus, the average data presented were limited to the central range of motion, where 3 or more participants were represented by the average.

Volume 90

Number 3

Physical Therapy f

413

Patellar Taping in Patients With Patellofemoral Pain

Figure 1. Patellofemoral kinematics in the untaped and taped conditions: Each variable is plotted relative to knee extension, with full extension being equal to zero (participant movement is read from left to right). Data were interpolated to 1-degree knee angle increments, but symbols are provided at every 5 degrees for clarity. One standard deviation bars are provided. The dotted line represents a knee extension angle of 8 degrees. The top row contains displacements (medial, superior, and anterior are the positive directions, respectively), and the bottom row contains the rotations (flexion, medial tilt, and varus rotation are the positive directions, respectively). ⌬disp⫽changed in displacement.

The kinematic data were reduced to the magnitudes of each of the 6 kinematic variables when the knee was at 10 degrees of knee extension (defined as the “value” of that variable). The untaped condition was considered baseline. A 10-degree knee flexion angle was used because not all participants were able to achieve full knee extension (0° flexion angle). In addition, patellar taping should have the greatest effect when the patella is in terminal extension and is most free from the constraints of the femoral groove.5,35,42

investigated using a paired 2-tailed Student t test. Correlations between the baseline patellofemoral kinematics and the change in kinematics due to taping were quantified. Minimal correlations between pain scores (VAS and Kujula) and patellofemoral kinematics have been documented (only the slope of patellofemoral varus relative to knee extension correlated with VAS pain score) in patients with PFPS.5 Because the current population was a subset of this earlier study, these correlations were not performed in the current study. Statistical significance was set

Data Analysis Statistical differences between the taped and untaped conditions were 414

f

Physical Therapy

Volume 90

Number 3

at Pⱕ.05, (SPSS, version 14.0).‡ An a priori sample size calculation determined that adequate power (␣⫽.05, ␤ⱖ.80) could be achieved with the inclusion of 15 knees, assuming that taping would result in a 2-mm medial patellar shift30 and that the standard deviation associated with this change would be 2.5 mm.34

Results The null hypothesis that McConnell taping does not alter the patellofemoral kinematics was rejected for a single variable only: IS displacement (Fig. 1, ⫺1.6 mm, P⫽.04). The ‡ SPSS Inc, 233 S Wacker Dr, Chicago, IL 60606.

March 2010

Patellar Taping in Patients With Patellofemoral Pain Table 2. Patellofemoral Kinematic Parametersa Displacement (mm)

Lateral-Medial

Inferior-Superior

Anterior-Posterior

Participants who were asymptomatic

⫺0.2⫾2.5

20.7⫾6.0

9.7⫾3.9

Participants with PFPS

⫺2.6⫾4.6

25.4⫾6.5

7.3⫾3.0

⫺0.7⫾2.7 P⫽.27

⫺1.6⫾3.1 P⫽.04

⫺0.3⫾2.3 P⫽.56

Rotation (°)

ExtensionFlexion

Lateral-Medial Tilt

Valgus-Varus

Participants who were asymptomatic

7.0⫾4.0

14.9⫾7.3

12.6⫾4.6

12.1⫾10.6

⫺2.1⫾2.5

0.0⫾3.4 P⫽.74

⫺0.2⫾6.0 P⫽.24

0.3⫾3.0 P⫽.53

Change with taping for participants with PFPS

Participants with PFPS Change with taping for participants with PFPS

0.5⫾2.3

a Values for patellofemoral displacements (mm) and rotations (°) are reported as mean⫾standard deviation for participants who were asymptomatic from a previous study,34 for the participants with patellofemoral pain syndrome (PFPS), and for the change in patellofemoral kinematics with taping in participants with PFPS. For the change in patellofemoral kinematics with taping, the P value is reported for the 2-tailed paired Student t test comparing the taped and untaped values of patellofemoral kinematics.

only significant effect of patellar taping was an inferior (downward) shift in the patella. This inferior shift resulted in an insignificant decrease in the patellar tendon length at 10 degrees of knee extension (⌬length ⫽⫺0.46, P⫽.79). However, there were significant differences in kinematics between the participants with PFPS included in the current study and a previous sample of individuals who were asymptomatic.34 On average, the participants with patellofemoral pain were 2.1 mm (P⫽.04), 4.2 mm (P⫽.03), and 2.3 mm (P⫽.05) more laterally, superiorly, and anteriorly displaced than the control population. In addition, these participants were 5.3 (P⬍.001) and 2.5 (P⬍.001) degrees more flexed and rotated into valgus than the control population (Tab. 2).

kinematic values (further from the asymptomatic average34) would see the greatest absolute change with taping. Interestingly, not all of the participants with patellofemoral pain had baseline values of LM displacement, LM tilt, and VV that were lateral and valgus, relative to a previously defined asymptomatic average.34 Thus, participants who began with medial displacement, medial tilt, or varus rotation, relative to the asymptomatic population, demonstrated a lateral shift in displacement, lateral tilt, or valgus rotation, respectively, after taping was applied. The participant with the largest values of

March 2010

Discussion This study was the first to quantify changes caused by McConnell taping in the complete 6-degrees-of-

Table 3. Correlations Between the Baseline Patellofemoral Kinematics (Untaped Condition) and the Change in Kinematics Due to Taping Parameter

The second null hypothesis that no correlations exist between the baseline patellofemoral kinematics and the change in kinematics due to taping was rejected for 3 variables (Tab. 3 and Fig. 2). The strongest inverse correlation existed for LM displacement (r⫽⫺.67), and moderate inverse correlations existed for LM tilt and VV (r⫽⫺.52 for both). The correlations indicate that participants with larger absolute baseline

lateral tilt and displacement proved to be an outlier in that taping caused a much smaller change in these variables than the correlation would suggest. Data for this participant were not removed from the analyses, but when these data are removed, the correlation between baseline and change with taping increases (r⫽⫺.54, P⫽.02) for LM tilt and the correlations for LM displacement and VV rotation increase minimally.

ra

P

⫺.60

.006b

Regression Equation

Displacement Lateral-medial

⌬⫽⫺0.31 ⫻ value ⫺ 1.81

Inferior-superior

⫺.19

.431

⌬⫽⫺0.06 ⫻ value ⫹ 0.15

Posterior-anterior

⫺.21

.385

⌬⫽⫺0.27 ⫻ value ⫹ 1.48

Extension-flexion

.02

.734

Lateral-medial tilt Valgus-varus

Rotation

a b c

⌬⫽0.08 ⫻ value ⫺ 0.26

⫺.51

.033

c

⌬⫽⫺0.26 ⫻ value ⫹ 2.88

⫺.52

.004b

⌬⫽⫺0.74 ⫻ value ⫺ 1.24

Pearson correlation coefficient. Significant at P⬍.01. Significant at P⬍.05.

Volume 90

Number 3

Physical Therapy f

415

Patellar Taping in Patients With Patellofemoral Pain

Figure 2. Change in kinematics with taping (taped and untaped conditions) versus baseline. The 3 variables with significant correlations between the change with taping and the baseline value are shown (top row: lateral to medial displacement and lateral to medial tilt; bottom row: valgus to varus rotation and legend to symbols). The x-axis indicates the untaped value of each variable (positive values indicate medial displacement, medial tilt, and varus rotation). The y-axis indicates the change in each variable with taping (positive values indicate medial displacement, medial tilt, and varus rotation with taping, shown by gray shading). The regression lines are shown for each variable set using a thick black line. Based on previous work,34 the average value for an asymptomatic cohort (untaped, n⫽34) for each of the 3 variables is shown using a vertical dashed line (medial displacement⫽⫺0.11 mm, medial tilt⫽14.7, varus rotation⫽0.52°). All graphs represent kinematics at 10 degrees of knee extension. Participants who were laterally translated, laterally tilted, and in valgus rotation (relative to the asymptomatic population) are shown in blue. This distinction is independent for each graph. Thus, a single participant who demonstrated medial displacement and lateral tilt at baseline would be designated by 2 different colors in each of the respective graphs.

416

f

Physical Therapy

Volume 90

Number 3

March 2010

Patellar Taping in Patients With Patellofemoral Pain freedom patellofemoral kinematics during volitional knee extension in a sample of patients with PFPS. On average, the only consistent influence McConnell taping provided on patellar kinematics was to shift the patella inferiorly. The benefit of inferiorly shifting the patella within the femoral groove was an increase in the patellofemoral contact area. In a previous study, Ward and colleagues43 defined a 19.3% decrease in cartilage contact area in a group of participants with a 24% increase in the Insall ratio (superior displaced relative to a control population). In the current study, the patellae in the participants with PFPS were 24% (5.2 mm) more superiorly placed (dynamically) compared with those of the asymptomatic population.34 The inferior shift in displacement with taping (1.6 mm) represents 34.8% of the difference in IS displacement between the participants with PFPS and the asymptomatic controls (Tab. 2, 4.6 mm). Increased contact area decreases contact stress at the patellofemoral joint by spreading out the load over a wider area. Ultimately, this decrease in contact stress may account for some of the previously reported decrease in patellofemoral pain after taping.21–26 The correlation between the baseline patellofemoral kinematics and the change in kinematics due to taping supports earlier work that identified the existence of subgroups within the patellofemoral pain population.33 The expected medial change in patellar displacement or tilt occurred in participants with lateral baseline values of these variables. Conversely, participants with medial displacement or tilt at baseline demonstrated a lateral change in displacement or tilt with taping. The same relationship held true for VV rotation. Therefore, taping shifts the patella medially in patients with lateral baseline values of patellofemoral displacement (relative to an asympMarch 2010

tomatic average), and it shifts the patella in the opposite direction in patients with medial baseline values of patellofemoral displacement. Similarly, taping is effective in medially tilting the patella in patients with PFPS and lateral baseline tilt, but it has the opposite effect for patients with medial baseline tilt. However, average differences in patellar kinematics after taping across all participants produced no significant change due to the contrary effect of patellar taping across participants. Thus, considering the participants in this study as a single population masked the true changes with taping seen on a person-by-person basis. Regardless, pain relief may have been further enhanced by the centralization of the patella in regard to patellar LM displacement, LM tilt, and VV rotation. One interesting exception in this study was a single participant who had the largest lateral displacement at baseline (⫺14.5 mm), yet demonstrated minimal medialization with taping. This participant had the highest baseline values of superior displacement (39.0 mm, asymptomatic average [⫾SD]⫽21.7⫾6.3 mm). If the inferior shift in patellar placement is considered the primary effect of taping, then the lack of kinematic change for this specific participant may have been due to patella alta. In general, for most patients with PFPS with lateral baseline values of patellofemoral LM displacement, the inferior shift and taping may lead to further patellar engagement with the femoral sulcus. Thus, both the force of the taping and the bone constraint may have medialized the patella placement. This concept is supported by a previous study that demonstrated a strong correlation between patellofemoral superior and lateral displacement.5 Yet, for this participant, the small inferior shift may not have resulted in patellar engagement with the sulcus, and tap-

ing alone was insufficient to cause a large medialization force. It is counterintuitive that a medial taping technique could cause lateralization of the patella in patients with PFPS. Yet, patients who begin with medial baseline values for LM displacement do not demonstrate patella alta.33 In the absence of patella alta, it is likely that the patella remained partially engaged with the femoral sulcus throughout extension, enabling the patella to remain medially placed. When the patella was inferiorly displaced with taping, the patella was further engaged with the sulcus. The bony constraints of the sulcus, due to either the shape of the groove or limb alignment, resulted in the patella shifting laterally relative to the femur, and taping could not overcome this bone force. The fact that any patient with PFPS had medial patellar placement relative to the femur may appear contrary to the fact that the clinical markers used as inclusion criteria should have confirmed the presence of excessive lateral tracking. This potential incongruity is explained by a recently published finding that the value of Q-angle is correlated with medial and not lateral patellofemoral displacement.33 The current results support and expand the findings of previous studies on the effectiveness of McConnell taping. Typically, previous taping studies focused on a static, 2-dimensional evaluation of patellar alignment with or without muscle contraction.21,27–30 On average, taping was shown to be ineffective in shifting and tilting the patella medially. Taping has been shown to be effective if it was evaluated prior to any exercise, but a short bout of exercise eliminated this effect.24,29 The current study design included a short bout of exercise prior to evaluating the taping condition, but our study differed from past studies in its

Volume 90

Number 3

Physical Therapy f

417

Patellar Taping in Patients With Patellofemoral Pain 3-dimensional dynamic evaluation of a volitional task. Based on this experimental design, McConnell taping was shown to be effective in shifting the patella inferiorly after a short bout of exercise. We also demonstrated that taping tended to shift patellar LM displacement, LM tilt, and VV rotation back toward the asymptomatic average. Thus, it is likely that the ineffectiveness of taping found in previous studies was due to the fact that some patients were not laterally tilted or displaced prior to taping. Although the current study used dynamic movement, which is representative of how muscle action can influence patellar kinematics, there were limitations. The movements were isolated to the knee, neglecting the influence of proximal or distal joints on patellofemoral kinematics. The range of motion was limited, but did cover from just prior to and through terminal extension, where altered patellofemoral kinematic patterns are most evident. The experimental design did not include an axial compressive load applied to the leg; thus, it is not known how such loading could affect patellar kinematics and taping. Another potential limitation of this study was the fact that the exact force applied with taping was not quantified. Yet, in previous studies, participants often applied their own tape,21,44 and no previous study measured the force applied with taping.17,21,24,29,30,44,45 Lastly, pain was not measured after tape application. This measure was not included in the study design, as previous studies21–26 have clearly shown the effectiveness of taping for pain reduction. Thus, the current study was focused specifically on accurately quantifying how taping alters 3-dimensional patellofemoral kinematics by measuring them noninvasively and in vivo during volitional activity. The noninvasive nature of this experiment and its high accu418

f

Physical Therapy

Volume 90

racy, excellent precision, and ability to demonstrate 3-dimensional kinematic differences during volitional activities justify these limitations.

Radiology Department at the National Institutes of Health for their support and research time.

Conclusions and Clinical Relevance

This research was supported, in part, by the Howard Hughes Foundation and the Intramural Research Program of the National Institutes of Health (Clinical Center and National Institute of Child Health and Human Development). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Institutes of Health or the US Public Health Service.

This is the first study to evaluate 3-dimensional changes in patellar kinematics with McConnell taping during voluntary active motion. For the participants in this study, patellar taping shifted the patella inferiorly with respect to the femur, which may account for some of the previously reported pain relief with taping, due to an increase in contact area. McConnell taping medialized the patella in participants who demonstrated lateral displacement at baseline and lateralized the patella in participants who demonstrated medial patellar displacement at baseline. Thus, medialization of the patella with taping was dependent upon the patellofemoral kinematic alterations present in each participant. This finding reinforces the need to clinically identify the specific alterations in patellofemoral kinematics present in each patient so that specific interventions can be used and optimized to correct these altered kinematics and reduce pain. All patients with PFPS and altered patellofemoral kinematics are likely to benefit from an inferior shift of the patella, which should reduce patellofemoral contact stress. Ms Derasari, Dr Alter, and Dr Sheehan provided concept/idea/project design, data analysis, and project management. Dr Brindle, Dr Alter, and Dr Sheehan provided writing. All authors provided data collection. Dr Sheehan provided fund procurement. Ms Derasari provided participants. Dr Brindle provided consultation (including review of manuscript before submission). The authors thank Elizabeth K. Rasch, PT, PhD, for support on the statistical analysis and Steven Stanhope, PhD, for guidance throughout the project. They also thank Bonnie Damaska, Jamie Fraunhaffer, Jere McLucas, Dr Ken Fine, and the Diagnostic

Number 3

This study was approved by the National Institute of Child Health and Development.

This article was received November 18, 2008, and was accepted October 31, 2009. DOI: 10.2522/ptj.20080365

References 1 Almeida SA, Trone DW, Leone DM, et al. Gender differences in musculoskeletal injury rates: a function of symptom reporting? Med Sci Sports Exerc. 1999;31:1807– 1812. 2 Rauh MJ, Koepsell TD, Rivara FP, et al. Epidemiology of musculoskeletal injuries among high school cross-country runners. Am J Epidemiol. 2006;163:151– 159. 3 Cibulka MT, Threlkeld-Watkins J. Patellofemoral pain and asymmetrical hip rotation. Phys Ther. 2005;85:1201–1207. 4 Sutlive TG, Mitchell SD, Maxfield SN, et al. Identification of individuals with patellofemoral pain whose symptoms improved after a combined program of foot orthosis use and modified activity: a preliminary investigation. Phys Ther. 2004;84: 49 – 61. 5 Sheehan FT, Derasari A, Brindle TJ, Alter KE. Understanding patellofemoral pain with maltracking in the presence of joint laxity: complete 3D in vivo patellofemoral and tibiofemoral kinematics. J Orthop Res. 2008;27:561–570. 6 Wilson NA, Press JM, Koh JL, et al. In vivo and noninvasive evaluation of abnormal patellar tracking during squatting in patellofemoral pain. J Bone Joint Surg Am. 2008;91:558 –566. 7 Boucher JP, King MA, Lefebvre R, Pepin A. Quadriceps femoris muscle activity in patellofemoral pain syndrome. Am J Sports Med. 1992;20:527–532. 8 Cowan SM, Bennell KL, Hodges PW, et al. Delayed onset of electromyographic activity of vastus medialis obliquus relative to vastus lateralis in subjects with patellofemoral pain syndrome. Arch Phys Med Rehabil. 2001;82:183–189.

March 2010

Patellar Taping in Patients With Patellofemoral Pain 9 Karst GM, Willett GM. Onset timing of electromyographic activity in the vastus medialis oblique and vastus lateralis muscles in subjects with and without patellofemoral pain syndrome. Phys Ther. 1995;75:813– 823. 10 Powers CM. Patellar kinematics, part I: the influence of vastus muscle activity in subjects with and without patellofemoral pain. Phys Ther. 2000;80:956 –964. 11 Adams WB. Treatment options in overuse injuries of the knee: patellofemoral syndrome, iliotibial band syndrome, and degenerative meniscal tears. Curr Sports Med Rep. 2004;3:256 –260. 12 Crossley K, Bennell K, Green S, et al. Physical therapy for patellofemoral pain: a randomized, double-blinded, placebocontrolled trial. Am J Sports Med. 2002; 30:857– 865. 13 Gilleard W, McConnell J, Parsons D. The effect of patellar taping on the onset of vastus medialis obliquus and vastus lateralis muscle activity in persons with patellofemoral pain. Phys Ther. 1998;78:25–32. 14 Moller BN, Krebs B. Dynamic knee brace in the treatment of patellofemoral disorders. Arch Orthop Trauma Surg. 1986; 104:377–379. 15 Sathe VM, Ireland ML, Ballantyne BT, et al. Acute effects of the Protonics system on patellofemoral alignment: an MRI study. Knee Surg Sports Traumatol Arthrosc. 2002;10:44 – 48. 16 McConnell J. The management of chondromalacia patellae: a long term solution. Aust J Physiother. 1986;32:215–223. 17 Crossley K, Cowan SM, Bennell KL, McConnell J. Patellar taping: is clinical success supported by scientific evidence? Man Ther. 2000;5:142–150. 18 Fredericson M, Yoon K. Physical examination and patellofemoral pain syndrome. Am J Phys Med Rehabil. 2006;85:234 – 243. 19 LaBella C. Patellofemoral pain syndrome: evaluation and treatment. Prim Care. 2004;31:977–1003. 20 Hergenroeder AC. Prevention of sports injuries. Pediatrics. 1998;101:1057–1063. 21 Bockrath K, Wooden C, Worrell T, et al. Effects of patella taping on patella position and perceived pain. Med Sci Sports Exerc. 1993;25:989 –992. 22 Christou EA. Patellar taping increases vastus medialis oblique activity in the presence of patellofemoral pain. J Electromyogr Kinesiol. 2004;14:495–504.

March 2010

23 Salsich GB, Perman WH. Patellofemoral joint contact area is influenced by tibiofemoral rotation alignment in individuals who have patellofemoral pain. J Orthop Sports Phys Ther. 2007;37:521–528. 24 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. 25 Warden SJ, Hinman RS, Watson MA Jr, et al. Patellar taping and bracing for the treatment of chronic knee pain: a systematic review and meta-analysis. Arthritis Rheum. 2008;59:73– 83. 26 Whittingham M, Palmer S, Macmillan F. Effects of taping on pain and function in patellofemoral pain syndrome: a randomized controlled trial. J Orthop Sports Phys Ther. 2004;34:504 –510. 27 Larsen B, Andreasen E, Urfer A, et al. Patellar taping: a radiographic examination of the medial glide technique. Am J Sports Med. 1995;23:465– 471. 28 Gigante A, Pasquinelli FM, Paladini P, et al. The effects of patellar taping on patellofemoral incongruence. A computed tomography study. Am J Sports Med. 2001; 29:88 –92. 29 Pfeiffer RP, DeBeliso M, Shea KG, et al. Kinematic MRI assessment of McConnell taping before and after exercise. Am J Sports Med. 2004;32:621– 628. 30 Worrell T, Ingersoll CD, Bockrath-Pugliese K, Minis P. Effect of patellar taping and bracing on patellar position as determined by MRI in patients with patellofemoral pain. J Athl Train. 1998;33:16 –20. 31 Bourne MH, Hazel WA Jr, Scott SG, Sim FH. Anterior knee pain. Mayo Clin Proc. 1988;63:482– 491. 32 Post WR. Clinical evaluation of patients with patellofemoral disorders. Arthroscopy. 1999;15:841– 851. 33 Sheehan FT, Derasari A, Fine KM, et al. Q-angle and J-sign: indicative of maltracking subgroups in patellofemoral pain. Clin Orthop Relat Res. 2009 May 9. [Epub ahead of print.] 34 Seisler A, Sheehan FT. Normative threedimensional patellofemoral and tibiofemoral kinematics: a dynamic, in vivo study. IEEE Trans Biomed Eng. 2007;54:1333– 1341.

35 Brossmann J, Muhle C, Schroder C, et al. Patellar tracking patterns during active and passive knee extension: evaluation with motion-triggered cine MR imaging. Radiology. 1993;187:205–212. 36 Draper CE, Besier TF, Santos JM, et al. Using real-time MRI to quantify altered joint kinematics in subjects with patellofemoral pain and to evaluate the effects of a patellar brace or sleeve on joint motion. J Orthop Res. 2009;27:571–577. 37 Shibanuma N, Sheehan FT, Lipsky PE, Stanhope SJ. Sensitivity of femoral orientation estimates to condylar surface and MR image plane location. J Magn Reson Imaging. 2004;20:300 –305. 38 Shibanuma N, Sheehan FT, Stanhope SJ. Limb positioning is critical for defining patellofemoral alignment and femoral shape. Clin Orthop Relat Res. 2005;(434): 198 –206. 39 Sheehan FT, Zajac FE, Drace JE. Using cine phase contrast magnetic resonance imaging to non-invasively study in vivo knee dynamics. J Biomech. 1998;31:21–26. 40 Rebmann AJ, Sheehan FT. Precise 3D skeletal kinematics using fast phase contrast magnetic resonance imaging. J Magn Reson Imaging. 2003;17:206 –213. 41 Sheehan FT, Drace JE. Human patellar tendon strain (a non-invasive, in vivo study). Clin Orthop Relat Res. 2000;370:201–207. 42 Powers CM. Patellar kinematics, part II: the influence of the depth of the trochlear groove in subjects with and without patellofemoral pain. Phys Ther. 2000;80:965– 978. 43 Ward SR, Terk MR, Powers CM. Patella alta: association with patellofemoral alignment and changes in contact area during weight-bearing. J Bone Joint Surg Am. 2007;89:1749 –1755. 44 Kowall MG, Kolk G, Nuber GW, et al. Patellar taping in the treatment of patellofemoral pain. A prospective randomized study. Am J Sports Med. 1996;24:61– 66. 45 Bennell K, Duncan M, Cowan S. Effect of patellar taping on vasti onset timing, knee kinematics, and kinetics in asymptomatic individuals with a delayed onset of vastus medialis oblique. J Orthop Res. 2006;24: 1854 –1860.

Volume 90

Number 3

Physical Therapy f

419

Case Report

Physical Therapy in the Emergency Department: Development of a Novel Practice Venue Debra Fleming-McDonnell, Sylvia Czuppon, Susan S. Deusinger, Robert H. Deusinger D. Fleming-McDonnell, PT, DPT, is Assistant Professor, Physical Therapy and Orthopaedic Surgery, Program in Physical Therapy, Washington University School of Medicine, Campus Box 8502, 4444 Forest Park Blvd, Suite 1101, St Louis, MO 63108 (USA). Address all correspondence to Dr Fleming-McDonnell at: flemingd @wusm.wustl.edu. S. Czuppon, PT, MS, is Instructor, Physical Therapy and Orthopaedic Surgery, Program in Physical Therapy, Washington University School of Medicine. S.S. Deusinger, PT, PhD, FAPTA, is Professor, Physical Therapy and Neurology, Program in Physical Therapy, Washington University School of Medicine. R.H. Deusinger, PT, PhD, is Associate Professor, Physical Therapy, Medicine, and Orthopaedic Surgery, Program in Physical Therapy, Washington University School of Medicine. [Fleming-McDonnell D, Czuppon S, Deusinger SS, Deusinger RH. Physical therapy in the emergency department: development of a novel practice venue. Phys Ther. 2010;90:420 – 426.]

Background and Purpose. The American Physical Therapy Association’s Vision 2020 advocates that physical therapists be integral members of health care teams responsible for diagnosing and managing movement and functional disorders. This report details the design and early implementation of a physical therapist service in the emergency department (ED) of a large, urban hospital and presents recommendations for assessing the effectiveness of physical therapists in this setting. Case Description. Emergency departments serve multiple purposes in the American health care system, including care of patients with non–life-threatening illnesses. Physical therapists have expertise in screening for problems that are not amenable to physical therapy and in addressing a wide range of acute and chronic musculoskeletal pain problems. This expertise invites inclusion into the culture of ED practice. This administrative case report describes planning and early implementation of a physical therapist practice in an ED, shares preliminary outcomes, and provides suggestions for expansion and effectiveness testing of practice in this novel venue. Outcomes. Referrals have increased and length of stay has decreased for patients receiving physical therapy. Preliminary surveys suggest high patient and practitioner satisfaction with physical therapy services. Outpatient physical therapy follow-up options were developed. Educating ED personnel to triage patients who show deficits in pain and functional mobility to physical therapy has challenged the usual culture of ED processes.

Discussion. Practice in the hospital ED enables physical therapists to fully use their knowledge, diagnostic skills, and ability to manage acute pain and musculoskeletal injury. Recommendations for future action are made to encourage more institutions across the country to incorporate physical therapy in EDs to enhance the process and outcome of nonemergent care.

© 2010 American Physical Therapy Association

Post a Rapid Response or find The Bottom Line: www.ptjournal.org 420

f

Physical Therapy

Volume 90

Number 3

March 2010

Physical Therapy in the Emergency Department

T

he hospital emergency department (ED) has become a common entry point into the health system for individuals with urgent, but noncritical, care needs.1 Estimates project that more than 80% of people seen in EDs have non–lifethreatening conditions, many involving chronic pain.2 Physical therapy intervention in the ED has been suggested to positively influence patient satisfaction and pain management for acute low back pain (LBP)3 and to shorten waiting time for referral to outpatient care.4 However, few studies have demonstrated the impact physical therapy could have in managing nonemergent patient cases, thus reducing unnecessary hospital admissions, costs, waiting time, elopement and frequent returns, and improvement of patient satisfaction and outcomes. Although physical therapist practice in the ED has been reported in a few locations in the United States,4 most reports regarding this practice paradigm are from other countries.3– 6 Traditionally, EDs have relied exclusively on nurses and physicians (MDs) whose short-term provider relationship with patients and training for emergent care may make managing acute and chronic pain difficult.7 As new roles have emerged in health care, advanced nurse practitioners (NPs) and physician assistants (PAs) have been integrated into the ED culture to improve care of patients with nonemergent conditions. Griffin and Melby8 demonstrated that NPs could be integrated effectively into the ED provider team as long as roles and responsibilities were clear and education and experience sufficiently enabled competence in this complex environment. Similarly, incorporating physical therapists into this setting requires the same careful personnel selection and role delineation. It also offers opportunities to enhance satisfaction of patients with

March 2010

nonemergent conditions who are seen in the ED. Because physicians’ education focuses on diagnosing medical illness, MDs may not be adequately prepared to manage musculoskeletal conditions without prescribing medications or surgery. This situation may be exaggerated in the ED setting, where quickly relieving symptoms and determining referral routes to fully address patients’ musculoskeletal problems are imperative.7 As recently as 2003, DiCaprio et al9 documented that nearly 50% of the 122 US medical schools required no training in musculoskeletal medicine. Childs et al10 showed that physical therapists are better prepared to manage common musculoskeletal conditions seen in primary care than other medical practitioners, except orthopedic surgeons, who typically provide only consultation in most EDs. In any ED, it is essential to rapidly identify primary movement impairments and provide specific interventions to relieve pain and improve function. Studies support that physical therapists can be effective and safe in collaborating with other primary care team members in diagnosing and managing musculoskeletal and neuromuscular disorders.11 This creates an ideal opportunity for partnership with other ED providers. Managing pain conditions in the ED can become costly, especially because these conditions may result in multiple ED visits by patients needing more than a short-term solution.12 Jorgensen13 suggested that costs associated with ED management of nonmalignant back pain may be unnecessarily high, especially for patients returning repeatedly for the same condition, and concluded that ED physicians may not be sufficiently prepared to address the functional problems associated with acute or chronic pain. In a retrospective study of data from the National

Hospital Ambulatory Medical Care Survey (NHAMCS), Isaacs et al14 found that 17.8% of patients with uncomplicated cases of LBP received unnecessary radiographs in the ED. Some authors have suggested that physical therapy intervention may be more cost-effective if more expensive options are avoided.11,15–17 Daker-White et al11 found the costs of managing specific musculoskeletal conditions by physical therapy to be less than if care was provided by an orthopedic surgeon because fewer radiographs were ordered or fewer referrals for surgery were made. Patients with chronic pain may wait longer in the ED due to their lower triage priority,18 an indirect health care cost. Time limitations felt throughout the ED, attitudes toward patients who return repeatedly, and limited primary care options outside the ED may cause tests to be ordered or pain medications prescribed inside the ED as short-term solutions to patients’ symptoms. These conditions invite including physical therapists into the provider team managing the myriad conditions seen in busy EDs. Historically, physical therapists may have not initiated hospital ED services because of 2 concerns. The first concern is that serious medical conditions could be overlooked without MD involvement. Contemporary practice requires physical therapists to screen for conditions not amenable to physical therapy intervention by identifying signs of medical pathology that do not fit the

Available With This Article at ptjournal.apta.org • Audio Abstracts Podcast This article was published ahead of print on January 7, 2010, at ptjournal.apta.org.

Volume 90

Number 3

Physical Therapy f

421

Physical Therapy in the Emergency Department patterns of musculoskeletal impairments.19 Stowell et al20 showed that physical therapists can independently distinguish medical conditions from problems of musculoskeletal origin and manage these pain problems in first-contact situations. Physical therapists have been able to associate complaints of LBP with medical pathologies, including endometriosis,21 hip pain with lymphoma,22 and hip pain and weakness with cervical cord compression.23 Each case resulted in referral to a more appropriate practitioner. The second concern is that patients would be at greater risk for adverse events without being first screened by a physician. However, Moore et al24 found no such risk in a pool of 50,799 patients. There were no reports of patient injury, adverse events, disciplinary action, revocation of licensure, or litigation. These results increase the confidence that physical therapists could be productive ED providers. Managing nonemergent acute and chronic pain is a primary obligation for physical therapists.19 A 2005 NHAMCS report documented a 23% increase in hospital ED visits over 10 years, despite a 15% decrease in the number of EDs operating nationally.1 Patients with musculoskeletal sprains, strains, and neck and back injuries accounted for 13.9% of ED visits, a 2% increase from 2002.1 These data suggest that EDs likely care for numerous patients with conditions appropriate for physical therapy intervention. Using physical therapy in the ED increases patient satisfaction with management of LBP3 and other musculoskeletal conditions compared to when NPs or MDs are involved.12,25 Overall waiting times have been shown to decrease, even though patients may spend more time receiving care from a physical therapist.25 When physical therapy is provided 422

f

Physical Therapy

Volume 90

in the ED, patients are more likely to be referred for further outpatient care,4 creating the possibility of earlier return to work.26 At least one study showed that outcomes of physical therapy intervention in the ED can last beyond the single intervention provided there. McClellan et al25 showed that improved function and decreased pain persisted 1 month after physical therapy intervention in the ED. However, neither Lau et al3 nor Richardson et al12 found such benefits lasted beyond an acute phase, even though satisfaction with the one-time encounter was high. These findings suggest that managing acute and chronic pain is a continuous process requiring not only episodic care in the ED but also appropriate referral to achieve follow-up.18,27 The primary purpose of this administrative case report is to describe the process of establishing a physical therapy service in a busy urban ED. Preliminary outcomes and recommendations for further assessment of physical therapy impact on ED cost of care, length of stay, pain, and patient and practitioner satisfaction are presented.

Target Setting After several years of planning, we initiated a demonstration project to evaluate the feasibility of physical therapy services in the Barnes-Jewish Hospital (BJH) Emergency Department in St Louis, Missouri.28 BarnesJewish Hospital is part of the Washington University Medical Center, which includes several collaborative components. Washington University School of Medicine (WUSM) provides all MDs, NPs, and PAs for this ED. All other personnel (eg, nurses, residents, orderlies) are hospital employees. Only WUSM physical therapy faculty practitioners participated in this demonstration project.

Number 3

The 52,000-sq ft BJH ED hosts the only level 1 trauma center in St Louis and, in 2005, provided care to 62,000 patients.28 The ED is divided into 4 separate areas associated with severity of medical presentation: trauma/critical care (12 beds), emergent care (31 beds), urgent care (12 beds), and observation (12 beds). Priority of care is determined through standardized triage processes that use indicators of urgency, manage patient waiting time, and identify nonemergent cases.29 In the BJH ED, nurses are specifically trained to categorize patients by signs of acuity and health risk and determine priority of service. Placement in 1 of 5 triage categories (A⫽resuscitation, B⫽emergent, C⫽urgent, D⫽semiurgent, E⫽nonurgent) reflects the number of resources (eg, laboratory work, imaging, specialty consults) each patient may require. Triage outcomes (eg, patients’ complaints and status) are available electronically, enabling ED providers to track the progress of patients through examination and intervention procedures. Results of diagnostic tests (eg, imaging, hematology) and some documentation also are available electronically.

Development of the Process One author (R.H.D.) created the concept, secured the funding, developed the administrative infrastructure, and implemented the plan for this physical therapy service. Developmental steps over several years preceded service delivery and built visibility: (1) observing in the ED and communicating with university and hospital leadership, (2) testing provider acceptance, (3) analyzing projected volumes and staffing needs, (4) planning for assessment, and (5) presenting a final proposal.28 Table 1 details activities related to these steps. When this project was first envisioned in 2004, back pain was the March 2010

Physical Therapy in the Emergency Department sixth most common complaint seen in this ED, accounting for 2,031 patients in that year. Combining this with other likely categories of musculoskeletal problems drawn from the top 75 chief complaints (totaling 10,737 patients),28 potential encounters amenable to physical therapy management were estimated. This estimate reflected indicators from the literature5 and information obtained from another Midwestern hospital with similar volumes that had implemented physical therapy services in the ED (personal communication, Pauline Flesch, February 2005). We projected that managing the resulting target volume (2,555 encounters or 4.3% of the top 75 chief complaints) could require at least a half-time staffing effort. A 3-day per week schedule was proposed with times that rotated each month so that all days could be tested for optimal service visibility. The proposal was presented to the BJH chief operating officer, who is responsible for ED management. Funding from hospital sources for salary, benefits, and supplies was requested. No requirements for developing billing services were built into the initial model. The WUSM Program in Physical Therapy leadership selected one of the authors (D.F.M.) to staff the ED because of her expertise in managing acute and chronic pain conditions; her experience in neuromuscular and musculoskeletal rehabilitation in inpatient, outpatient, and long-term care; and her ability to build collaborative networks in practice. These characteristics were viewed as critical, especially in the early phases of service implementation. The original proposal recommended evaluating effectiveness of patient care, patient waiting time, cost-effectiveness of care, and efficiency of the ED health care team (Tab. 2). Annual reports of these measures— or others if the original measures were not feasible

March 2010

Table 1. Service Development Processa Activity

Outcome

Observed triage and patient care

Selected times and days projected to be optimal for patient flow and staffing

Proposed service and its benefits to WUSM and BJH leadership

Discovered support from all levels of leadership

Surveyed physicians, nurses, residents, and physician assistants to test perceptions of physical therapy impact on ED culture

Received unanimous response that physical therapy could contribute to diagnosis, management, and follow-up of patients; some doubt about potential to decrease costs or waiting time

Analyzed data for proposal development

Projected volumes, costs to support initial fulltime equivalent position and supplies

Selected assessment directions

Targeted volumes, referral source, encounter time and type, and patient satisfaction and disposition after ED discharge

Considered documentation formats

Designed initial and discharge forms

Presented case studies for attending physicians and residents to illustrate inclusion of a physical therapist as an ED provider

Identified providers willing to refer patients to a physical therapist and interested in understanding physical therapist’s ability to assist in differential diagnosis

a WUSM ⫽Washington University School of Medicine, BJH⫽Barnes-Jewish Hospital, ED⫽emergency department.

or meaningful—were pledged to hospital administration.

Application of the Process The service began with a funding commitment from BJH for the proposed half full-time equivalent position, a commitment that increased to one full-time equivalent position in the second year. This cost accounts for less than 1% of the entire ED budget. To assist in resident training, a series of abbreviated case studies were developed that illustrated the scope of interventions used and how physical therapists can assist in diag-

nostic and disposition decisions. Flexibility and patience were required to achieve sufficient visibility given the persistence of traditional triage processes and the need to continually educate providers who returned to those traditions at times when the physical therapist was offsite. Early signs of acceptance into the care team resulted in the ED creating an electronic physical therapy consult icon and inviting us to create a physical therapy documentation form for the new electronic medical record. Awareness of the service prompted ED clinicians to page the

Table 2. Outcome Measures Proposed Outcome Target

Specific Measures

Method

Effectiveness of care

Initial and discharge pain

Analog pain scale Medication type/amount Frequency of returns

Patient waiting time

Time to triage and intervene

Patient satisfaction surveys Length of stay

Cost-effectiveness

Use of radiographs and medications

Radiographs for select diagnoses Medication timing during care pathway

Efficiency

Triage trajectories

Referral patterns Staff satisfaction surveys

Volume 90

Number 3

Physical Therapy f

423

Physical Therapy in the Emergency Department Table 3. Case Examples of Patients Seen by Physical Therapist Age (y)/Sex

a

Triage Categorya

Chief Complaint

Imaging Prior to Physical Therapy

Physical Therapist Examination

Intervention

32/female

D

Ankle and foot pain after tripping. Second emergency department visit for same complaint.

Prior and current radiograph of foot: negative.

Foot and ankle screening inconclusive. Correcting fibular head alignment decreased symptoms.

Education, gait training, outpatient physical therapy referral.

55/male

C

Chest and shoulder pain, hand tingling after scaffolding fell onto shoulder. Cardiac issues and fractures ruled out.

Computed tomography of shoulder: negative.

Cervical screening reproduced symptoms.

Education, neck and shoulder postural correction, outpatient physical therapy referral.

22/male

B

Single-car accident, intoxicated; lost consciousness. Head and chest trauma ruled out. Knee pain.

Computed tomography of knee: negative.

Knee screening not consistent with primary musculoskeletal pain problem.

Referred back to physician. Magnetic resonance imaging showed complete tears of all knee ligaments and popliteus muscle. Hospital admission.

Triage categories: B⫽emergent, C⫽urgent, and D⫽semiurgent.

physical therapist to provide telephone consultation when the physical therapist was not on-site. Relationships were built with BJH and WUSM information systems personnel to enable surveys of patient satisfaction and analysis of length-ofstay patterns in the ED.

Preliminary Outcomes The case examples shown in Table 3 demonstrate the types of patients seen in the ED and suggest how physical therapy can influence movement, function, pain management, and disposition in the emergent care setting. Between August 2005 and May 2007 (6 months of a half fulltime equivalent position and 12 months of a full-time equivalent position), 316 patients were seen, with referrals highly variable from month to month. Most referrals (72%) occurred between 8:00 a.m. and 4:00 p.m.; MDs and NPs provided 93% of those referrals. Eighty-nine percent of referrals were from the emergent or urgent care areas of the ED. Between June 2007 and May 2008, 518 patients were referred (average of 1.98 patients per day), and between 424

f

Physical Therapy

Volume 90

June 2008 and May 2009, 565 patients were referred (average of 2.56 patients per day). Available data regarding the chief pain complaints of patients seen by the physical therapist in 2008 are presented in Table 4. This increase in consult requests reflects the gradual acceptance of the physical therapist by ED providers, particularly attending physicians. Written feedback from ED personnel complimented the effective manageTable 4. 2008 Chief Pain Complaints of Patients Seen by Physical Therapists (n⫽422) Chief Complaint Back

Percentagea 43.6

Neck

11.6

Knee

11.6

Ankle and foot

8.8

Shoulder

8.3

Other

8.1

Hip

5.9

Hand and wrist

2.1

Elbow

0.5

a

Total percentage exceeds 100% due to patients having multiple complaints.

Number 3

ment of musculoskeletal pain, ability to provide follow-up resources, and contributions to differential diagnosis and disposition planning. Concerns were voiced only about the inability to provide services during all hours of ED operation. Patient perspectives, obtained using a short written survey and a telephone follow-up by BJH’s customer satisfaction research team, suggested that physical therapy intervention helps patients learn to reduce pain and avoid subsequent problems of the same type. A more systematic test of patient satisfaction is required to make definitive conclusions about patient regard for physical therapy in the ED. The complexity of the BJH ED environment, including financial reporting, staffing patterns of all providers, and documentation paradigms, limited our ability to comprehensively assess cost-effectiveness and service efficiency. However, length-of-stay data were tracked for patients who received physical therapy between 2005 and 2006. Compared with the average BJH ED length of stay of 6.3 March 2010

Physical Therapy in the Emergency Department hours, 50% of patients seen by the physical therapist showed a length of stay of less than 5 hours. Average encounter time with the physical therapist was 54 minutes (range⫽ 15–105 minutes). The accuracy of this measure is complicated by the periodic interruption of care that occurs when patients are sent for diagnostic tests or moved to other ED areas during the course of physical therapy intervention. Further measures (eg, total cost of care, number of returns within 72 hours for the same complaint) are being pursued, but such data are not easily accessed due to a limited number of BJH information system staff who can analyze hospital data. Ensuring appropriate follow-up of patients seen in the ED is challenging but important. In comparison with national data reported in 2005,1 the case mix within the BJH ED has more patients funded by Medicare (27.2%) and Medicaid (27.2%), fewer supported by commercial health insurance (24.4%), and more who lack insurance completely (23%). To provide uninsured patients with follow-up physical therapy care, a Saturday pro bono clinic was established. This clinic is staffed by WUSM professional doctor of physical therapy students supervised by faculty practitioners. In the first year of the pro bono clinic, 168 patients were referred and 82 patients were seen. In the next year, 236 patients were referred and 104 patients were seen.

Discussion Thus far, physical therapy has been used in all areas of the BJH ED, which now hosts 88,000 visits annually. Physical therapists have evaluated and treated patients with many different medical diagnoses and have assisted with pain management, safety assessments, differential diagnosis of complex medical conditions, and discharge planning. Although the frequency of physical therapy consults continues to increase, the most chalMarch 2010

Table 5. 2008 Resources for Follow-up (n⫽231) Follow-up

Percentage

Outpatient physical therapy

84.4

Referred back to physician

7.8

Hospital admission

4.8

Home physical therapy

1.7

Other

1.3

lenging role has been to educate other ED providers about the knowledge and skills a physical therapist contributes in managing musculoskeletal problems. Our current average of approximately 3 patients per day (compared with the expectation in faculty practice of 12 patients per day) reflects how physical therapy intervention is complicated by acuity and severity of pain, specialty consultations, medical testing and medication regimens, and transfers to other areas within the ED. To reduce the number of return visits to the ED in less than 72 hours, especially because of persistent pain, patients are provided with extensive education and appropriate follow-up resources (Tab. 5). Unfortunately follow-up in our pro bono clinic is compromised by the numerous patients who do not keep appointments, possibly because of transportation issues and family obligations.30 This project has expanded the visibility of physical therapy among providers in the BJH ED and begun to demonstrate how using physical therapy services may help improve overall patient care in the emergent care setting. Physical therapist practice in the ED requires adapting to many complexities while providing rapid and effective patient care services. Exceptional skills in identifying movement and postural faults and the ability to systematically assess their origin and meaning are essential. Equally important is the ability to identify patient problems that are not amenable to physical ther-

apy. Consistent with the literature, the preliminary data suggest high levels of patient satisfaction with the service3,12,25 and decreased waiting time.25 Although costs have not been shown to be reduced, most ED personnel now understand how physical therapists can identify conditions appropriate for physical therapy referral. This ability could enable physical therapists to enter the triage process at an earlier stage to decrease some patients’ need for radiographs and pain medication and to decrease further overall patient waiting time by expediting physical therapy consultation requests for patients with musculoskeletal complaints. These changes could allow other ED providers to focus on more urgent patient cases.29 These initiatives are expected to require continual reinforcement and preliminary testing because their implementation requires modifying traditional ED processes and influencing existing ED culture. Additional recommendations in expanding and refining physical therapy service in the ED include: • Establishing standing orders that enable triage of patients with musculoskeletal pain directly to physical therapists while ensuring appropriate precautions to avoid clinical error. • Building a financial model to ensure the stability of the service. • Developing a physical therapy staffing model that optimizes ED coverage while permitting practitioners to pursue other professional obligations. • Improving service assessment by comparing outcomes of care (eg, cost of care, pain, length of stay, function) for patients reporting specific musculoskeletal complaints who do and do not receive physical therapy.

This practice venue has enabled physical therapists to use their knowledge, diagnostic skills, and ability to manage pain and musculo-

Volume 90

Number 3

Physical Therapy f

425

Physical Therapy in the Emergency Department skeletal injury as they are seen in the hospital ED. The partnership with a teaching hospital may have had some initial benefits in welcoming new consultative services. However, building the service required multiple levels of approval and visibility that may not be needed in less complex hospitals. Independent of the environment, this practice has required creativity, flexibility, persistence, and an appreciation for other practitioners’ resistance to changing the traditional patterns of triage and care within EDs. The future intention is to incorporate clinical education opportunities during physical therapists’ professional training, residencies, and fellowships to reinforce this practice direction as a viable career option for physical therapists in this country. The opportunity to bridge organizational lines has enhanced the professional development of those involved and cemented relationships across disciplines. Dr Deusinger and Dr Deusinger provided concept/idea/project design, data analysis, project management, fund procurement, institutional liaisons, and consultation (including review of manuscript before submission). All authors provided writing. Ms Czuppon provided data collection and patients. The authors thank Lawrence Lewis, MD, and Brent Ruoff, MD, Washington University School of Medicine, for strong advocacy to their colleagues for inclusion of physical therapy in the emergency department; Sharon O’Keefe, RN, MBA, Chief Operating Officer, Barnes-Jewish Hospital, for approving the initial financial support for this service; Damon Braggs, MBA, Manager of Business Medicine, for his assistance with the financial management of this program; and all participating personnel in the Barnes-Jewish Hospital Emergency Department for welcoming this new service into this very complex clinical environment. Funding for this demonstration project was provided by Barnes-Jewish Hospital and the Program in Physical Therapy at Washington University School of Medicine. This article was received September 3, 2008, and was accepted October 12, 2009.

References 1 Nawar EW, Niska RW, Xu J. National Hospital Ambulatory Medical Care Survey: 2005 Emergency Department Summary (NHAMCS). Advance Data from Vital and Health Statistics—Centers for Disease Control and Prevention, Department of Health and Human Services. No. 386, June 2007. 2 Wilsey BL, Fishman SM, Ogden C, et al. Chronic pain management in the emergency department: a survey of attitudes and beliefs. Pain Med. 2008;9:1073–1080. 3 Lau PM, Chow DH, Pope MH. Early physiotherapy intervention in an accident and emergency department reduces pain and improves satisfaction for patients with acute low back pain: a randomized trial. Aust J Physiother. 2008;54:243–249. 4 Lebec MT, Jogodka CE. The physical therapist as a musculoskeletal specialist in the emergency department. J Orthop Sports Phys Ther. 2009;39:221–229. 5 Graeme AC, Jones MB. Musculoskeletal problems at an accident and emergency department and in general practice. NZ Med J. 1983;98:529 –531. 6 Darwent M, Gamon A, McLoughlin F. Early physiotherapy within the accident and emergency department. Physiotherapy. 1998;83:281. 7 Garbez R, Puntillo K. Acute musculoskeletal pain the emergency department: a review of the literature and implications for the advanced practice nurse. AACN Clin Issues. 2005;16:310 –319. 8 Griffin M, Melby V. Developing and advanced nurse practitioner service in emergency care: attitudes of nurses and doctors. J Adv Nurs.2006;56:292–301. 9 DiCaprio MR, Covey A, Bernstein J. Curricular requirements for musculoskeletal medicine in American medical schools. J Bone Joint Surg Am. 2003;85:565–567. 10 Childs JD, Whitman JM, Sizer PS, et al. A description of physical therapists’ knowledge in managing musculoskeletal conditions. BMC Musculoskelet Disord. 2005; 6:32. 11 Daker-White G, Carr AJ, Harvey I, et al. A randomised controlled trial: shifting boundaries of doctors and physiotherapists in orthopaedic outpatient departments. J Epidemiol Community Health. 1999;53:643– 650. 12 Richardson B, Shepstone L, Poland F, et al. Randomised controlled trial and cost consequences study comparing initial physiotherapy assessment and management with routine practice for selected patients in an accident and emergency department of an acute hospital. Emerg Med J. 2005;22: 87–92. 13 Jorgensen DJ. Fiscal analysis of emergency admissions for chronic back pain: a pilot study from a Maine hospital. Pain Med. 2007;8:354 –358. 14 Isaacs DM, Marinac J, Sun C. Radiograph use in low back pain: a United States emergency department database analysis. J Emerg Med. 2004;26:37– 45.

15 Anaf S, Sheppard LA. Describing physiotherapy interventions in an emergency department setting: an observational pilot study. Accid Emerg Nurs. 2007;15:34 –39. 16 McKinney LA, Dornan JO, Ryan M. The role of physiotherapy in the management of acute neck sprains following road-traffic accidents. Arch Emerg Med. 1989;6:27–33. 17 Hourigan PG, Weatherly 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 Wilsey BL, Fishman SM, Crandall M, et al. A qualitative study of the barriers to chronic pain management in the ED. Amer J Emerg Med. 2008;26:255–263. 19 Guide to Physical Therapist Practice. 2nd ed. Phys Ther. 2001;81:9 –746. 20 Stowell T, Cioffredi W, Greiner A, Cleland J. Abdominal differential diagnosis in a patient referred to a physical therapy clinic for low back pain. J Orthop Sports Phys Ther. 2005;35:755–764. 21 Troyer MR. Differential diagnosis of endometriosis in a young adult woman with nonspecific low back pain. Phys Ther. 2007;87:801– 810. 22 Browder DA, Erhard RE. Decision making for a painful hip: a case requiring referral. J Orthop Sports Phys Ther. 2005; 35:738 –744. 23 Sasaki M. Cervical cord compression secondary to ossification of the posterior longitudinal ligament. J Orthop Sports Phys Ther. 2005;35:722–729. 24 Moore JH, McMillian DJ, Rosenthal MD, Weishaar MD. Risk determination for patients with direct access to physical therapy in military health care facilities. J Orthop Sports Phys Ther. 2005;35:674 – 678. 25 McClellan CM, Greenwood R, Benger JR. Effect of an extended scope physiotherapy service on patient satisfaction and the outcome of soft tissue injuries in an adult emergency department. Emerg Med J. 2006;23:384 –387. 26 Hackett GI, Bundred P, Hutton JL, et al. Management of joint and soft tissue issues in three general practices: value of on-site physiotherapy. Br J Gen Pract. 1993;43: 61– 64. 27 Kuritzky L. Current management of acute musculoskeletal pain in the ambulatory care setting. Am J Ther. 2008;15(supp 10): S7–S11. 28 Deusinger RH. Demonstration project proposal: Physical therapy care within BJH emergency services: a Washington University physical therapy clinics and Barnes Jewish Hospital patient care innovation. March 15, 2005. 29 Derlet RW, Kinser D, Ray L, et al. Prospective identification and triage of nonemergency patients out of an emergency department: a 5-year study. Ann Emerg Med. 1995;25:215–223. 30 Handel DA, McConnell KJ, Allen H, et al. Outpatient follow-up in today’s health care environment. Ann Emerg Med. 2007; 49:288 –292.

DOI: 10.2522/ptj.20080268

426

f

Physical Therapy

Volume 90

Number 3

March 2010

Case Report

Aerobic Exercise During Early Rehabilitation for Cervical Spinal Cord Injury Amira E. Tawashy, Janice J. Eng, Andrei V. Krassioukov, William C. Miller, Shannon Sproule

Background and Purpose. People with spinal cord injuries (SCIs), particularly those with injuries causing tetraplegia, are at risk for cardiovascular illnesses. There is a compelling need to address poor cardiovascular health as early as possible after cervical SCI. The purpose of this case report is to illustrate the process of aerobic exercise prescription during inpatient rehabilitation for cervical SCI.

Case Description. The patient was a 22-year-old man who had sustained a complete C5 SCI during a swimming accident 12 weeks before he participated in an aerobic exercise program. The program was developed to facilitate aerobic capacity while minimizing muscular fatigue. The patient attended 18 sessions over a 2-month period.

Outcomes. The patient’s exercise tolerance increased in terms of both exercise duration and exercise intensity. Measurements of cardiovascular health, taken before and after training, revealed substantial increases in peak oxygen uptake (20%) and orthostatic tolerance over the course of the program.

Discussion. The patient experienced typical complications associated with acute SCI (eg, orthostatic hypotension, urinary tract infections). He exhibited several signs of improved exercise tolerance and wheelchair mobility during the 2-month program, indicating potential cardiovascular and functional improvements from the exercise training.

A.E. Tawashy, MSc, Graduate Program in Rehabilitation Sciences, University of British Columbia, and G.F. Strong Rehabilitation Centre, Vancouver, British Columbia, Canada. She was a graduate student at the time of this project. J.J. Eng, PT/OT, PhD, is Professor, Department of Physical Therapy, University of British Columbia, 212-2177 Wesbrook Mall, Vancouver, British Columbia, Canada V6T 1Z3. She also is affiliated with the International Collaboration on Repair Discoveries (ICORD) and the G.F. Strong Rehabilitation Centre, Vancouver, British Columbia, Canada. Address all correspondence to Dr Eng at: [email protected]. A.V. Krassioukov, MD, PhD, is Associate Professor, Division of Physical Medicine and Rehabilitation, Department of Medicine, and the International Collaboration on Repair Discoveries (ICORD), University of British Columbia. W.C. Miller, OT, PhD, is Associate Professor, Department of Occupational Science and Occupational Therapy, and the International Collaboration on Repair Discoveries (ICORD), University of British Columbia. S. Sproule, PT, BSc(PT), is Physical Therapist, G.F. Strong Rehabilitation Centre, Vancouver, British Columbia, Canada. [Tawashy AE, Eng JJ, Krassioukov AV, et al. Aerobic exercise during early rehabilitation for cervical spinal cord injury. Phys Ther. 2010; 90:427– 437.] © 2010 American Physical Therapy Association

Post a Rapid Response or find The Bottom Line: www.ptjournal.org March 2010

Volume 90

Number 3

Physical Therapy f

427

Aerobic Exercise in Spinal Cord Injury Rehabilitation

C

ervical spinal cord injuries (SCI) significantly disrupt connections in the spinal cord, causing severe muscle paralysis, loss of sensation, and autonomic dysfunction. People with cervical SCI are likely to become inactive1 and exhibit low levels of cardiovascular fitness.2 The latter significantly increases the risk of cardiovascular disease. Cardiovascular disease is the leading cause of death in both people who are able-bodied and people with SCI.3 However, the disease occurs earlier and is more prevalent in the latter population, and tetraplegic injuries are associated with a 16% higher risk of developing the disease than paraplegic injuries.4 Although reduced muscle mass is a fundamental factor contributing to inactivity, autonomic dysfunction compounds this problem. The failure of sympathetically driven cardiac acceleration and contractility mechanisms, combined with venous pooling, results in severely decreased cardiac output and therefore a reduction in the overall ability to transport oxygen to the exercising muscle. These factors lead to a vicious cycle of further decline because they can result in reduced functional capacity and, therefore, threaten the ability to live independently.

Physical inactivity directly predisposes people to metabolic syndrome.5 Metabolic syndrome, a prediabetic state characterized by a group of metabolic risk factors (obesity, high blood pressure, insulin resistance, and atherogenic dyslipide-

Available With This Article at ptjournal.apta.org • Audio Abstracts Podcast This article was published ahead of print on January 21, 2010, at ptjournal.apta.org.

428

f

Physical Therapy

Volume 90

mia6), is considered to be a major risk factor for heart disease7; its presence approximately doubles the risk of cardiovascular disease mortality.4 Metabolic syndrome has been shown to be present in 23% of people with SCI (approximately double its presence in people who are able-bodied and of similar ages6); the highest incidence of the disease is apparent in people with tetraplegia.4 The increased risk for metabolic syndrome in the latter population is likely multifactorial. A loss of muscle mass attributable to paralysis causes a shift in body composition from lean mass (composed of skeletal muscle and bone) to fat mass, thus promoting insulin resistance and physical inactivity.4 Another cardiovascular dysfunction observed in people with SCI is orthostatic hypotension (OH), which is prevelant in both the acute8 and the chronic9 stages of SCI. Orthostatic hypotension can significantly complicate and delay rehabilitation for cervical SCI because 58.9% of patients experience symptomatic OH with mobilization.10 It has even been suggested that OH can lead to neurological deterioration in people who may otherwise have stable SCI.11 Because prolonged bed rest and cardiac insufficiency are 2 primary causes of OH in people who are able-bodied,12,13 it is assumed that severe cardiovascular deconditioning in the initial months after SCI increases the risk of OH and, therefore, threatens efficient rehabilitation. It is critical to address cardiovascular function as early as possible after newly sustained SCI. People with acute, motor-complete cervical SCI can spend up to 4 weeks in bed during acute care.10 The negative cardiovascular consequences of such prolonged bed rest have been well documented for people who are able-bodied14 and are particularly detrimental for people with cervical SCI. Thus, cardiovascular training, in

Number 3

addition to increasing sitting tolerance and functional ability, during subacute rehabilitation may halt the cardiovascular deterioration that results from prolonged immobility and reduce OH symptoms. Cardiovascular training is needed if people with SCI are to reach and maintain the level of cardiovascular fitness that is desirable for functioning in daily life. Thus, the effort to achieve optimal levels of fitness through sustained aerobic activity should start during primary rehabilitation. Despite the compelling rationale to optimize physical fitness levels in people with SCI, research investigating conventional aerobic exercise (eg, sustained wheelchair propulsion [wheeling], arm ergometry) during subacute rehabilitation is scarce. There are currently no guidelines for prescribing aerobic exercise for people with cervical SCI, nor is it an integral part of subacute rehabilitation. To our knowledge, only 2 studies have explored the use of aerobic exercise as an adjunct to primary rehabilitation.15,16 These studies aimed to maintain an exercise frequency of 3 to 5 days per week at an intensity of 70% to 80% of heart rate reserve (HRR). However, the 2003 study by De Groot et al15 (n⫽6) included only 1 individual with cervical SCI, sustained more than 6 months before participation in the program, and the 2005 study by Bizzarini et al16 did not include any people with complete cervical SCI and did not involve any cardiovascular assessments. Although previous studies17–19 demonstrated the beneficial effects of upper-extremity training on cardiovascular status in people with cervical injuries, no study has assessed the feasibility (or the effects) of an aerobic exercise program designed specifically for people with tetraplegia during inpatient rehabilitation. Because people with cervical SCI are severely limited by autonomic nerMarch 2010

Aerobic Exercise in Spinal Cord Injury Rehabilitation vous system impairments and paralysis, they are not always able to participate in programs designed for people with full upper-body strength (force-generating capacity) and intact autonomic control of the cardiovascular system. The purpose of this case report is to outline the process used to develop an exercise program to enhance cardiovascular fitness in a patient undergoing primary rehabilitation for a motor-complete C5-C6 SCI. We used a novel upperextremity circuit training program that aimed to minimize muscular fatigue to facilitate gains in aerobic capacity. Secondary outcomes included measures of OH, lipids, functional wheeling, and program satisfaction.

tolerate a full day in his manual wheelchair and expressed no complaints of pain, shortness of breath, or cough. He was managing occasional OH with abdominal binders and thromboembolism deterrent stockings. He was participating in standard-of-care physical therapy (1 hour per day) and occupational therapy (1 hour per day), which addressed range of motion, functional mobility, activities of daily living, and wheelchair skills training.

Clinical Impression The patient appeared to be a potential candidate for our novel treatment because he was highly motivated to work on his cardiovascular fitness. An American Spinal Injury Association Impairment Scale (AIS) examination20 was planned to clarify the completeness and severity of the injury.

Clinical Impression The patient was diagnosed with motor-complete tetraplegia. He was a good candidate for our novel treatment because his motor function and strength were such that he was able to perform all exercises in the proposed aerobic exercise program (outlined below). Measurements of cardiovascular health (peak oxygen ˙ O2], 6-minute arm ergomeuptake [V ter test, lipid profile, sit-up test, and timed functional wheeling) were to be taken before the commencement of the exercise program, at the midpoint of the program, and at program completion. Because of the medical management of the patient’s injury (cervical collar) and the development of deep vein thrombosis, baseline assessments for the case report were completed 6 weeks after admission to the rehabilitation center, and the exercise regimen commenced 3 months after the patient sustained his injury. The patient provided informed consent for the treatment program.

Examination

Intervention

The patient rated his activity level before the injury as “very active.” The AIS examination revealed a C5 sensory-C6 motor AIS-A SCI. The patient had full passive shoulder range of motion, with an overall motor score of 20/100, and was able to perform some activities of daily living (eg, brushing teeth, upper-body dressing) with assistance. He could

The American College of Sports Medicine (ACSM) recommends a minimum exercise frequency of 3 times per week and a minimum duration of 20 to 30 minutes.21 The ACSM suggests that an exercise intensity of 70% to 80% of HRR is necessary to produce training effects in deconditioned people.21 Designing an effective aerobic exercise program for

Patient History and Review of Systems The patient was a 22-year-old man (weight⫽62 kg, height⫽188 cm) admitted for inpatient rehabilitation after a traumatic SCI sustained during a swimming accident. He spent 50 days in acute care before being transferred to the inpatient rehabilitation center.

March 2010

people with tetraplegia is challenging because of physiological and musculoskeletal considerations. Cervical SCIs disrupt sympathetic outflow to the heart, resulting in a blunted chronotropic response to exercise. Normally, the adrenal medulla, which is innervated by cholinergic preganglionic sympathetic neurons originating between T5 and T9,22 releases epinephrine and norepinephrine into the blood in response to stress or exercise, causing an increase in both heart rate (HR) and blood pressure as well as blood vessel constriction. Whereas people who are able-bodied will show exponential increases in free plasma epinephrine and norepinephrine levels during maximal effort, people with tetraplegia will exhibit only small increases in the levels of these same catecholamines, indicating no considerable stimulation of the sympathetic nervous system.22 This loss of sympathetic outflow to the heart and adrenal glands results in bradycardia.23 Consequently, people with tetraplegia show a blunted chronotropic response to exercise and are unable to attain their age-predicted maximal HR; the peak HR in people with cervical injuries rarely exceeds 120 bpm.24 Thus, exercise intensity in this population cannot be monitored solely with HR and must include ratings of perceived exertion. Because arm exercises involve a relatively small muscle mass, the primary factors limiting performance may be peripheral in nature; thus, local fatigue of the highly stressed arm musculature may occur despite sufficient systemic oxygen availability.15 Accordingly, we used interval training to facilitate longer cumulative exercise durations and incorporated several activities to minimize muscle fatigue and boredom. We developed a circuit that included 4 activities (arm ergometry, boxing,

Volume 90

Number 3

Physical Therapy f

429

Aerobic Exercise in Spinal Cord Injury Rehabilitation Table 1. Guidelines for Progression of Intermittent Exercisea

Level

Value on Borg Rating of Perceived Exertion (RPE)

Total Minutes at Indicated Value on Borg RPE

1

12–14

18

3

2

12–14

21

3.5

1.5

3

12–14

24

4

1

4

14–16

24

4

1

5

14–16

27

4.5

0.5

6

14–18 ([1 min at RPE value of 14 ⫹ 1 min at RPE value of 18] ⫻ 2)

24

4

1

Minutes of Exercise

Minutes of Rest 2

a Six repetitions were performed at each level. The patient was considered to be able to progress to the next level of exercise when both of the following criteria were met: first, the patient was able to complete the time goal (exercise minutes) at the current level without pain or overbearing fatigue (fatigue that interferes with activities of daily living or regular therapy, as reported by the patient), and second, the patient was able to complete the exertion goal at the current level without pain or overbearing fatigue.

sliding motion, and wheeling), 2 of which were repeated, for a total of 6 timed stations. The patient chose which 2 stations to repeat; these often were the boxing and sliding stations. Emphasis was placed on proper technique during all stations to minimize the risk of shoulder pain. The patient could perform all of the activities independently. Five minutes was allotted to each station, for a total session duration of 30 minutes. The patient began and finished each session with 2 minutes on the arm ergometer against no resistance. The circuit was to be performed 3 times per week. The exercise protocol was made more challenging throughout the treatment.

encouraged to punch the bag with the thumb pointing upward to mitigate any potential shoulder impingement.25 A VitaGlide wheelchair fitness machine† with Tri-Post Adaptors was used for sliding. The Tri-Post Adaptors provided support for the patient’s limited gripping ability. Wheeling was performed forward and backward along a 20-m hallway.

Arm ergometry was performed on an arm cycle ergometer (Monark Rehab Trainer 881E*) against 2 to 5 W of resistance. Upright hand cuffs were used so that the hand was placed with the thumb pointing upward, and the ergometer was positioned so that the arm never exceeded the height of the shoulder. A small boxing bag was positioned at shoulder height on the wall. The patient was

Exercise intensity was monitored with an HR monitor and as perceived exertion. Both exercise duration and intensity progressed over the course of the program in accordance with the ACSM guidelines for exercise progression for individuals with chronic conditions21: (1) duration: stations initially consisted of 3 minutes of exercise followed by 2 minutes of rest (18 cumulative minutes of exercise) and gradually increased to 4.5 minutes of exercise followed by 30 seconds of rest (27 cumulative minutes of exercise); (2) intensity: target work rates (HR as determined ˙ O2 tests and correfrom initial peak V sponding values on the Borg Rating of Perceived Exertion [RPE]26) were

* Monark, Kroons va¨g 1, Vansbro, Sweden 780 50.

† RehaMed International, LCC, 522 W Mowry Dr, Homestead, FL 33030

430

f

Physical Therapy

Volume 90

Number 3

initially set to 50% of HRR and then gradually increased to 70% to 80% of HRR. The HRR was calculated from a resting HR of 73 bpm and a maximal HR of 109 bpm. Progression guidelines are shown in Table 1.

Outcome All assessments were performed at 3 time points: baseline (before the exercise program), midpoint (40 days after the first training session, the point at which the patient could complete level 3 of the exercise program [Tab. 1]), and endpoint (80 days after the first training session). The patient did not wear thromboembolism deterrent stockings for any of the tests. Primary Outcome Measure Cardiopulmonary status: peak V˙O2. A maximal graded exercise test was performed before the initiation of training to assess cardiopulmonary status. The protocol for as˙ O2 in people with SCI sessing peak V is well established27 and documented.28 In brief, the test consisted of 1 minute of arm cycling against no resistance at a comfortable cadence (generally 50 –70 rpm) and subsequent work load increases of 5 W/min29 until volitional exhaustion. The test was performed on an electronically braked arm ergometer (Excaliber‡). Cardiovascular and metabolic measurements were collected with a 12-lead electrocardiograph (Quark C12§) and a non-rebreathing face mask (K4b2§). The patient rated his perceived exertion by using the Borg RPE immediately after the test. Secondary Outcome Measures Functional capacity: 6-minute submaximal arm ergometer task. As a measure of functional capacity, the patient completed a single-stage, ‡ Lode BV Medical Technology, Zernikepark 16, Groningen, the Netherlands 9747AN. § COSMED Srl, Via dei Piani di Mt Savello 37, Pavona di Albano, Rome I-00041, Italy.

March 2010

Aerobic Exercise in Spinal Cord Injury Rehabilitation 6-minute submaximal arm ergometer task shown to be valid and reliable in people with SCI.28 He cycled against 5 W of resistance at a comfortable cadence (50 –70 rpm) for 6 minutes on the arm cycle ergometer used in the training circuit. During arm cycling, the patient wore an HR monitor (Polar A3㛳) and the nonrebreathing face mask used in the ˙ O2 test. The HR was recorded peak V every 5 seconds. An average HR over a 30-second period during the last minute of this test was used to determine functional capacity. Immediately after the test, the patient rated his perceived exertion by using the Borg RPE. Lipid profile. Levels of total cholesterol, high-density lipoprotein (HDL), low-density lipoprotein (LDL), and triglycerides in serum were analyzed by use of fasting blood samples. Orthostatic tolerance: sit-up test. The presence of OH was assessed with a sit-up test.9 In the supine position, the patient was fitted with a 12-lead electrocardiograph. Systolic blood pressure, diastolic blood pressure, and mean blood pressure were measured with a calibrated DINAMAP 300 monitor.# This monitor accurately measures blood pressure (mean error⫽ⱕ5 mm Hg, SD⫽8) and HR (⫾3.5%).30 A cuff was placed on the patient’s right bicep, and an oximeter was placed on his left index finger. Baseline HR and blood pressure recordings were made during a 10-minute supine rest period. The patient then was passively moved into an upright seated position by raising the head of the plinth by 90 degrees and dropping the base of the plinth by 90 degrees from the knees. This “sit-up” posi㛳

Polar Electro Inc, 1111 Marcus Ave, Suite M15, Lake Success, NY 11042-1034 # Critikon Inc, 4110 George Rd, Tampa, FL 33634.

March 2010

tion is essentially the same as that of a patient seated in a wheelchair. The patient was informed about the importance of the sit-up maneuver being passive and was instructed not to assist during the procedure. The upright seated position was maintained for 10 minutes, during which HR and blood pressure recordings were continued. Values from minute 3 after the patient was moved into an upright posture are reported. Orthostatic hypotension was defined as a decrease in systolic blood pressure of ⱖ20 mm Hg or in diastolic blood pressure of ⱖ10 mm Hg in the upright position, whether or not symptoms occurred.31 Visual signs of OH (eg, yawning, pallor) and patientreported symptoms (eg, lightheadedness, dizziness) also were recorded. Timed functional wheeling. The patient completed 2 items from the Functional Tasks for Persons Who Self-Propel a Manual Wheelchair.32 Because the reliability of this measure is reported for each task individually (ie, single-item results can be used32), we used the timed forward wheeling (distance⫽23 m) and ramp ascent (10-m ramp, 1:12 grade) tasks in the treatment program. Patient perception of the program. At 1 week after exercise completion, the patient completed a 7-item questionnaire regarding his satisfaction with the program (Appendix). We designed the questionnaire on the basis of a literature review of satisfaction criteria and input from 2 SCI rehabilitation researchers. The items in each category were rated on an ordinal scale from 1 to 5. A rating of 5 indicated that the patient was extremely satisfied with the program, and a rating of 1 indicated that he was not satisfied. We calculated a total score from this questionnaire; a higher score indicated a greater degree of satisfaction.

Training intervention. The patient had expressed interest in participating in the exercise program at admission to the rehabilitation center. Although he eagerly attended 18 of 35 scheduled exercise sessions over a 2-month period (average attendance⫽2 sessions per week), his participation in the program (and required assessment sessions) was disrupted by several common medical setbacks (Fig. 1). Seventeen sessions were rescheduled because of statutory holidays or recreational day trips outside the center (n⫽5), specialist appointments outside the center (n⫽4), illness (n⫽5), or fatigue associated with illness (n⫽3). During the course of the exercise program, the patient contracted 3 urinary tract infections (UTIs; confirmed by urine cultures), developed heterotopic ossification over his right hip (confirmed on radiographs and bone scans), and sustained a shear wound on his left ischial tuberosity. Although every effort was made to reschedule sessions when possible (eg, to compensate for recreational programs, holidays, or specialist appointments), rescheduling was not always feasible because the patient was very active in his rehabilitation program (he participated in most recreational opportunities) and enjoyed spending time with family and friends. The progression of the exercise program with regard to percent of HRR and value on the Borg RPE is shown in Figure 1. During the program, exercise duration increased by 6 minutes (from 18 to 24 minutes), and exercise intensity increased in terms of both the average value on the Borg RPE (values of 10 [“light effort”]–18 [“very hard effort”]) and the average HRR (56%–106%). The average value on the Borg RPE in the first half of the program was 12 (“moderately hard”), and the average HRR was 65%. In the second half of the program, the average value on

Volume 90

Number 3

Physical Therapy f

431

Aerobic Exercise in Spinal Cord Injury Rehabilitation

Figure 1. Exercise program progression. The time course of the exercise program shows exercise intensity (HR reserve [HRR] and Borg Rating of Perceived Exertion [RPE]) over time. Two endpoint assessments were conducted because the patient felt too ill to continue with testing on day 182 (subsequent testing confirmed a urinary tract infection). Thus, the assessments were readministered once the patient felt well enough to perform the tests, and the results of this second endpoint evaluation (final evaluation 2) were analyzed. HO⫽heterotopic ossification (confirmed by radiography), SCI⫽spinal cord injury, UTI⫽urinary tract infection (confirmed by urine culture).

Table 2.

the Borg RPE was 15 (“hard”), and the average HRR was 85%.

Results of Cardiovascular Assessments Value at: Assessment

Baseline

Midpoint

Endpoint

V˙O2 (mL/kg/min)

11.78

14.16

13.72

Heart rate (bpm)

109

115

117

Power output (W)

20

20

30

Value on Borg Rating of Perceived Exertion

19

19

19

V˙O2 (mL/kg/min)

8.47

8.68

7.59

Heart rate (bpm)

110

114

99

11

16

11

Stress test (peak values)

Six-minute submaximal arm ergometer task (average values)

Value on Borg Rating of Perceived Exertion Lipid profile, mmol/L (mg/dL) Triglycerides

0.41 (36.49)

0.50 (44.50)

0.41 (36.49)

High-density lipoprotein cholesterol

1.08 (42.12)

1.11 (43.29)

1.30 (50.70)

Low-density lipoprotein cholesterol

0.87 (33.93)

0.75 (29.25)

1.05 (40.95)

1.98

1.88

1.95

Ratio of triglycerides to low-density lipoprotein

432

f

Physical Therapy

Volume 90

Number 3

Cardiovascular Assessments The results of the cardiovascular assessments are shown in Table 2. Cardiopulmonary status: peak V˙O2. Peak HR increased from 109 bpm (baseline evaluation) to 117 bpm (endpoint evaluation). Peak ˙ O2 increased from 11.78 mL/kg/ V min (baseline) to 13.72 mL/kg/min (endpoint). Power output, measured during the peak cardiovascular stress test, increased from 20 W (baseline) to 30 W (endpoint) during the 2-month period. Functional capacity: 6-minute submaximal arm ergometer task. ˙ O2 and HR values were lower at The V ˙ O2⫽7.59 the endpoint evaluation (V mL/kg/min, HR⫽99 bpm) than at the ˙ O2⫽8.47 mL/ baseline evaluation (V kg/min, HR⫽110 bpm); however,

March 2010

Aerobic Exercise in Spinal Cord Injury Rehabilitation Table 3. Patient’s Response to Orthostatic Challengea Supine Values (Average of First 10 Minutes) Evaluation

SBP (mm Hg)

DBP (mm Hg)

HR (bpm)

Final Values (Average of Last 3 Minutes)

Values at 3-Minute Mark SBP (mm Hg)

DBP (mm Hg)

HR (bpm)

SBP (mm Hg)

DBP (mm Hg)

HR (bpm)

Before program

107.3

58.2

72.8

75

37

87

86.7

49.3

90.7

Midpoint

124

75.6

59.6

112

72

67

98

66.7

87.7

After program

111.1

57.8

76.9

62

32

101

67.7

32

98.3

a

Average values from 10-minute supine rest period at beginning of challenge, values at 3 minutes after sit-up maneuver, and average values from final 18 to 20 minutes of challenge are shown for systolic blood pressure (SBP), diastolic blood pressure (DBP), and heart rate (HR). bpm⫽beats per minute.

the values on the Borg RPE remained the same (ie, 11). Lipid profile. Lipid changes were minimal. The patient’s HDL values increased from the baseline evaluation (1.08 mmol/L, 42.12 mg/dL) to the endpoint evaluation (1.30 mmol/L, 50.70 mg/dL), whereas the ratio of total cholesterol to HDL decreased by 0.03 (baseline⫽1.98, endpoint⫽1.95) during the course of the exercise program. Orthostatic tolerance: sit-up test. The patient’s response to an orthostatic challenge improved from the baseline evaluation to the midpoint evaluation (Tab. 3, Fig. 2). During the baseline evaluation, at 3 minutes after passive movement of the patient from a supine position to a sitting position, his systolic blood pressure decreased from 107.3 mm Hg to 75 mm Hg and his diastolic blood pressure decreased from 58.2 mm Hg to 37 mm Hg, confirming the presence of OH. These decreases in blood pressure were accompanied by feelings of dizziness and lightheadedness. During the midpoint evaluation, blood pressure did not show a large decrease at the 3-minute mark (like that observed during the baseline evaluation), and the patient had no complaints of adverse symptoms. Because of an acute UTI, however, these improvements were not apparent during the endpoint evaluation. Two endpoint asMarch 2010

sessments were required because the patient felt too ill to continue with testing on day 182 (subsequent testing confirmed a UTI). Thus, the assessments were readministered once the patient felt well enough to perform the tests. The results of this second endpoint evaluation were analyzed and are presented in Table 3 and Figure 2. Dehydration and fatigue (common consequences of a UTI) likely impaired orthostatic tolerance.12 However, the patient’s HR at 3 minutes after the sit-up maneuver during the endpoint evaluation (101 bpm) was 14 beats higher than

that at the baseline evaluation (87 bpm), suggesting an improvement in chronotropic compensation during the orthostatic challenge. Timed functional wheeling. The patient showed improvements on the 2 wheeling tests during the course of the exercise program, with substantially faster times to cover the flat distance (from 19 seconds at the baseline to 12 seconds at the endpoint) and the ramp (from 23 seconds [with 4 rest breaks] at the baseline to 12 seconds [without stopping] at the endpoint).

Figure 2. Sit-up test. Minute-by-minute measurements of mean arterial blood pressure in response to an orthostatic challenge test before the program, at the midpoint, and after the program are shown. Two endpoint assessments were conducted because the patient felt too ill to continue with testing on day 182 (subsequent testing confirmed a urinary tract infection). Thus, the assessments were readministered once the patient felt well enough to perform the tests. The results of this second endpoint evaluation were analyzed and are shown in Table 3.

Volume 90

Number 3

Physical Therapy f

433

Aerobic Exercise in Spinal Cord Injury Rehabilitation Patient perception of the program. The patient rated the program 31 out of a possible 35 points, with a score of 3 on items 3 and 5 (satisfaction with content and manageability of the program) and a score of 5 (extremely satisfied) on all other items. He stated that each station had its own unique challenges and functional outcomes. For example, he noticed and appreciated increases in hand-eye coordination, balance, and body awareness from practicing the boxing station. He also believed that using the arm ergometer was instrumental in alleviating some mild shoulder pain, stating that this was likely attributable to balanced strengthening from the push-pull motion. He also commented on the psychological benefits of exercise, reporting that the “biggest thing” about the program was that the exercise increased his mood, stating that “it [the exercise] is good for the body and, even more importantly, good for the mind.”

Discussion The aerobic exercise program appeared to be a feasible addition to the patient’s inpatient rehabilitation schedule because he was able to participate without adverse effects, was able to increase his exercise tolerance, and reported that he enjoyed the program. The upper-extremity circuit that we developed is a unique alternative to a standard arm ergometry exercise program. It provides the opportunity to increase aerobic capacity while minimizing muscular fatigue and boredom. Increased exercise tolerance, evident during both training and testing sessions, suggested gains in aerobic capacity, and orthostatic testing indicated increased orthostatic tolerance. The patient also commented on the psychological benefits of aerobic exercise. The activities and intensity used in our program were modified to con434

f

Physical Therapy

Volume 90

tinually challenge the patient’s increasing capacity for exercise (as measured by changes in HR and value on the Borg RPE). Careful adjustment of exercise prescription is of utmost importance in people with SCI because training intensities must not compromise the muscles required for wheeling and transferring yet must be high enough to induce training effects.33 The patient was able to sustain a relatively high level of physical effort during inpatient rehabilitation without compromising standard therapy or experiencing overbearing fatigue (fatigue that would consistently prohibit participation in regular therapy). Although the protocol of sessions on 3 days per week seemed to be relatively manageable at the outset of the program, participating in 3 sessions per week was often difficult. Medical complications and numerous specialist appointments outside the center impeded regular adherence to the program. These disturbances are commonly encountered during SCI rehabilitation34 and need to be recognized and accounted for to achieve the realistic implementation of any program. Thus, although a frequency of 3 sessions per week often is thought to be optimal for exercise training, it has been noted that twice-weekly programs may be more suitable for people with SCI because they offer benefits similar to those of more frequent training yet provide a prescription to which it is easier to adhere.33 The patient exhibited several signs of increased exercise tolerance (eg, sustained increases in exercise duration and intensity) throughout the program, indicating potential cardiac or peripheral skeletal muscle adaptations to the exercise training. Aerobic changes during exercise (sustained elevation of HRR and values on the Borg RPE) seemed to occur during the second half of the program. This result was likely attribut-

Number 3

able to the fact that the patient was accustomed to the exercises and, therefore, could challenge himself with respect to intensity and his ability to sustain longer exercise dura˙ O2 changed by 2 tions. His peak V mL/kg/min (with a corresponding 10-W increase in power output) from the baseline evaluation to the endpoint evaluation. Functionally, this result was previously shown to positively relate to wheeling perfor˙ O2 of 2 mance; a change in the peak V mL/kg/min was sufficient to increase the ability and decrease the time to complete a wheelchair circuit in 74 people with SCI.35 Similarly, our patient showed improvements on the 2 wheeling tests, indicating a concomitant increase in functional wheeling ability. We found that orthostatic tolerance in our patient improved after 6 weeks of aerobic exercise. Heart rate responses to orthostatic stress normally are attributable to baroreflexmediated parasympathetic (vagal) withdrawal and sympathetic activation.12 However, because of our patient’s cardiovascular sympathetic insufficiency, it is likely that any increase in HR during the orthostatic challenge was predominantly attributable to baroreceptor reflex– mediated reductions in vagal tone. Although there is some evidence to support the role of exercise in increasing baroreceptor activity in people who are able-bodied,36 similar research in people with SCI is scarce, although Engelke et al suggested that it may play a similar role in people with SCI.37 They found that exercise induced a 4-fold increase in the cardiac baroreceptor reflex response, increased the sensitivity of the aortic baroreceptor reflex control of HR, and ultimately eliminated OH.37 The deterioration of the patient’s orthostatic tolerance after an episode of UTI was likely multifactorial. On March 2010

Aerobic Exercise in Spinal Cord Injury Rehabilitation the one hand is the effect of deconditioning because the patient likely experienced a decrease in activity level subsequent to more time spent in bed as a result of increased spasticity (reflex activity) and discomfort. In fact, it was previously shown that 48 hours of bed rest can impair orthostatic tolerance in people during hospitalization.38 On the other hand, increased sweating (because of the high temperature associated with a UTI) may have contributed to a loss of intravascular volume and dehydration, thereby resulting in decreased orthostatic tolerance. Previous researchers reported unfavorable lipid profiles in people with SCI and suggested that low levels of HDL are mainly attributable to low physical activity secondary to wheelchair dependency.39 Thus, HDL correlates positively with physical activity and negatively with the risk of cardiovascular disease.15 Although De Groot et al15 demonstrated that a physical activity program enhanced the lipid profiles of people undergoing primary rehabilitation for SCI,15 Solomonow et al40 found that people with normal cholesterol levels did not exhibit changes in HDL or LDL after a 14-week exercise program.40 Because our patient’s lipid profile was within normal guidelines,41 minimal changes were observed. Like those of previously published work,15 the results of this case report suggested that being physically active can increase HDL and that this mechanism may have decreased the risk for cardiovascular disease in our patient with tetraplegia. The medical complications mentioned in this case report are common occurrences during inpatient rehabilitation for cervical SCI.34 Clinicians and scientists performing research with inpatients who have SCI need to be aware of the typical interruptions during rehabilitation. Although participation in our exerMarch 2010

cise program was complicated by various medical setbacks, our patient was able to tolerate the training sessions, suggesting that this program is a feasible addition to standard therapy. Thus, it would appear, from the results of this case report, that a patient with a motorcomplete C5 SCI is able to sustain a relatively high level of physical effort during initial rehabilitation without compromising standard therapy or experiencing overbearing fatigue. Early aerobic training has the potential to mitigate cardiovascular deterioration and metabolic alterations, thereby decreasing cardiovascular risk factors and achieving optimal physical fitness for function in daily life. Limitations The findings described in this case report have limitations that must be taken into consideration. First, our patient was a recently injured individual undergoing subacute rehabilitation; thus, physiologic changes may have been influenced by the typical rehabilitation program and natural processes of adaptation, recovery, or both. Adding elements of a controlled experiment is necessary to address internal validity concerns about the effects of history or maturation. Second, the difficulty in accurately measuring exercise intensity in people with tetraplegia cannot be overlooked. Although the Borg RPE is commonly used as an index of the physiologic response to exercise in people with SCI, no correlations have been found between ˙ O2 at high workloads and HR or V exercise intensities in people with tetraplegia.42 However, HRR generally mirrored values on the Borg RPE throughout the program, indicating that the Borg RPE provided a reflection of our patient’s exertion. Third, because we did not do any follow-up on exercise behavior af-

ter the termination of the program, we could not examine long-term exercise adherence after discharge from inpatient rehabilitation. Dr Eng, Dr Krassioukov, Dr Miller, and Ms Sproule provided concept/idea/project design. Ms Tawashy and Dr Eng provided writing. Ms Tawashy and Dr Krassioukov provided data collection. Dr Eng and Dr Miller provided data analysis. Ms Sproule provided the patient. Dr Eng, Dr Krassioukov, and Dr Miller provided consultation (including review of manuscript before submission). This project was approved by the G.F. Strong Rehabilitation Centre and the University of British Columbia research ethics boards in Vancouver, British Columbia, Canada. Dr Eng received salary support from the Michael Smith Foundation for Health Research. Data from the manuscript were presented as a poster at the Society for Neuroscience Annual Meeting; November 17, 2008; Washington, DC. This article was received January 23, 2009, and was accepted September 21, 2009. DOI: 10.2522/ptj.20090023

References 1 Tawashy AE, Eng JJ, Lin KH, et al. Physical activity is related to lower levels of pain, fatigue and depression in individuals with spinal-cord injury: a correlational study. Spinal Cord. 2008;47:301–306. 2 Hoffman MD. Cardiorespiratory fitness and training in quadriplegics and paraplegics. Sports Med. 1986;3:312–330. 3 Garshick E, Kelley A, Cohen SA, et al. A prospective assessment of mortality in chronic spinal cord injury. Spinal Cord. 2005;43:408 – 416. 4 Myers J, Lee M, Kiratli J. Cardiovascular disease in spinal cord injury: an overview of prevalence, risk, evaluation, and management. Am J Phys Med Rehabil. 2007; 86:142–152. 5 Dallmeijer AJ, Hopman MT, Angenot EL, van der Woude LH. Effect of training on physical capacity and physical strain in persons with tetraplegia. Scand J Rehabil Med. 1997;29:181–186. 6 Lee MY, Myers J, Hayes A, et al. C-reactive protein, metabolic syndrome, and insulin resistance in individuals with spinal cord injury. J Spinal Cord Med. 2005;28:20 –25. 7 American Heart Association. Metabolic syndrome. Available at http://www.ameri canheart.org/presenter.jhtml?identifier⫽ 4756. Accessed October 30, 2009.

Volume 90

Number 3

Physical Therapy f

435

Aerobic Exercise in Spinal Cord Injury Rehabilitation 8 Sidorov EV, Townson AF, Dvorak MF, et al. Orthostatic hypotension in the first month following acute spinal cord injury. Spinal Cord. 2008;46:65– 69. 9 Claydon VE, Krassioukov AV. Orthostatic hypotension and autonomic pathways after spinal cord injury. J Neurotrauma. 2006;23:1713–1725. 10 Illman A, Stiller K, Williams M. The prevalence of orthostatic hypotension during physiotherapy treatment in patients with an acute spinal cord injury. Spinal Cord. 2000;38:741–747. 11 El Masry WS. Physiological instability of the spinal cord following injury. Paraplegia. 1993;31:273–275. 12 Mathias CJ. Orthostatic hypotension: causes, mechanisms, and influencing factors. Neurology. 1995;45(suppl):S6 –S11. 13 Claydon VE, Steeves JD, Krassioukov AV. Orthostatic hypotension following spinal cord injury: understanding clinical pathophysiology. Spinal Cord. 2006;44:341– 351. 14 Saltin B, Blomqvist G, Mitchell JH, et al. Response to exercise after bed rest and after training. Circulation. 1968;38(5 suppl):VII1–VII78. 15 De Groot PCE, Hjeltnes N, Heijboer AC, et al. Effect of training intensity on physical capacity, lipid profile and insulin sensitivity in early rehabilitation of spinal cord injured individuals. Spinal Cord. 2003;41: 673– 679. 16 Bizzarini E, Saccavini M, Lipanje F, et al. Exercise prescription in subjects with spinal cord injuries. Arch Phys Med Rehabil. 2005;86:1170 –1175. 17 DiCarlo SE. Improved cardiopulmonary status after a two-month program of graded arm exercise in a patient with C6 quadriplegia: a case report. Phys Ther. 1982;62:456 – 459. 18 DiCarlo SE. Effect of arm ergometry training on wheelchair propulsion endurance of individuals with quadriplegia. Phys Ther. 1988;68:40 – 44. 19 McLean KP, Skinner JS. Effect of body training position on outcomes of an aerobic training study on individuals with quadriplegia. Arch Phys Med Rehabil. 1995;76:139 –150. 20 Marino RJ, Barros T, Biering-Sorensen F, et al. International standards for neurological classification of spinal cord injury. J Spinal Cord Med. 2003;26(suppl 1): S50 –S56.

436

f

Physical Therapy

Volume 90

21 American College of Sports Medicine. ACSM’s Guidelines for Exercise Testing and Prescription. 8th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2009. 22 Bravo G, Guizar-Sahagun G, Ibarra A, et al. Cardiovascular alterations after spinal cord injury: an overview. Curr Med Chem Cardiovasc Hematol Agents. 2004;2:133– 148. 23 Krassioukov AV, Karlsson AK, Wecht JM, et al. Assessment of autonomic dysfunction following spinal cord injury: rationale for additions to International Standards for Neurological Assessment. J Rehabil Res Dev. 2007;44:103–112. 24 Freyschuss U, Knutsson E. Cardiovascular control in man with transverse cervical cord lesions. Life Sci. 1969;8:421– 424. 25 Kirshblum S, Campagnolo DI, DeLisa JA. Spinal Cord Medicine. Philadelphia, PA: Lippincott Williams & Wilkins; 2001. 26 Borg G. Perceived exertion as an indicator of somatic stress. Scand J Rehabil Med. 1970;2:92–98. 27 Lasko-McCarthey P, Davis JA. Protocol de˙ O2max during arm cycle erpendency of V gometry in males with quadriplegia. Med Sci Sports Exerc. 1991;23:1097–1101. 28 Hol AT, Eng JJ, Miller WC, et al. Reliability and validity of the six-minute arm test for the evaluation of cardiovascular fitness in people with spinal cord injury. Arch Phys Med Rehabil. 2007;88:489 – 495. 29 Lasko-McCarthey P, Davis JA. Effect of work rate increment on peak oxygen uptake during wheelchair ergometry in men with quadriplegia. Eur J Appl Physiol Occup Physiol. 1991;63:349 –353. 30 Critikon. DINAMAP Pro Series 100 – 400 Monitor Operation Manual. Tampa, FL: Critikon, Inc; 1999. 31 Consensus Committee of the American Autonomic Society and the American Academy of Neurology. Consensus statement on the definition of orthostatic hypotension, pure autonomic failure, and multiple system atrophy. Neurology. 1996;46:1470. 32 May LA, Butt C, Minor L, et al. Measurement reliability of functional tasks for persons who self-propel a manual wheelchair. Arch Phys Med Rehabil. 2003;84:578 – 583. 33 Martin Ginis KA, Hicks AL. Considerations for the development of a physical activity guide for Canadians with physical disabilities. Appl Physiol Nutr Metab. 2007; 32(suppl):S135–S147.

Number 3

34 Chen D, Apple DF, Hudson LM, Bode R. Medical complications during acute rehabilitation following spinal cord injury: current experience of the model systems. Arch Phys Med Rehabil. 1999;80:1397– 1401. 35 Kilkens OJ, Dallmeijer AJ, Nene AV, et al. The longitudinal relation between physical capacity and wheelchair skill performance during inpatient rehabilitation of people with spinal cord injury. Arch Phys Med Rehabil. 2005;86:1575–1581. 36 Convertino VA, Adams WC. Enhanced vagal baroreflex response during 24 h after acute exercise. Am J Physiol Regul Integr Comp Physiol. 1991;260:570 –575. 37 Engelke KA, Shea JD, Doerr DF, Convertino VA. Enhanced carotid-cardiac baroreflex response and elimination of orthostatic hypotension 24 hours after acute exercise in paraplegics. Paraplegia. 1992; 30:872– 879. 38 Schneider SM, Robergs RA, Amorim FT, et al. Impaired orthostatic response in patients with type 2 diabetes mellitus after 48 hours of bed rest. Endocr Pract. 2009; 15:104 –110. 39 Dearwater SR, LaPorte RE, Robertson RJ, et al. Activity in the spinal cord-injured patient: an epidemiologic analysis of metabolic parameters. Med Sci Sports Exerc. 1986;18:541–544. 40 Solomonow M, Reisin E, Aguilar E, et al. Reciprocating gait orthosis powered with electrical muscle stimulation (RGO II), part II: medical evaluation of 70 paraplegic patients. Orthopedics. 1997;20:411– 418. 41 Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA. 2001;285:2486 – 2497. 42 Lewis JE, Nash MS, Hamm LF, et al. The relationship between perceived exertion and physiologic indicators of stress during graded arm exercise in persons with spinal cord injuries. Arch Phys Med Rehabil. 2007;88:1205–1211.

March 2010

Aerobic Exercise in Spinal Cord Injury Rehabilitation Appendix. Patient Satisfaction Scale

I am interested in how you feel about the exercise program you participated in over the past 2 months. Please tell me how you would rate the following (response categories: 1⫽not at all satisfied, 2⫽somewhat satisfied, 3⫽satisfied, 4⫽very satisfied, 5⫽extremely satisfied). 1. How satisfied were you with the frequency of the program (3 times per week)? 2. How satisfied were you with the duration of the program (30 minutes)? 3. How satisfied were you with the content of the program (6 stations)? 4. How satisfied were you with the intensity of the program (12–16 on the Borg Rating of Perceived Exertion)? For each of the following statements, please tell me how manageable/beneficial/enjoyable you found the exercise program (response categories [eg, for question 1]: 1⫽not at all manageable, 2⫽somewhat manageable, 3⫽ manageable, 4⫽very manageable, 5⫽extremely manageable). 1. How manageable is the program given an inpatient rehabilitation schedule? 2. How beneficial did you find the program? 3. How enjoyable did you find the program?

March 2010

Volume 90

Number 3

Physical Therapy f

437

Perspective

Increasing Muscle Extensibility: A Matter of Increasing Length or Modifying Sensation? Cynthia Holzman Weppler, S. Peter Magnusson C.H. Weppler, PT, MPT, is Independent Researcher, Am Honigbaum 20, 65817 Niederjosbach, Germany. Address all correspondence to Ms Weppler at: [email protected]. S.P. Magnusson, PT, DSc, is Professor, University of Copenhagen, Faculty of Health Sciences, Bispebjerg Hospital, Copenhagen, Denmark. He also is affiliated with the Institute of Sports Medicine Copenhagen and the Musculoskeletal Rehabilitation Research Unit at Bispebjerg Hospital. [Weppler CH, Magnusson SP. Increasing muscle extensibility: a matter of increasing length or modifying sensation? Phys Ther. 2010;90:438 – 449.]

Various theories have been proposed to explain increases in muscle extensibility observed after intermittent stretching. Most of these theories advocate a mechanical increase in length of the stretched muscle. More recently, a sensory theory has been proposed suggesting instead that increases in muscle extensibility are due to a modification of sensation only. Studies that evaluated the biomechanical effect of stretching showed that muscle length does increase during stretch application due to the viscoelastic properties of muscle. However, this length increase is transient, its magnitude and duration being dependent upon the duration and type of stretching applied. Most of these studies suggest that increases in muscle extensibility observed after a single stretching session and after short-term (3- to 8-week) stretching programs are due to modified sensation. The biomechanical effects of long-term (⬎8 weeks) and chronic stretching programs have not yet been evaluated. The purposes of this article are to review each of these proposed theories and to discuss the implications for research and clinical practice.

© 2010 American Physical Therapy Association

Post a Rapid Response or find The Bottom Line: www.ptjournal.org 438

f

Physical Therapy

Volume 90

Number 3

March 2010

Increasing Muscle Extensibility

V

arious theories have been proposed to explain increases in muscle extensibility observed after intermittent stretching. Most of these theories suggest a mechanical increase in length of the stretched muscle. The mechanical theories include viscoelastic deformation, plastic deformation, increased sarcomeres in series, and neuromuscular relaxation. More recently, a sensory theory has been proposed suggesting instead that increases in muscle extensibility are due to a modification of sensation only. The purposes of this article are to review each of these theories and to discuss the implications for research and clinical practice.

Muscle Length, Length Measurements, and Muscle Extensibility According to the science of biomechanics, muscle length is multidimensional.1 Length measurements are only one dimension of muscle length. When more than one dimension is included in muscle length assessment, important biomechanical properties of the muscle can be determined. These additional dimensions include tension, cross-sectional area, and time. From these added dimensions, the biomechanical properties of stiffness, compliance, energy, hysteresis, stress, viscoelastic stress relaxation (VESR), and creep can be derived (Table).1,2 Because muscle comprises deformable material, its length measurement at a given moment in time is always dependent upon the amount of tensile force (force that pulls the specimen in the direction of elongation) applied.1 Tension is the passive resistance of the muscle being stretched and is equal to the applied tensile force. The relationship between length and tension can be described by a passive length/tension curve on which multiple individual

March 2010

length measurements are plotted according to the amount of passive tension required to reach each measurement.1,2 Human muscle length measurements are, with few exceptions, measurements of joint angles, and the tensile force is applied in a rotational manner (ie, a torque). For this reason, length/tension curves are commonly presented as torque/angle curves in human studies. Physical therapy texts describe techniques for measuring muscle length in human subjects. However, this is traditionally presented as a one-dimensional concept of muscle length, describing only the measurement of end-range joint angles, and does not clearly distinguish between the single and multi-dimensional concepts of muscle length. Throughout this perspective article, one-dimensional measurement of muscle length will be referred to as “muscle extensibility.” The term “muscle length” will be reserved to refer to the multidimensional concept of length as a function of tension. For the purposes of this article, muscle extensibility is defined as the ability of a muscle to extend to a predetermined endpoint. The endpoint of stretch varies depending on the intent of the study. In human research, this endpoint is most often subject sensation. For this reason, when referring to human studies throughout this article, the term “extensibility” assumes an endpoint of subject sensation unless otherwise noted. Skeletal muscles comprise contractile tissue intricately woven together by fibrous connective tissue that gradually blends into tendons. The tendons are made of fibrous connective tissue and attach the muscle to bone.3 Although the contractile tissue and tendons are sometimes evaluated separately for research purposes, they cannot be separated

during routine clinical testing and stretching procedures, nor during functional activity. Both the muscular contractile tissue and tendon exhibit changes in biomechanical properties and cross-sectional area in response to exercise, disuse and aging.4 For these reasons, the term “muscle” is used in this article to indicate the entire skeletal muscle, including the contractile tissue and tendon components. Animal studies of muscle length are able to purely test the mechanical properties of the muscle-tendon unit (MTU) as other overlying and adjoining tissues—skin, connective tissue, muscles, and neurovascular structures— can be surgically reflected. These tissues remain fully intact during human muscle length testing, so the passive resistance and extensibility measured may not be attributable solely to the tested muscles.5–11 When assessing muscles that cross at least 2 joints in human subjects, each joint can be tested separately to ensure that a joint restriction is not responsible for motion limitations and end-range passive resistance. With appropriate joint positioning, the stretched muscle can be placed under maximal stretch,12 ensuring that the passive resistance to stretch is due primarily to the muscle being stretched and conjoining soft tissues. However, when testing muscles that cross only one joint, it may not be possible to determine to what degree the joint itself and its capsular structures contribute to extensibility limitations and passive resistance.7 Available With This Article at ptjournal.apta.org • Audio Abstracts Podcast This article was published ahead of print on January 14, 2010, at ptjournal.apta.org.

Volume 90

Number 3

Physical Therapy f

439

Increasing Muscle Extensibility Table. Muscle Length Dimensions and Biomechanical Properties That Can Be Derived From Each Added Dimension Muscle Length Dimension

Biomechanical Property

Length measurement

Muscle extensibility: ability of a muscle to extend to a predetermined endpoint. When referring to human studies, “extensibility” assumes an endpoint of subject sensation unless otherwise noted.

Tension

Stiffness: change in tension per unit change in length Compliance: change in length per unit change in tension Energy: area under the length/tension curve Hysteresis: energy dissipated during the unloading phase

Cross-sectional area

Time

Stress: tension per unit of cross-sectional area Stiffness, compliance, energy, and hysteresis normalized for muscle thickness Viscoelastic stress relaxation: decrease in resistance that occurs during a passively applied static stretch, the percentage difference between peak and final torque Creep: increase in muscle length as applied force is held constant

Figure 1. Model of shifting length/tension curve. When a change in muscle length occurs, there is a shift in the entire passive length/tension curve. When “shortening” occurs, the curve shifts to the left, reflecting shorter muscle length measurements at a given passive tensile force. When lengthening occurs, the curve shifts to the right, reflecting a longer muscle length measurement at a given passive tensile force. Note: Number values are absolute; curve is a theoretical illustration.

Increasing Muscle Extensibility Increases in human muscle extensibility are demonstrated by an increase in end-range joint angles. When an increase in muscle extensibility is observed, it is possible that the increase is due to a simple decrease in muscle stiffness or an increase in muscle length. A simple 440

f

Physical Therapy

Volume 90

decrease in muscle stiffness is demonstrated by a decrease in the slope of the torque/angle curve. Increases in muscle length are reflected on the torque/angle curve by a shift to the right of the entire curve.1,2,12,13 This right shift results in decreased stiffness and an increased length measurement (joint angle) for any given tension (Fig. 1). Muscle extensibility

Number 3

also can increase—without a change in muscle length or stiffness— due to a simple increase in applied tension, which causes the muscle to stretch further (Fig. 2). Without information about applied tension, there is no way to differentiate between these possibilities.14,15

Mechanical Theories for Increasing Muscle Extensibility The rehabilitation literature often suggests that increases in muscle extensibility observed after intermittent stretching involve an increased mechanical length of the stretched muscle. These mechanical theories include viscoelastic deformation, plastic deformation, increased sarcomeres in series, and neuromuscular relaxation. Viscoelastic Deformation Many human studies11,16 –18 suggest that increases in muscle extensibility observed immediately after stretching are due to a lasting viscoelastic deformation. Skeletal muscles are considered to be viscoelastic. Like solid materials, they demonstrate elasticity by resuming their original length once tensile force is removed. Yet, like liquids, they also behave viscously because their response to tensile force is rate and time dependent.1,14 An immediate increase in muscle length can occur due to the viscous behavior of muscles whenever they undergo stretch of sufficient magnitude and duration or frequency. This increased length is a viscoelastic deformation because its magnitude and duration are limited by muscles’ inherent elasticity.1 Viscoelastic deformation has been tested in research using various stretching methods such as “static” (constant joint angle) stretches,19 –23 constant load,24 contract/relax,25 and repeated cyclic stretches.23,26 Static stretching can be used to evaluate the property of viscoelastic

March 2010

Increasing Muscle Extensibility stress relaxation. When stretch is applied to a muscle and the muscle is held in the stretched position for a period of time, as is the case with normal static stretching techniques, the muscle’s resistance to stretch gradually declines (Fig. 3).1,2,14,27 This decline in resistance to stretch is called viscoelastic stress relaxation and is expressed as a percentage of the initial resistance.14,19,20 Constant load stretching, such as stretching that uses a fixed torque, can be used to evaluate the property of creep. Creep occurs when mechanical length gradually increases in response to a constant stretching force.1,2,23 The study most commonly used to support the theory that viscoelastic deformation is responsible for increases in human muscle extensibility is an animal study by Taylor et al.23 The results of this study showed an immediate increase in MTU length induced by repeated cyclic and static stretches. The authors suggested that the observed length increases should be lasting due to the viscous properties of the MTU.23 However, no further testing was performed to determine the duration and residual magnitude of these length increases.23 In human studies, viscoelastic deformation and recovery have been tested on hamstring and ankle plantar-flexor muscles.20 –22,24,28 The results refute viscoelastic deformation as a mechanism for lasting increases in muscle length and extensibility. These studies showed that the magnitude and duration of the length increases vary depending on the duration of the stretch and the type of stretching applied. All of these studies consistently showed viscoelastic deformation of human muscle to be transient in nature. With stretch application typical of that practiced in rehabilitation and sports, the biomechanical effect of March 2010

Figure 2. No shift in passive length/tension curve model. When there is no mechanical change in muscle length after an intervention, there is no shift in passive length/tension curves, and tension required to achieve the preintervention muscle length measurement remains unchanged. In human studies, if the endpoint of the stretch is determined by subject sensation, the increased measurements may be attributed to sensory modification. Note: Number values are absolute; curve is a theoretical illustration.

viscoelastic deformation can be quite minimal and so short-lived that it may have no influence on subsequent stretches. In one hamstring muscle study, a static stretch of 45 seconds’ duration was found to have no significant effect on the next stretch performed 30 seconds later.28 With 3 consecutive 45-second static stretches (30-second rest intervals between stretches), each stretch showed VESR of 20% during the static holding phase. However, the muscles had already recovered from the relaxation by the next stretch.28 Similar results were demonstrated in a study of ankle plantar-flexor muscles.21 There was no change in stiffness of the ankle plantar-flexor muscles that underwent static stretches of: (1) 4 sets of 15 seconds’ duration and (2) 2 sets of 30 seconds’ duration (10-second rest intervals between stretches).21

ter stretching are due to “plastic,”17,29 –32 or “permanent”17,30 –36 deformation of connective tissue.37 The classical model of plastic deformation would require a stretch intensity sufficient to pull connective tissue within the muscle past the elastic limit and into the plastic region of the torque/angle curve so that once the stretching force is removed, the muscle would not return to its original length but would remain permanently in a lengthened state (Fig. 4).1,2 In 10 studies17,29 –37 that suggested plastic, permanent, or lasting deformation of connective tissue as a factor for increased muscle extensibility, none of the cited evidence was found to support this classic model of plastic deformation. The term “plastic deformation” often was considered only to be a synonym for deformation that is permanent in nature.31,32

Plastic Deformation of Connective Tissue Another popular theory suggests that increases in human muscle extensibility observed immediately af-

The evidence cited29 –31,33–35,37 in support of this theory can be traced almost entirely to a study by Warren et al38 performed on rat tail tendons and a review article by Sapega et al.32

Volume 90

Number 3

Physical Therapy f

441

Increasing Muscle Extensibility There was no evidence of a classic plastic deformation phase occurring in any of the cited studies.

Figure 3. Viscoelastic stress relaxation during static stretch. Peak torque occurs when muscle first reaches the final stretch position. Final torque is measured at the end of the static stretch holding phase. Viscoelastic stress relaxation (delta torque) is the decrease in torque and can be expressed as a percentage of peak torque: (peak torque ⫺ final torque) ⫼ peak torque. Reprinted with permission of Wiley-Blackwell from: Magnusson SP. Passive properties of human skeletal muscle during stretch maneuvers: a review. Scand J Med Sci Sports. 1998;8:65–77.

Neither of these works recommended the classical model of plastic deformation, which requires high stretching loads, but instead suggested viscoelastic deformation: using lower stretching loads with prolonged stretch duration in order to facilitate “viscous flow” within the connective tissue.

442

f

Physical Therapy

Volume 90

Although model passive length/tension curves that include a plastic deformation phase may be applicable for some types of biological tissue, studies of muscle demonstrate a markedly different typical curve. A plastic deformation phase would be reflected on the passive length/tension curve by a decrease in its slope.2

Number 3

Increased Sarcomeres in Series Animal studies have demonstrated that the number of sarcomeres in series of a muscle can be changed by prolonged immobilization in extreme positions. That is, when muscles are immobilized in fully extended positions, there is an increase in the number of sarcomeres in series. Although often reported otherwise, these muscles demonstrated no overall change in muscle length because increases in the number of sarcomeres in series were offset by a concurrent decrease in sarcomere length.39 – 41 When muscles are immobilized in shortened positions, there is a decrease in the number of sarcomeres in series and a concurrent decrease in muscle length.39 – 41 Sarcomere number and muscle length in the shortened muscles have been found to increase to normal levels after recovery from immobilization.39,40 These animal studies suggest that muscles adapt to new functional lengths by changing the number and length of sarcomeres in series in order to optimize force production at the new functional length.39,41 Despite substantial differences between muscle immobilization and intermittent stretching, this research has been generalized to suggest that short-term (3- to 8-week) human stretching regimens cause similar increases in sarcomeres in series and a concurrent increase in length of the stretched muscles.7,11,12,17,31,42– 45 For obvious practical and ethical reasons, there are no human stretching studies that evaluated on a histological level whether the number of sarcomeres in series changes due to therapeutic intervention. Perhaps with development of imaging techniques, this will someday be a possibility.

March 2010

Increasing Muscle Extensibility Neuromuscular Relaxation The rehabilitation literature often suggests that involuntary contraction of muscles due to a neuromuscular “stretch reflex” can limit muscle elongation during static stretching procedures.16,33,34,46 –50 In order to increase muscle extensibility, it often has been proposed that slowly applied static stretch (used alone or in combination with therapeutic techniques associated with proprioceptive neuromuscular facilitation) stimulates neuromuscular reflexes that induce relaxation of muscles undergoing static stretch.16,17,34,36,46,47,49 –53 Some authors,34,53 furthermore, have suggested that neuromuscular reflexes adapt to repeated stretch over time, which enhances the stretched muscles’ ability to relax and results in increased muscle extensibility. Experimental evidence does not support any of these assertions.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 asymptomatic and whose muscles were subjected to a long, slow, passive stretch into endrange positions did not demonstrate significant activation of stretched muscles.14,54,56,57 Even studies that simulated ballistic (cyclic and highvelocity) 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” stretch25 and in short-term (3 and 6 weeks’ duration) stretching studies,45,58 no significant electromyographic activity of the stretched muscles was found and no shift of passive torque/angle curves was observed. The increase in end-range joint angles, therefore, could not be attributed to neuromuscular relaxation.14,25,45,58

March 2010

Figure 4. Model passive length/tension curve for biological tissue. The elastic region begins at the initial length and ends at the elastic limit. Increases in the length measurement due to applied tensile force are temporary. When tensile force is removed, the specimen will resume its original length. In the plastic region, with application of tensile force beyond the elastic limit, permanent deformation of the specimen will occur. In this case, once the tensile force is removed, the specimen will not return to its original length. Failure or rupture point is the last point on the curve. The length attained at the rupture point is the maximum length of the specimen. Note: Number values are absolute; curve is a theoretical illustration.

Sensory Theory for Increasing Muscle Extensibility In the early 1990s, several researchers put these mechanical theories to the test by assessing the biomechanical effects of stretching. By including the dimension of tension in muscle length evaluation, they were able to construct torque/angle curves and assess biomechanical properties of the muscles before and after stretching. If the increases in muscle extensibility observed after stretching were due to an increase in length of the muscles caused by any of these mechanical explanations, there should have been a lasting right shift in passive torque/angle curves (Fig. 1). Instead, the only change observed in passive torque/angle curves was an increase in end-range joint angles and applied torque (Fig. 2).15,25,58,59 Because the endpoint of these stretches was subject sensation (pain onset,15,25,26,58 maximum stretch,59 or maximum pain tolerated15), the only observable expla-

nation for these results was that subjects’ perception of the selected sensation occurred later in stretch application. These studies suggest that increases in muscle extensibility observed immediately after stretching and after short-term (3- to 8-week) stretching programs are due to an alteration of sensation only and not to an increase in muscle length.15,25,53,59 This theory is referred to as the sensory theory throughout this article because the change in subjects’ perception of sensation is the only current explanation for these results. To what extent this adaptation is a peripheral or central phenomenon or a combination thereof remains to be established. It is possible that psychological factors also play a role in the observed increases in muscle extensibility. Because there is no way to keep subjects from knowing that they are participating in a stretching study,

Volume 90

Number 3

Physical Therapy f

443

Increasing Muscle Extensibility subjects may demonstrate an increase in extensibility because they expect this to be the result of stretching. Increased extensibility then may be due to a psychological alteration in sensory perception or to a willingness of subjects to tolerate greater torque application.16,19,42,47 Single Stretching Session A modification in sensation that occurs immediately after a single stretching session was first reported in 1996 in a study by Magnusson et al25 (using a passive knee extension test involving a Kin-Com dynamometer* with a modified thigh pad) that investigated the effects of a 10second static hold versus a single contract/relax stretch on human hamstring muscles. Halbertsma et al,59 using an instrumented passive straight-leg raise method, tested subjects just prior to and immediately following a 10-minute hamstring muscle stretching session. In both of these studies, there was no shift of passive torque/angle curves (Fig. 1). but an increase in applied torque and increased end-range joint angles were observed (Fig. 2). Subsequent studies showed that sensory perception in response to stretching of human hamstring muscles is acutely modified by assuming a stooped versus an upright trunk position60 and is similarly modified by a single 90second static stretch and 10 repeated cyclic (“ballistic”) stretches.26 Short-Term (3- to 8-Week) Stretching Programs In a study investigating the biomechanical effects of a 4-week hamstring muscle stretching program, Halbertsma and Go ¨ eken15 (using an instrumented passive straight-leg raise test) found that sensations of pain onset and pain or stretch tolerance occurred at increased torques, resulting in increases in hamstring * Isokinetic International, 6426 Morning Glory Dr, Harrison. TN 37341.

444

f

Physical Therapy

Volume 90

muscle extensibility (pain onset: mean increase⫽10°, 95% confidence interval⫽5°–14°; pain or stretch tolerance: mean increase⫽5°, 95% confidence interval⫽1°–9°). Concurrently, no shift of passive torque/ angle curves was observed. Magnusson et al58 reported similar results in a 3-week hamstring muscle stretching study that induced a 17degree mean increase in end-range joint angles using an endpoint of pain onset. These results—no shift in passive torque/angle curves accompanied by increases in end-range joint angles— have been supported repeatedly in studies involving hamstring muscles and using various stretching and testing methods.17,42,43,45,61,62 Modification of subjects’ sensory response to stretch after short-term stretching programs also has been demonstrated in the rectus femoris muscle63 and in ankle plantar-flexor muscles.44,64 Studies involving subjects with spinal cord injuries showed no evidence of a shift in torque/angle curves after 4-week programs of sustained 30minute daily stretching of hamstring65 and ankle plantar-flexor66 muscles, further supporting the notion that short-term stretching does not alter torque/angle relationships. Long-Term (>8 Weeks) and Chronic Stretching Programs The effect of longer-term stretching programs (⬎8 weeks) and rigorous chronic stretching regimens on passive torque/angle curves has not yet been evaluated.14

Conflicts in Research Conflicting Terminology Throughout the rehabilitation literature regarding the effects of stretching, confusion arises due to inconsistent use of terminology among studies. Some of the above-cited studies confirmed that increases in muscle extensibility occurred after stretching, whereas others claimed that muscle extensibility did not in-

Number 3

crease. On the surface, it appears that these studies had conflicting results, but the difference merely resides in the definition of muscle extensibility. The studies reporting increases in muscle extensibility used a sensory endpoint, which indicates that the selected sensation had onset later during stretch application, allowing increases in end-range joint angles.15,59 The studies reporting no increases in muscle extensibility used an endpoint of standardized torque, which gives some evidence that there was no shift of the torque/angle curves or change in muscle stiffness.61,62,65 Taken together, the findings of all of these studies support the sensory theory to explain increases in end-range joint angles. Conflicting Interpretations Although the results of many of the supporting studies were similar, not all of these studies attributed the findings to a change in sensory perception.43– 45,64 Some studies43– 45 suggest instead that because there is increased applied torque, a longer torque/angle curve, and increased end-range joint angles, the stretching program has induced structural changes within the muscle, such as increased sarcomeres in series and a mechanical increase in muscle length. If an increase in the number of serial sarcomeres is accompanied by an increase in length of the muscle, there should be an observable right shift in the entire passive torque/angle curve similar to the shift in passive length/tension curves shown in the animal studies (Fig. 1).39,40,67 Without a concurrent right shift in passive torque/angle curves, there is no evidence of an increase in muscle length. The theory of structural adaptation occurring after a short-term stretching program also does not explain similar increases seen immediately after a single stretching session that occur

March 2010

Increasing Muscle Extensibility without a regimented stretching program.14,25,59,60 Conflicting Results Although there is growing evidence to support the theory that increases in muscle extensibility observed after stretching are due to modified sensation only, there are a few conflicting reports. In a study of ankle plantar-flexor muscles, Guissard and Duchateau68 observed a right shift of torque/angle curves that occurred over a 6-week training period. This result may have been due to the vigorous design of the stretching program, which was performed 5 days per week and took 20 minutes to complete. This stretching dosage for a single muscle group on a single limb is well in excess of the 15 to 150 seconds29,44,64,69,70 of daily stretch typically used in sports and research but may be applicable in rehabilitation settings. Thirty days after the stretching program ended, increases in extensibility and muscle length were partially maintained. More research is needed to determine: (1) whether increases in muscle length are an appropriate and desirable outcome of treatment and (2) the most efficient therapeutic intervention and dosage to induce and maintain length increases.

A Multidimensional Approach to Evaluating Muscle Length The sensory theory of increasing muscle extensibility demonstrates how multidimensional muscle length testing can enhance basic knowledge about muscle adaptation. Evaluation techniques that include multiple dimensions of muscle length (eg, extensibility measurements, torque and cross-sectional area) provide tools to better assess muscle status and the effect of therapeutic interventions.

March 2010

Muscle Extensibility Muscle extensibility is a critical dimension of muscle length. Tests of extensibility (traditionally called “muscle length tests”) were developed with the idea that there is a “normal” or ideal range of muscle extensibility that promotes optimal kinematics, resulting in efficient motion, enhancing the ability to adapt to imposed stresses, and potentially decreasing the risk of injury.71,72 It is suggested that when a particular muscle or muscle group demonstrates insufficient extensibility (appearing to be “short”), motion between joint surfaces that the muscle crosses may be limited, resulting in restricted joint motion. When the muscle or muscle group demonstrates excessive extensibility (appearing to be “long”), motion between the joint surfaces also may be excessive, resulting in excessive joint motion. Whether insufficient or excessive, a deviation from optimal extensibility is thought to precipitate unusual wear patterns on capsular structures and articular surfaces of involved joints. It is suggested that deviations from optimal extensibility contribute to muscle imbalances, faulty posture, and dysfunctional movement.71,73 Although guidelines for what constitutes insufficient, optimal, and excessive extensibility measurements are based on the science of kinematics, their clinical validity has rarely been studied. Although kinematic analysis is concerned with the motion that occurs at the joint and can identify the clinical extensibility measurements that are theoretically optimal, it is not concerned with analyzing the forces causing the motion.1,72 Except in cases where bony approximation is the limiting factor, this type of analysis does not clearly define what should constitute the endpoint of stretch application. Perhaps this is

the reason that stretch endpoints often are poorly defined and inconsistent among texts and research studies and in clinical practice. An endpoint of the examiner’s perception of “(firm) resistance” is suggested in some texts72,74 and often is used in research, although many studies did not measure the amount of applied torque required to reach this point.35,49,53,69,70 The validity and reliability of this endpoint are highly questionable because without quantitative measurement, there is no way to be certain that torque is being applied consistently.75 There also is evidence that the amount of torque applied by trained therapists can vary markedly—as much as 40fold for a single subject.76 Even if torque were standardized, how would the most clinically relevant torque be determined? The importance of subject sensation as an endpoint has largely been overlooked. To date, endpoints of subject sensation are widely used in research, but basic texts describing muscle extensibility assessment have not clearly and unequivocally made this recommendation. Is passive muscle stiffness necessary to stop joint motion, or is it possible that just the subject’s sensory perception of stiffness or perception of moderate stretch can be a limiting factor? Studies evaluating the biomechanical effects of stretching reveal that in controlled clinical settings under the condition of slowly applied passive stretch, it is subject sensation—not the degree of stiffness—that limits joint motion. Researchers have been able to apply passive torque up to the sensory endpoint of pain25,58 or stretch tolerance15,59 without being limited by stiffness. It seems reasonable that subject sensation could both alter and reflect the way the tested muscle is routinely used in function. Further research is needed to determine whether subject sensa-

Volume 90

Number 3

Physical Therapy f

445

Increasing Muscle Extensibility tion is a significant factor in limiting joint motion during functional motion and whether muscle extensibility measurements are truly a reflection of the way muscles are used during function. Although subject sensation is the most frequently used endpoint in human stretching research, there is little consensus regarding which sensation is most relevant clinically. A wide range of sensations has been used in research, from the subject’s perception of a “pull,”47 to varying degrees of: • • • • • •

“resistance”16,37 “stretch”15,17,19,20,22,42,44,45,59,63 “discomfort”7,29,33,69 –71 “tightness”25,26,28,58,69,70,77 “stiffness”78 “pain”7,15,18,25,26,58,60,64,79

A change in endpoints from detection of the “first sensation of pain” to pain or stretch tolerance can result in a change in end-range joint angles that varies markedly among subjects.15 This has been demonstrated in the hamstring musculature of otherwise “normal” subjects assessed with “short” hamstring muscles to range anywhere from no change at all to a 20-degree increase in endrange joint angle values.15 Further research is needed to assess which sensation is most clinically relevant. Subject sensation is—at the very least—an important endpoint of the torque/angle curve and may give information regarding how the muscle is routinely used during functional activity. However, extensibility measurements alone are only one dimension of muscle length and may not accurately reflect the actual length of the muscle. This has been demonstrated repeatedly in studies that included evaluation of torque80 and cross-sectional area.81,82 Torque and cross-sectional area measurements provide critical information that allows a more precise muscle length assessment. 446

f

Physical Therapy

Volume 90

Torque When the dimension of tension is included in muscle length evaluation, a passive torque/angle curve can be constructed. This curve shows the relationship between individual extensibility measurements and the torque required to attain each measurement (the torque/angle relationship). Using this curve, important biomechanical properties such as stiffness, compliance, energy, and hysteresis can be assessed.2 This information allows evaluation of an individual’s muscle for comparison before and after an intervention, thus showing the effect of the intervention on the tested muscle’s biomechanical properties. Use of this type of testing led to the development of the sensory theory.15,25,58,59 Torque/angle curves, however, may not fully reflect actual muscle length. Torque measurements quantify a muscle’s resistance to passive stretch, and this resistance is partly determined by thickness of the muscle. Other factors being equal, a thicker muscle demonstrates increased stiffness at a given joint angle, which causes the muscle to appear shorter on a torque/angle curve. A thinner muscle, other factors being equal, demonstrates decreased stiffness at a given joint angle, causing the muscle to appear longer on a torque/angle curve. In order to evaluate the contribution of muscle thickness to passive resistance, measurement of crosssectional area is required. Cross-Sectional Area Measurement of cross-sectional area is, by itself, valuable. Changes in cross-sectional area indicate an intervention effect of muscle hypertrophy (when increased) and atrophy (when decreased). When assessment of muscle crosssectional area is combined with

Number 3

torque and joint angle, the biomechanical properties of muscles of different thicknesses can be compared.1,81,82 Measurements of stress (tension/cross-sectional area), as well as normalized stiffness, compliance, energy, and hysteresis values, can be derived. These normalized values allow researchers to determine to what degree muscle crosssectional area contributes to observed passive resistance and biomechanical properties.

Implications for Research and Clinical Practice Despite its fundamental role in rehabilitation, as well as sports and fitness, very little is actually known about muscle length: what constitutes optimal extensibility, torque/ angle parameters, and crosssectional area. Future research is needed to address which biomechanical properties and measures (or combination thereof) reflect an optimal muscle length. An optimal muscle length would allow not only an optimal range of muscle extensibility and joint motion but also optimal tendon length, overlap of contractile tissue filaments, and overall muscle thickness so that the muscle can generate the amount of passive and active tension required during function. Further research could address the extensibility, torque/angle relationship, and cross-sectional area considered optimal and how these parameters vary among individuals, between the sexes, over the lifespan, and for various muscles and subject groups. With continued research, muscle length disorders may someday be more precisely assessed, allowing selection of the intervention that will best address the specific disorder. This research also has relevance in developing general fitness guidelines. For example, a muscle that is too short is operating in a range that is left of optimal torque/angle curves. March 2010

Increasing Muscle Extensibility Clinically, this would be considered to be a “contracture.” Whether a muscle exhibits decreased extensibility or if it is truly shortened cannot be determined by extensibility measures alone. Two studies on ankle plantar-flexor muscles compared different subject groups (elderly women7 and subjects diagnosed with diabetes mellitus and peripheral neuropathies80) with control subjects and found that the test groups exhibited decreased extensibility but that torque/angle curves, besides being shorter, were not significantly different. These findings would suggest that the test subjects’ muscles were not actually shortened (torque/angle curves were not shifted left) but were lacking in extensibility. The commonly prescribed treatment of stretching would address this clinical problem by increasing extensibility without shifting torque/angle curves. Neither of these studies assessed cross-sectional area of the tested muscles, however, so it is not known to what extent muscle thickness contributed to passive resistance. For example, it is possible that shortened muscles combined with decreased cross-sectional area could have confounded the results.80 Another study involving male endurance athletes found that subjects whose hamstring muscles were classified as “tight” did have passive torque/angle curves that were shifted left compared with control subjects’ hamstring muscles.82 Both groups were similar in age, height, weight, training history, and hamstring muscle cross-sectional area. Decreased hamstring muscle extensibility also has been associated with a left shift in active torque/angle curves.83 Does this left shift in torque/angle curves predispose the subjects to be less efficient in functional motion or more prone to musculoskeletal pain syndromes and injury? Does the change in biomeMarch 2010

chanical properties that accompanied the shorter muscles enhance or detract from functional performance? It appears that standard dosage stretching regimens do not change the torque/angle relationship in the short term, and 8 of 10 of the athletes with “tight” hamstring muscles were already performing regular stretching exercises. If the left shifted position was found to be detrimental, the challenge would be to find a therapeutic intervention that would induce a lasting right shift of these subjects’ torque/angle curves. A study involving subjects diagnosed with benign joint hypermobility syndrome (BJHS) suggests that hamstring muscles attaining greater than “normal” extensibility may not actually be longer than those of “normal” control subjects.81 Both groups were matched for age and sex and were similar in hamstring muscle crosssectional area. The biomechanical properties (VESR and passive energy absorption [area under the stress/angle curve] at mutual joint angles) of the subjects with BJHS were not significantly different from those of “normal” controls.1,81 The excessive range of muscle extensibility in the subjects diagnosed with BJHS was attributed to altered sensation and not to mechanically longer muscles.81 Using just the end-range joint angles as a guide, the typical treatment recommendation would be strengthening of the muscles and avoiding stretched positions.71 In this case, the biomechanical analysis suggests that the primary problem is sensory in nature—a late onset of sensation in response to stretch. Strengthening has been shown to affect torque/angle curves by increasing passive stiffness84 but would not address the sensory problem. Instead, perhaps treatment should focus primarily on avoidance of overstretching the muscle. It is not known whether the sensory perception of stretch could return to an

optimal range over time with appropriate treatment and adherence to kinematic guidelines. These examples suggest 3 different potential muscle length disorders and how treatment can be specifically directed to address them. As research continues, there are likely to be more disorders of muscle length (involving different combinations of altered extensibility, torque/ angle curves, and cross-sectional area) discovered that may be able to explain clinical anomalies. One possible example could be an athletic subject with a history of recurrent hamstring muscle strains who stretches regularly and demonstrates an optimal range of extensibility. Perhaps the root of the problem is a torque/angle curve that is left of optimal. The challenge, once again, would be to find an intervention that can induce a lasting shift in the torque/angle curve toward an optimal range. Multidimensional muscle length testing also can be important in developing fitness guidelines. There is a growing popularity of various exercise regimens that encourage stretching to a degree considered excessive by kinematic analysis. Little is known about the short- and long-term effects of this type of stretching and what accounts for the increased extensibility it induces. Are the adaptations sensory in nature, as was suggested by the study of subjects diagnosed with BJHS, or is there a long-term increase in muscle length or a change in other biomechanical properties? Are these adaptations reversible once this type of stretching is stopped? Are these adaptations desirable, despite kinematic evidence to the contrary?

Conclusion Traditionally, rehabilitation literature has attributed increases in muscle extensibility observed after stretching to a mechanical increase in muscle length. A growing body of re-

Volume 90

Number 3

Physical Therapy f

447

Increasing Muscle Extensibility search refutes these mechanical theories, suggesting instead that in subjects who are asymptomatic, increases in muscle extensibility observed immediately after a single stretching session and after shortterm (3- to 8-week) stretching regimens are predominantly due to modification in subjects’ sensation. This research brings to light the importance of using sensory endpoints when assessing muscle extensibility, the value of multidimensional muscle length assessment, and the need for basic research in this field. Multidimensional evaluation of muscle length can lead to a more comprehensive and effective approach to addressing disorders of muscle length and has application in developing fitness guidelines. Both authors provided concept/idea/project design and writing. Ms Weppler collected and analyzed information. Dr Magnusson provided consultation (including review of manuscript before submission). This article was received January 15, 2009, and was accepted October 12, 2009. DOI: 10.2522/ptj.20090012

References ¨ zkaya N, Nordin M. Fundamentals of 1 O Biomechanics: Equilibrium, Motion, and Deformation. 2nd ed. New York, NY: Springer; 1999. 2 Enoka RM. Neuromechanics of Human Movement. 3rd ed. Champaign, IL: Human Kinetics Publishers; 2002. 3 Hollinshead WH, Rosse CM. Textbook of Anatomy. 4th ed. Baltimore, MD: Lippincott Williams & Wilkins; 1985. 4 Magnusson SP, Narici MV, Maganaris CN, Kjaer M. Human tendon behaviour and adaptation, in vivo. J Physiol (Lond). 2008; 586:71– 81. 5 Gajdosik RL, Rieck MA, Sullivan DK, Wightman SE. Comparison of four clinical tests for assessing hamstring muscle length. J Orthop Sports Phys Ther. 1993; 18:614 – 618. 6 Gajdosik RL, Lusin G. Hamstring muscle tightness: reliability of an active-kneeextension test. Phys Ther. 1983;63: 1085–1090. 7 Gajdosik RL, Vander Linden DW, Williams AK. Influence of age on length and passive elastic stiffness characteristics of the calf muscle-tendon unit of women. Phys Ther. 1999;79:827– 838.

448

f

Physical Therapy

Volume 90

8 Johnson EK, Chiarello CM. The slump test: the effects of head and lower extremity position on knee extension. J Orthop Sports Phys Ther. 1997;26:310 –317. 9 Kornberg C, Lew P. The effect of stretching neural structures on grade one hamstring injuries. J Orthop Sports Phys Ther. 1989;10:481– 487. 10 Turl SE, George KP. Adverse neural tension: a factor in repetitive hamstring strain? J Orthop Sports Phys Ther. 1998; 27:16 –21. 11 Webright WG, Randolph BJ, Perrin DH. Comparison of nonballistic active knee extension in neural slump position and static stretch techniques on hamstring flexibility. J Orthop Sports Phys Ther. 1997;26: 7–13. 12 Gajdosik RL. Passive extensibility of skeletal muscle: review of the literature with clinical implications. Clin Biomech (Bristol, Avon). 2001;16:87–101. 13 Liebesman J, Cafarelli E. Physiology of range of motion in human joints: a critical review. Crit Rev Phys Rehabil Med. 1994; 6:131–160. 14 Magnusson SP. Passive properties of human skeletal muscle during stretch maneuvers: a review. Scand J Med Sci Sports. 1998;8:65–77. 15 Halbertsma JP, Go ¨ eken LN. Stretching exercises: effect on passive extensibility and stiffness in short hamstrings of healthy subjects. Arch Phys Med Rehabil. 1994; 75:976 –981. 16 de Weijer VC, Gorniak GC, Shamus E. The effect of static stretch and warm-up exercise on hamstring length over the course of 24 hours. J Orthop Sports Phys Ther. 2003;33:727–733. 17 Chan SP, Hong Y, Robinson PD. Flexibility and passive resistance of the hamstrings of young adults using two different static stretching protocols. Scand J Med Sci Sports. 2001;11:81– 86. 18 Willy RW, Kyle BA, Moore SA, Chleboun GS. Effect of cessation and resumption of static hamstring muscle stretching on joint range of motion. J Orthop Sports Phys Ther. 2001;31:138 –144. 19 Magnusson SP, Simonsen EB, Aagaard P, et al. Viscoelastic response to repeated static stretching in the human hamstring muscle. Scand J Med Sci Sports. 1995;5: 342–347. 20 Magnusson SP, Simonsen EB, Aagaard P, Kjaer M. Biomechanical responses to repeated stretches in human hamstring muscle in vivo. Am J Sports Med. 1996;24: 622– 628. 21 McNair PJ, Dombroski EW, Hewson DJ, Stanley SN. Stretching at the ankle joint: viscoelastic responses to holds and continuous passive motion. Med Sci Sports Exerc. 2001;33:354 –358. 22 Duong B, Low M, Moseley AM, et al. Time course of stress relaxation and recovery in human ankles. Clin Biomech (Bristol, Avon). 2001;16:601– 607.

Number 3

23 Taylor DC, Dalton JD, Seaber AV, Garrett WE. Viscoelastic properties of muscletendon units: the biomechanical effects of stretching. Am J Sports Med. 1990;18: 300 –309. 24 Ryan ED, Beck TW, Herda TJ, et al. The time course of musculotendinous stiffness responses following different durations of passive stretching. J Orthop Sports Phys Ther. 2008;38:632– 639. 25 Magnusson SP, Simonsen EB, Aagaard P, et al. Mechanical and physical responses to stretching with and without preisometric contraction in human skeletal muscle. Arch Phys Med Rehabil. 1996;77: 373–378. 26 Magnusson SP, Aagard P, Simonsen E, Bojsen-Møller F. A biomechanical evaluation of cyclic and static stretch in human skeletal muscle. Int J Sports Med. 1998;19: 310 –316. 27 De Deyne PG. Application of passive stretch and its implications for muscle fibers. Phys Ther. 2001;81:819 – 827. 28 Magnusson SP, Aagaard P, Nielson JJ. Passive energy return after repeated stretches of the hamstring muscle-tendon unit. Med Sci Sports Exerc. 2000;32:1160 –1164. 29 Feland JB, Myrer JW, Schulthies SS, et al. The effect of duration of stretching of the hamstring muscle group for increasing range of motion in people aged 65 years or older. Phys Ther. 2001;81:1110 –1107. 30 Draper DO, Castro JL, Feland B, et al. Shortwave diathermy and prolonged stretching increase hamstring flexibility more than prolonged stretching alone. J Orthop Sports Phys Ther. 2004;34: 13–20. 31 Wessling KC, DeVane DA, Hylton CR. Effects of static stretch versus static stretch and ultrasound combined on triceps surae muscle extensibility in healthy women. Phys Ther. 1987;67:674 – 679. 32 Sapega AA, Quedenfeld TC, Moyer RA, Butler RA. Biophysical factors in range-ofmotion exercise. Phys Sportsmed. 1981;9: 57– 65. 33 Zito M, Driver D, Parker C, Bohannon RL. Lasting effects of one bout of two 15second passive stretches on ankle dorsiflexion range of motion. J Orthop Sports Phys Ther. 1997;26:214 –221. 34 Hortoba´gyi T, Faludi J, Tihanyi J, Merkely B. Effects of intense “stretching”-flexibility training on the mechanical profile of the knee extensors and on the range of motion of the hip joint. Int J Sports Med. 1985;6:3173–3121. 35 Knight CA, Rutledge CR, Cox ME, et al. Effect of superficial heat, deep heat, and active exercise warm-up on the extensibility of the plantar flexors. Phys Ther. 2001; 81:1206 –1214. 36 Smith CA. The warm-up procedure: to stretch or not to stretch: a brief review. J Orthop Sports Phys Ther. 1994;19: 12–17. 37 Taylor BF, Waring CA, Brashear TA. The effects of therapeutic application of heat or cold followed by static stretch on hamstring muscle length. J Orthop Sports Phys Ther. 1995;21:283–286.

March 2010

Increasing Muscle Extensibility 38 Warren CG, Lehmann JF, Koblanski JN. Elongation of rat tail tendon: effect of load and temperature. Arch Phys Med Rehabil. 1971;52:465– 474. 39 Tabary JC, Tabary C, Tardieu C, et al. Physiological and structural changes in the cat’s soleus muscle due to immobilization at different lengths by plaster casts. J Physiol. 1972;224:231–244. 40 Goldspink G, Tabary C, Tabary JC, et al. Effect of denervation on the adaptation of sarcomere number and muscle extensibility to the functional length of the muscle. J Physiol. 1974;236:733–742. 41 Williams PE, Goldspink G. Changes in sarcomere length and physiological properties in immobilized muscle. J Anat. 1978; 127(pt 3):459 – 468. 42 Folpp H, Deall S, Harvey LA, Gwinn T. Can apparent increases in muscle extensibility with regular stretch be explained by changes in tolerance to stretch? Aust J Physiother. 2006;52:45–50. 43 Gajdosik RL. Effects of static stretching on the maximal length and resistance to passive stretch of short hamstring muscles. J Orthop Sports Phys Ther. 1991;14: 250 –255. 44 Gajdosik RL, Vander Linden DW, McNair PJ, et al. Effects of an eight-week stretching program on the passive-elastic properties and function of the calf muscles of older women. Clin Biomech (Bristol, Avon). 2005;20:973–983. 45 Reid DA, McNair PJ. Passive force, angle, and stiffness changes after stretching of hamstring muscles. Med Sci Sports Exerc. 2004;36:1944 –1918. 46 Spernoga SG, Uhl TL, Arnold BL, Gansneder BM. Duration of maintained hamstring flexibility after a one-time, modified hold-relax stretching protocol. J Athl Train. 2001;36:44 – 48. 47 Tanigawa MC. Comparison of the holdrelax procedure and passive mobilization on increasing muscle length. Phys Ther. 1972;52:725–735. 48 Alter MJ. Science of Flexibility. 3rd ed. Champaign, IL: Human Kinetics Publishers; 2004. 49 Bandy WD, Irion JM, Briggler M. The effect of static stretch and dynamic range of motion training on the flexibility of the hamstring muscles. J Orthop Sports Phys Ther. 1998;27:295–300. 50 Etnyre BR, Lee EJ. Comments on proprioceptive neuromuscular facilitation stretching techniques. Res Q Exerc Sport. 1987; 58:184 –188. 51 Depino GM, Webright WG, Arnold BL. Duration of maintained hamstring flexibility after cessation of an acute static stretching protocol. J Athl Train. 2000;35:56 –59. 52 Winters MV, Blake CG, Trost JS, et al. Passive versus active stretching of hip flexor muscles in subjects with limited hip extension: a randomized clinical trial. Phys Ther. 2004;84:800 – 807. 53 Nelson RT, Bandy WD. Eccentric Training and static stretching improve hamstring flexibility of high school males. J Athl Train. 2004;39:254 –258.

March 2010

54 Chalmers G. Re-examination of the possible role of Golgi tendon organ and muscle spindle reflexes in proprioceptive neuromuscular facilitation muscle stretching. Sports Biomech. 2004;3:159 –183. 55 Sharman MJ, Cresswell AG, Riek S. Proprioceptive neuromuscular facilitation stretching : mechanisms and clinical implications. Sports Med. 2006;36:929 –939. 56 Moore MA, Hutton RS. Electromyographic investigation of muscle stretching techniques. Med Sci Sports Exerc. 1980;12: 322–329. 57 McHugh MP, Kremenic IJ, Fox MB, Gleim GW. The role of mechanical and neural restraints to joint range of motion during passive stretch. Med Sci Sports Exerc. 1998;30:928 –932. 58 Magnusson SP, Simonsen EB, Aagaard P, et al. A mechanism for altered flexibility in human skeletal muscle. J Physiol. 1996; 497(pt 1):291–298. 59 Halbertsma JP, van Bolhuis AI, Go ¨ eken LN. Sport stretching: effect on passive muscle stiffness of short hamstrings. Arch Phys Med Rehabil. 1996;77:688 – 692. 60 Laessøe U, Voigt M. Modification of stretch tolerance in a stooping position. Scand J Med Sci Sports. 2004;14:239 –244. 61 Ben M, Harvey LA. Regular stretch does not increase muscle extensibility: a randomized controlled trial. Scand J Med Sci Sports. 2009 May 28. [Epub ahead of print] 62 Law RYW, Harvey LA, Nicholas MK, et al. Stretch exercises increase tolerance to stretch in patients with chronic musculoskeletal pain: a randomized controlled trial. Phys Ther. 2009;89:1016 –1026. 63 Bjo ¨ rklund M, Hamberg J, Crenshaw AG. Sensory adaptation after a 2-week stretching regimen of the rectus femoris muscle. Arch Phys Med Rehabil. 2001;82: 1245–1250. 64 Gajdosik RL, Allred JD, Gabbert HL, Sonsteng BA. A stretching program increases the dynamic passive length and passive resistive properties of the calf muscletendon unit of unconditioned younger women. Eur J Appl Physiol. 2007;99: 449 – 454. 65 Harvey LA, Byak AJ, Ostrovskaya M, et al. Randomised trial of the effects of four weeks of daily stretch on extensibility of hamstring muscles in people with spinal cord injuries. Aust J Physiother. 2003;49: 176 –181. 66 Harvey LA, Batty J, Crosbie J, et al. A randomized trial assessing the effects of 4 weeks of daily stretching on ankle mobility in patients with spinal cord injuries. Arch Phys Med Rehabil. 2000;81: 1340 –1347. 67 Williams P, Simpson H, Kyberd P, et al. Effect of rate of distraction on loss of range of joint movement, muscle stiffness, and intramuscular connective tissue content during surgical limb-lengthening: a study in the rabbit. Anat Rec. 1999;255:78 – 83. 68 Guissard N, Duchateau J. Effect of static stretch training on neural and mechanical properties of the human plantar-flexor muscles. Muscle Nerve. 2004;29:248 –255.

69 Bandy WD, Irion JM. The effect of time on static stretch on the flexibility of the hamstring muscles. Phys Ther. 1994;74:845– 580; discussion 850 – 852. 70 Bandy WD, Irion JM, Briggler M. The effect of time and frequency of static stretching on flexibility of the hamstring muscles. Phys Ther. 1997;77:1090 –1096. 71 McCreary EK, Provance PG, Rodgers MM, Romani WA. Muscles: Testing and Function, With Posture and Pain. 5th ed. Baltimore, MD: Lippincott Williams & Wilkins; 2005. 72 Reese NB, Bandy WD. Joint Range of Motion and Muscle Length Test. Philadelphia, PA: Saunders; 2002. 73 Sahrmann S. Diagnosis and Treatment of Movement Impairment Syndromes. St Louis, MO: Mosby; 2002. 74 Norkin CC, White DJ. Measurement of Joint Motion: A Guide to Goniometry. 3rd ed. Philadelphia, PA: FA Davis Co; 2003. 75 Bohannon RL. Commentary on “Hamstring Muscle Tightness.” Phys Ther. 1983; 63:1088 –1089. 76 Harvey LA, McQuade L, Hawthorne S, Byak A. Quantifying the magnitude of torque physiotherapists apply when stretching the hamstring muscles of people with spinal cord injury. Arch Phys Med Rehabil. 2003;84:1072–1075. 77 Magnusson SP, Simonsen EB, Aagaard P, et al. Contraction specific changes in passive torque in human skeletal muscle. Acta Physiol Scand. 1995;155:377–386. 78 Stephens J, Davidson J, Derosa J, et al. Lengthening the hamstring muscles without stretching using “awareness through movement”. Phys Ther. 2006; 86:1641–1650. 79 Halbertsma JP, Go ¨ eken LN, Hof AL, et al. Extensibility and stiffness of the hamstrings in patients with nonspecific low back pain. Arch Phys Med Rehabil. 2001; 82:232–238. 80 Salsich GB, Mueller MJ, Sahrmann SA. Passive ankle stiffness in subjects with diabetes and peripheral neuropathy versus an age-matched comparison group. Phys Ther. 2000;80:352–362. 81 Magnusson SP, Julsgaard C, Aagaard P, et al. Viscoelastic properties and flexibility of the human muscle-tendon unit in benign joint hypermobility syndrome. J Rheumatol. 2001;28:2720 –2725. 82 Magnusson SP, Simonsen EB, Aagaard P, et al. Determinants of musculoskeletal flexibility: viscoelastic properties, crosssectional area, EMG and stretch tolerance. Scand J Med Sci Sports. 1997;7:195–202. 83 Alonso J, McHugh MP, Mullaney MJ, Tyler TF. Effect of hamstring flexibility on isometric knee flexion angle-torque relationship. Scand J Med Sci Sports. 2009;19: 252–256. 84 Klinge K, Magnusson SP, Simonsen EB, et al. The effect of strength and flexibility training on skeletal muscle electromyographic activity, stiffness, and viscoelastic stress relaxation response. Am J Sports Med. 1997;25:710 –716.

Volume 90

Number 3

Physical Therapy f

449

CARE V Conference Series I. Kjeken, OT, PhD, is Senior Researcher, National Resource Centre for Rehabilitation in Rheumatology, Diakonhjemmet Hospital, Oslo, Norway. Address all correspondence to Dr Kjeken at: [email protected]. C. Ziegler, Bachelor in German and Danish, is Project Manager, Danish Rheumatism Association, Gentofte, Denmark. J. Skrolsvik, Bachelor in Marketing Management, is General Secretary, Norwegian Rheumatism Association, Oslo, Norway. J. Bagge, Master of Social Science, is Research Executive, Swedish Rheumatism Association, Stockholm, Sweden. G. Smedslund, PhD, is Senior Researcher, National Resource Centre for Rehabilitation in Rheumatology, Diakonhjemmet Hospital. A. Tøvik, Bachelor in Social Work, is Social Worker, National Resource Centre for Rehabilitation in Rheumatology, Diakonhjemmet Hospital.

How to Develop Patient-Centered Research: Some Perspectives Based on Surveys Among People With Rheumatic Diseases in Scandinavia Ingvild Kjeken, Connie Ziegler, Jack Skrolsvik, Jan Bagge, Geir Smedslund, Anne Tøvik, Hanne S. Dagfinrud, Ingemar F. Petersson, Kåre Birger Hagen Patient-centered research addresses the research agenda of patients and captures aspects of health and functioning that they consider important. Yet, those who live with a disease or condition have limited influence when it comes to setting the research agenda, and we know little about how they experience being participants in research studies. Furthermore, knowledge is limited concerning factors enhancing or hindering patients’ participation in trials and the format that people with rheumatic diseases and their families prefer for dissemination of the results from clinical research. This perspective article describes the research priorities of people with rheumatic diseases in Scandinavia, their experiences and attitudes concerning participation in research projects, and which format for research information they prefer. Based on results from 3 surveys organized by the Scandinavian Rheumatism Associations and on related research literature, the possible implications for future research also are discussed.

H.S. Dagfinrud, PT, PhD, is Senior Researcher, National Resource Centre for Rehabilitation in Rheumatology, Diakonhjemmet Hospital. I.F. Petersson, MD, PhD, is Associate Professor, Departments of Orthopaedics and Rheumatology, Lund University and Lund University Hospital, Lund, Sweden. KB Hagen, PhD, is Senior Researcher, National Resource Centre for Rehabilitation in Rheumatology, Diakonhjemmet Hospital. [Kjeken I, Ziegler C, Skrolsvik J, et al. How to develop patientcentered research: some perspectives based on surveys among people with rheumatic diseases in Scandinavia. Phys Ther. 2010;90: 450 – 460.] © 2010 American Physical Therapy Association

Post a Rapid Response or find The Bottom Line: www.ptjournal.org 450

f

Physical Therapy

Volume 90

Number 3

March 2010

Patient-Centered Research

T

here is a growing consensus among health care providers and researchers that patients should be involved as partners in clinical practice and research. On an individual level, patients participate in medical decisions and take increased responsibility for managing their disease.1,2 On a broader system level, patients’ feedback on the content and provision of health care is used to develop services that correspond to patients’ needs.3 Moreover, patients identify and assess important outcomes, and they evaluate the effect of interventions and programs.4,5 From the very beginning, research has depended upon the willingness of patients to volunteer as participants in studies. Recently, patients also have been involved as working partners in research projects.6 –10 Yet, we know little about how they experience being subjects in research studies, and those who live with a disease or condition still have limited influence when it comes to setting the research agenda.11,12 According to Bem et al,13 patientcentered research designs address the research agenda of patients and capture aspects of health and functioning that they consider to be important. However, knowledge concerning patients’ research priorities is still scarce. Researchers in 3 studies involving participants with cancer,14 chronic obstructive pulmonary disease (COPD),15 and ulcerative colitis16 all concluded that there is a need to broaden the research agenda in order to capture issues that are important to patients but poorly covered by current research. In all 3 studies, patients and researchers highly agreed that research on biological and treatment-related aspects such as causes, risk factors, early detection and prevention, and development of effective treatment is important. However, the top priority among people with cancer was reMarch 2010

search on the management of practical, social, and emotional issues, whereas the participants with COPD prioritized research on comorbidity, side effects of and interaction between medications, and psychosocial aspects.14,15 Participants with ulcerous colitis called for research on topics such as the effect of complementary therapies and how their colitis affected other family members.16 In a project involving people with low back pain, discussions with participants yielded 3 research themes: diagnosis and causality, understanding and proving the pain, and the impact of low back pain on quality of life.17 Within rheumatology, a study of people with knee osteoarthritis (OA) concluded that although the participants favored conservative treatments such as physical therapy and complementary medicine, the research was concentrated on studies on drugs and surgical treatment.18 Participants also wanted more research on psychological impact, education, and self-help strategies.

Research Priorities Among People With Arthritis in Scandinavia Within rheumatology, there have been several initiatives to strengthen the collaboration between patients and researchers. The CARE conferences were initiated in 2002 as an international forum where patient representatives and researchers meet and discuss issues related to care and research within rheumatology.19 Close collaboration with consumers has been central to the CARE conferences,20,21 and at the fifth CARE conference, consumers’ research agenda and experiences were specifically addressed. Preceding the conference, surveys addressing these issues had been administered on the Web site of the Danish Rheu-

matism Association and to a sample of members of the Norwegian Rheumatism Association, and members of the Swedish Rheumatism Association had been asked to evaluate a 10-year program within the association to facilitate transference of research results to people with arthritis and their families (for more detailed information regarding the 3 surveys, see Appendix 1). The results from the surveys indicate a general consensus across diagnoses regarding priorities for research. As in the other diagnostic groups,14 –16 etiology, prevention, and early diagnosis of disease were the most frequently chosen research topics among the Danish participants, and even if these topics were not listed as options in the Norwegian survey, they often were raised in the comments from the participants (Fig. 1). This finding probably reflects the understanding that increased knowledge of disease causes is essential to be able to prevent diseases and develop effective medications. In general, there seems to be consensus between consumers and researchers on the importance of biomedical research. Regarding research on treatments, the participants in the Danish and Norwegian surveys most frequently prioritized new medications and exercises as important issues (Figs. 1 and 2). However, there were interesting differences among diagnostic groups with regard to treatment options. In both surveys, participants

Available With This Article at ptjournal.apta.org • Audio Abstracts Podcast This article was published ahead of print on January 28, 2010, at ptjournal.apta.org.

Volume 90

Number 3

Physical Therapy f

451

Patient-Centered Research with rheumatoid arthritis (RA) most frequently chose research on medications, whereas participants with OA most often chose exercises. Over the last decades, new medical treatments have improved prognosis and quality of life in people with RA, and the priorities of the participants with this diagnosis may reflect a hope for even more progress in this area. Concerning exercises, there is an increasing body of evidence supporting the effectiveness of this intervention in the management of RA, ankylosing spondylitis (AS), and knee OA, although the research on hip and hand OA is still scarce. In general, there is currently little evidence for what types of exercises or doses are the most effective for which patients.22–24 Thus, there is a need for further research in this area, and the consensus among consumers regarding the importance of this research topic should be a strong encouragement for researchers within this field. Regarding research on symptoms and consequences, pain was the most frequently prioritized topic in all diagnostic groups in both the Danish and Norwegian surveys. Pain has been reported previously as the area of highest priority for improvement among people with RA25 and knee OA.18 Apparently, development of effective strategies aimed at pain relief and coping should have a high priority among health care professionals and researchers. In the Norwegian survey, fatigue and economy were the second and third most frequently listed topics for research on disease consequences. Accessibility to buildings and products was chosen as an important research topic by a substantial number of the participants in the Danish survey. In previous studies, people with RA have described fatigue as overwhelming, uncontrollable, and ignored by health care professionals,26 452

f

Physical Therapy

Volume 90

and they rated decreased fatigue as an extremely important outcome.27 A few studies have addressed societal strategies to reduce barriers such as limited personal economic resources, inaccessible buildings, and poorly designed devices and technology.28 Thus, the priorities of consumers timely reflect important research gaps and contradict a notion held by some researchers that consumers are incapable of adequate research prioritization due to a lack of relevant knowledge, a failure to look beyond their own individual problems, or an inability to consider long-term targets.15 However, to close this gap, there is a need for funding resources and processes where patients can significantly influence strategic decisions. It should be noted that study participants frequently chose research topics where the evidence is scarce or conflicting, such as nutrition and alternative therapy.29 People with arthritis are frequent users of such therapies.30 Still, the evidence for these interventions is not always reviewed when guidelines for treatment of different rheumatic diseases are developed,31 and when the evidence is reviewed, the lack of highquality studies on effect is striking.32–35 More studies, therefore, are needed to allow for informed choices regarding the use of these therapies. Whereas level of education had little influence on the Norwegian survey participants’ research priorities, those still working chose research on workability 3 times more frequently than those who were out of work (P⬍.001), and a greater proportion also prioritized research on exercise (P⬍.001). These results indicate that people with rheumatic diseases have a strong motivation to stay in the workforce and may be interpreted as a call for strategies to enable them to do so. Recent studies

Number 3

indicate that early and aggressive treatment, including use of antitumor necrosis factor agents, increase ability to work among people with shorter RA duration.36,37 However, current vocational rehabilitation programs have not proven effective to increase work capacity among patients with rheumatic diseases at risk for job loss.38 – 40 More attention, therefore, should be directed to development and testing of the effect of evidence-based vocational rehabilitation programs to enable people with rheumatic diseases to remain working or return to work. Even if there seems to be high agreement among consumers regarding the research agenda, the rationale behind research priorities probably varies across people and diagnoses. Studies using qualitative methods, therefore, are needed to gain more insight into the reasoning behind participants’ priorities.

Factors Influencing Participation in Clinical Studies Studies reveal that there are numerous reasons why people agree to enroll in studies. Some common reasons are an altruistic hope that the research will benefit future patients or a wish to gain personal profit by learning more about one’s own health, having extra health checkups, and getting access to better treatment.41– 44 A general trust in health care researchers or in specific researchers and the ability of the researcher to establish a friendly relationship with the patient are further underscored as important factors enhancing participation.42,45– 49 However, studies indicate that trust also may easily be broken due to misunderstandings or erroneous descriptions of the extent, time, and inconvenience of research-based tasks; feelings of being treated as an object rather than a study participant; or a

March 2010

Patient-Centered Research

Figure 1. Danish consumers’ priorities for research within rheumatology (%) (n⫽451). RA⫽rheumatoid arthritis, OA⫽osteoarthritis.

Figure 2. Priorities for research on treatments and consequences of rheumatic diseases among 516 members of the Norwegian Rheumatism Association (%). RA⫽rheumatoid arthritis, AS⫽ankylosing spondylitis, OA⫽osteoarthritis, FM⫽fibromyalgia, ADL⫽activities of daily living.

March 2010

Volume 90

Number 3

Physical Therapy f

453

Patient-Centered Research Table. Summary of Research Attitudes Among 319 Members of the Norwegian Rheumatism Association Who Would Like to Participate in Research Projects Yes

No

Do Not Know

Ability to contribute with useful information

80

10

10

Economy

45

41

14

Time consumption

63

28

9

Travel distance

77

16

7

Availability of personal computer/e-mail

56

36

8

Aspects of Research (% of Respondents) My participation in research projects depends on:

I may be interested in contributing by: Answering questionnaires

96

2

2

Participating in individual interviews

78

12

10

Participating in focus groups

70

16

14

Donating blood samples

93

2

5

Medication

52

29

19

Surgery

27

50

23

Psychological interventions

55

27

18

Alternative treatment

82

8

10

Physical activities and exercising

95

2

3

Diets

91

3

6

I am interested in participating in research projects on the following treatments:

Physical therapy

90

5

5

Assistive technology and environmental adaptations

75

13

12

Patient education

76

10

14

Designing and planning a project

47

29

24

Analyzing the results

52

25

23

Disseminating the results

49

26

25

As co-worker in a research project, I would like to participate in:

general experience that the research did not bring forward any important knowledge.42,46,47 Concerning barriers for entering trials, painful procedures, time consumption, travel distance and expenses, and problems taking time off from work are frequently listed as de-motivating factors.42 Furthermore, personal embarrassment caused by a need to reveal intimate information or to expose body parts may be an important barrier against entering the trial.42,45,46 Ethnicity also may play a role, as people from ethnic minorities may lack fluency in 454

f

Physical Therapy

Volume 90

the language used in the trial setting.42 Again, the results from the Scandinavian surveys, to a large degree, are in line with research within other medical areas. A total of 69% and 62% of the Danish and Norwegian survey respondents, respectively, were positive toward participating in research projects. The most frequently cited reason for participation among the Norwegian survey participants was a feeling that they could contribute important information, whereas time consumption and travel distance were the 2 most important

Number 3

practical barriers to participation (Table). Some suggestions for improving recruitment to studies were providing more information about the study aims and what participation would comprise and making study information available at drugstores and in the waiting areas of rheumatologists’ offices and rheumatology departments. However, the results of the Scandinavian surveys indicate that willingness to participate depends to a large degree on the aim of the study. Although more than 90% of the survey participants indicated they would participate in studies concerning physical activities and exercising, diets, and physical therapy, approximately half were willing to enroll in studies on medications or psychological interventions, and only one fourth were willing to enroll in studies on surgery. One reason for these findings could be that patients know that an operation often is followed by a period of pain and limited function and prefer interventions where they can contribute more actively and have more control. The high willingness to participate in research on exercises and diets supports this hypothesis. It also is interesting to note that even though research on new medications was one of the most frequently listed important research topics, only half of the Norwegian survey participants were willing to participate in pharmacological studies. In a recent study, findings indicated a general lack of trust in pharmaceutical companies among potential trial participants, which was based on an image that such institutions are driven by the pursuit of profit more than a pursuit of knowledge and the best interests of the participants.47 This finding underscores that trust is a dynamic concept, which is built and negotiated within the specific research setting.

March 2010

Patient-Centered Research Level of education and working status, to some degree, did influence people’s willingness to enroll in research projects, as 82% of the Norwegian survey participants with higher education levels were positive toward such participation compared with 59% of those with lower education levels (P⬍.001), and working participants were more likely than those who were out of work to report time consumption as a barrier to participation (73% versus 50%, P⫽.001). Both the Danish and Norwegian survey participants frequently commented that participation would depend on coverage of travel expenses and that good communication skills and a relaxed atmosphere would enhance their willingness to participate in future studies. Many respondents also emphasized that researchers and clinicians should understand that living with a disease is an individual experience, that the patient is the one with the most reliable knowledge about his or her situation and symptoms, and that patients develop a variety of strategies to cope with their disease. The comments indicate that participants want to be regarded as individuals with expertise rather than as study objects in a trial. It seems vital, therefore, that researchers establish workable relationships with participants by giving clear and concise information about the trial and what kind of performance participation requires. Furthermore, researchers should discuss mutual expectations and be willing to adapt time schedules and procedures to participants’ needs as much as possible.46 Great effort also should be taken to minimize the number of tests and questionnaires that patients need to complete and to optimize patient flow to avoid unnecessary waiting. From our own experience, we know that the possibility of performing tests and interviews on afternoons or on Saturdays March 2010

made it easier for working participants to attend a study, and a modest coverage of travel expenses was highly appreciated by the participants.50 To enhance participation from ethnic minorities, researchers should provide translated information sheets and use trained translators to inform potential participants. Concerning embarrassing procedures, care should be taken to minimize both the amount of time and parts of the body exposed and to ensure that the examination takes part in a shielded environment. Furthermore, some people may prefer to have the examination performed by a person of the same sex. In the Danish and Norwegian surveys, participants frequently commented that routinely providing feedback on study results would increase their willingness to participate in future studies, thereby underscoring the importance of research dissemination.

Dissemination of Study Results Study results may be disseminated to participants at 3 levels: (1) by reporting of individual results to each participant in a study; (2) by providing information of the aggregated results to study participants after study completion; and (3) by making results from new research available to those with the disease or condition, their families, and the general population. The International Ethical Guidelines for Biomedical Research Involving Human Subjects (Guideline 5: Obtaining Informed Consent: Essential Information for Prospective Research Subjects)51 states that before requesting an individual’s consent to participate in research, the investigator must provide information “that, after the completion of the study, subjects will be informed of the findings of the research in general, and individual subjects will be informed of any finding that relates to their particular health status.” In Appendix 1 in the guidelines, it is

further written that the following items, among others, should be included in the protocol: (1) plans and procedures, and the people responsible, for communicating to subjects information arising from the study (on harm or benefit, for example), or from other research on the same topic, that could affect subjects’ willingness to continue in the study; and (2) plans to inform subjects about the results of the study. Still, there is an ongoing discussion within the research community as to whether individual information should be disclosed to study participants.52,53 Results from a qualitative study, however, indicate that many participants are interested in receiving personal rather than, or in addition to, summary results.54 It also is argued that providing individual results will make the process of research more transparent and increase participants’ willingness to enroll in future studies.55 Some of the participants in the Danish and Norwegian surveys specifically commented that study participants should receive individual feedback on the results of their tests and assessments routinely. Such practice, however, raises both practical and ethical dilemmas. Producing valid, reliable, and comprehensible individual reports may be timeconsuming for researchers. Researchers should be careful not to force unwanted results on participants and should be aware that sharing the results has the potential to harm recipients by causing distress and anxiety.53,54 An invitation in the patient information sheet to request individual trial results, followed by a contact telephone number, may be a method to ensure disclosure to those participants who want such information. Obviously, researchers and ethical committees should consider carefully when and how participants should be informed of individual research results and include proce-

Volume 90

Number 3

Physical Therapy f

455

Patient-Centered Research dures addressing this issue in the study protocol. Studies indicate that offering a summary of aggregate research results to participants after study completion has numerous benefits, including having a positive impact on the future health of the participants. Providing a summary affirms appreciation to the participants of their important contribution and may improve the perception of research in the public sphere by allowing the participants to directly see the impact of their contribution to scientific knowledge.56,57 Findings from a study involving survivors of cancer or their parents also indicate that research participants want feedback even when the information is upsetting.58 Furthermore, only 5% of these study participants called one of the investigators listed in the result leaflet. The results indicate that additional work related to returning results may be overestimated. However, studies demonstrate that even if many researchers support the practice of providing research results to participants, very few actually do so.57,59 – 62 Reasons for not providing results to participants include anticipation of potential follow-up or contact difficulties and anticipation of participant difficulty understanding the results.60 An inconsistent practice also was demonstrated in the Danish survey, as only half of those who had participated in a study (42 out of 83 respondents) reported having received any feedback of the study results. However, both the Norwegian and Danish survey participants frequently commented that the final study results should be communicated to the participants, the patient population, and the public in general. The method preferred by participants to receive study results seems to be a posted letter in which the main results are summarized in lay language, to456

f

Physical Therapy

Volume 90

gether with contact telephone numbers.50,54,58 Additionally, planned, submitted, and published research articles could be listed in the leaflet, allowing those who want full-text versions of the articles to request them when they are published.58 It is further suggested that participants should have access to psychological support following dissemination of results and to medical follow-up if health risks are identified.60

and the patient population that is specific for the genre of research conducted.

Involving consumers in the development of patient information material has proven to be an effective method to ensure that the text is relevant, readable, and understandable to patients, without affecting participants’ anxiety.3 A recent article reporting communication of study results to participants in a trial for Huntington disease used modern technology in the dissemination process.63 The communication strategy included a media release that was posted on the Huntington Study Group’s public Web site and e-mailed to members of the Huntington disease community; a telephone call from the site staff to research participants providing results and next steps for their participation; and a joint telephone conference for the investigators, sponsor, and study participants to discuss study results.63

To reach a wider audience, relevant research could be summarized and discussed in open meetings announced in the local media. Publishing short reports in local and national patient magazines and on the Web sites of the patient associations may be an effective method for dissemination of study results. The power of these media to reach people with arthritis is confirmed by the results from the Swedish survey, as almost all respondents (93%) reported that they read about research in the Swedish Rheumatism Magazine and more than half of the respondents read the annual research supplement, whereas only 17% read about rheumatological research in journals and the popular press. The Swedish Rheumatism Magazine also was the most important source for research information, followed by the Internet and physicians. Nearly 40% of the Swedish consumers had participated at a lecture with popular presentations of rheumatological research, but even though more than one third had visited the Swedish Rheumatism Association’s Web page, only half of them had read research results published on the association’s Web site.

To improve future practice, existing guidelines for provision of study results should be followed. These guidelines include: (1) offering research results at the time of study enrollment; (2) disclosure following peer review but prior to public disclosure; (3) presenting participants with the harms and benefits of receiving the results; and (4) budgeting for the costs, including maintaining contact with research participants.56 All research projects also should include a communication plan for dissemination of the results to the participants and their families

Building on these experiences, the Norwegian Rheumatism Association recently published a report summarizing new research within rheumatology.64 The report is richly illustrated and consists of short summaries written by journalists based on interviews with researchers. Furthermore, a consumer and 2 researchers participated in the drafting committee, giving advice concerning content and language. The report was published on the association’s Web site, and printed copies have been available for free through the local branches of the association,

Number 3

March 2010

Patient-Centered Research as well as in the waiting areas of rheumatology departments and rehabilitation institutions. However, researchers should recognize that the majority of people with chronic diseases are not members of a patient association. Researchers, therefore, need to emphasize other channels for research dissemination, such as open conferences, newspapers, magazines, television, and radio. Suggestions for new strategies to inform current and coming generations of people with chronic conditions include use of movies, animation, and online technology.65,66 Rheumatic diseases are chronic conditions primarily affecting middleaged or elderly women. The fact that research regarding acute and lifethreatening diseases usually has a greater potential to reach public media indicates that researchers need to ally with professional information providers such as journalists and lobbyists to reach policy makers and the general population with new and important research results and information.

Participatory Research Involvement of patients as working partners in research projects has proven to be an effective method to ensure patient-centered research.6 –10 In close collaboration with researchers, partners can contribute throughout the whole research process, from identification and prioritizing of research topics to dissemination of results and decisions about which research is funded (for a more detailed description, see Appendix 2).5,6,26,67–70 To ensure active involvement of patients, researchers should follow published recommendations, principles, and indicators of successful consumer involvement.6,67,69 –72 Some central principles are time and climate for role negotiation; an approMarch 2010

priate budget for covering the costs of consumer involvement; a mutual respect for the differing skills, knowledge, and experience of consumers and researchers; appropriate training of consumers and researchers; involving consumers throughout the whole research process and acknowledging such involvement in the research reports; and making the findings available to consumers in formats and languages they can easily understand.71 There is an increasing body of experience concerning participatory research. Some of the challenges described are that patient research partners may feel uncertain concerning their ability to contribute or doubt if they will really have any significant influence in the research process. Furthermore, they may have a tendency to look up to experts and question their own experiential knowledge.70 Research partners have emphasized how important it is to create a safe and respectful environment, especially in the beginning, and that researchers take time to share a cup of coffee, listen to the concerns of the research partners, provide reassurance, adjust time schedules, and help them with difficult tasks.6,67,70 Care should be taken not to overburden the research partners by expecting them to read large number of documents or participate in long meetings with little time for breaks.70 Researchers also should be alert to improper reasons and inappropriate forms for collaboration, such as involving patient research partners for external or “aesthetic” reasons, or to enhance the adherence of patients in a trial.6,70 One indicator of successful consumer involvement in research is that the roles of consumers are agreed on between the researchers and consumers involved in research.6,69,72 The Scandinavian sur-

veys involved rheumatism associations, consumers, and researchers in 3 countries. One lesson learned from this experience is that the roles and responsibilities of the participants concerning coordination of the work among the 3 associations, as well as within each of the countries, should have been discussed and clarified from the beginning. Furthermore, reporting and dissemination of the results beyond the CARE conference should have been included in the project plan. These actions would have saved us much time spent sorting out these matters along the way, and probably would have increased the scientific rigor of the surveys. However, our experiences are also in line with those of other researchers and research partners, who report that collaboration generates substantial benefits for the research process.6,67,70 Most important is the notion that research that reflects the needs and experiences of patients is more likely to produce knowledge that can be used to improve clinical practice. Other examples of benefits are revealing prejudices, translating complex and abstract language, gaining access to networks and organizations, establishing trust among study participants, minimizing the risk of misinterpretations, and developing new perspectives and friendships thorough dialogues and mutual processes within the research team.6,70

Conclusions The Scandinavian surveys demonstrate that people with rheumatic diseases respond positively to participating in research and are highly competent at identifying important research issues. In general, participants had similar priorities for research. New pharmacological treatments, physical exercise, and management of pain and fatigue were rated as the most important issues for research on treatment and disease consequences. There were,

Volume 90

Number 3

Physical Therapy f

457

Patient-Centered Research however, some differences among diagnostic groups, as participants with RA gave highest priority to research on medications, whereas participants with OA gave highest priority to research on physical exercise. Research on surgery generally had low priority among participants with RA and OA. The results also indicate that willingness to participate varies, depending upon the aim of the study. Fewer than half of those enrolled in studies had received any information on the final results. To enhance future participation, researchers should develop a patientfriendly study design, including routines to ensure that participants receive feedback on the study results. Patients’ preferences for communication and dissemination of research also should be taken into account in future research projects. Involving consumers as research partners is an effective way to enhance patient-centered research. To ensure good practice, existing principles and recommendations for successful consumer involvement should be used to guide researchers who are new to participatory research. All authors provided concept/idea/project design and writing. Dr Kjeken, Ms Ziegler, Mr Bagge, Dr Smedslund, Ms Tøvik, and Dr Hagen provided data collection. Dr Kjeken, Ms Ziegler, Mr Skrolsvik, Mr Bagge, Ms Tøvik, Dr Dagfinrud, and Dr Petersson provided data analysis. Dr Kjeken, Ms Ziegler, Mr Bagge, Ms Tøvik, Dr Petersson, and Dr Hagen provided project management. Dr Kjeken and Dr Hagen provided fund procurement and institutional liaisons. Mr Skrolsvik and Mr Bagge provided patients. Dr Kjeken, Ms Ziegler, Mr Bagge, Dr Petersson, and Dr Hagen provided facilities/equipment. Dr Kjeken, Ms Ziegler, Mr Skrolsvik, and Mr Bagge provided clerical support. Dr Kjeken, Dr Smedslund, Dr Dagfinrud, Dr Petersson, and Dr Hagen provided consultation (including review of manuscript before submission).

23–25, 2008; Oslo, Norway. The results also were presented orally at the European League Against Rheumatism (EULAR) Congress; June 10 –13, 2009; Copenhagen, Denmark. This article was received December 1, 2008, and was accepted November 12, 2009. DOI: 10.2522/ptj.20080381

References 1 Ward MM. Patient-centered care and health outcomes. Curr Opin Rheumatol. 2004;16:89 –90. 2 Hewlett S, Kirwan J, Pollock J, et al. Patient initiated outpatient follow-up in rheumatoid arthritis: six-year randomised controlled trial. BMJ. 2005;330:171. 3 Nilsen ES, Myrhaug HT, Johansen M, et al. Methods of consumer involvement in developing healthcare policy and research, clinical practice guidelines and patient information material. Cochrane Database Syst Rev. 2006;3:CD004563. 4 Kvien TK, Heiberg T. The patient perspective in outcome assessments: perceptions or something more? J Rheumatol. 2003; 30:873– 876. 5 Kirwan JR, Hewlett SE, Heiberg T, et al. Incorporating the Patient perspective into outcome assessment in rheumatoid arthritis: progress at OMERACT 7. J Rheumatol. 2005;32:2250 –2256. 6 Hewlett S, Wit M, Richards P, et al. Patients and professionals as research partners: challenges, practicalities, and benefits. Arthritis Rheum. 2006;55:676 – 680. 7 Boote J, Telford R, Cooper C. Consumer involvement in health research: a review and research agenda. Health Policy. 2002; 61:213–236. 8 Hanley B, Truesdale A, King A, et al. Involving consumers in designing, conducting, and interpreting randomised controlled trials: questionnaire survey. BMJ. 2001;322:519 –523. 9 Fudge N, Wolfe CD, McKevitt C. Involving older people in health research. Age Ageing. 2007;36:492–500. 10 Hubbard G, Kidd L, Donaghy E, et al. A review of literature about involving people affected by cancer in research, policy and planning and practice. Patient Educ Couns. 2007;65:21–33. 11 Li LC, Koehn C, Lehman AJ. People with arthritis and their families in rehabilitation, care and research. Curr Opin Rheumatol. 2007;19:174 –178. 12 Tallon D, Chard J, Dieppe P. Relation between agendas of the research community and the research consumer. Lancet. 2000; 355:2037–2040. 13 Bem C, Lee C, Dawson R, Watkinson J. Is clinical otolaryngology publishing patientcentred research? Clin Otolaryngol Allied Sci. 2004;29:84 –93.

14 Corner J, Wright D, Hopkinson J, et al. The research priorities of patients attending UK cancer treatment centres: findings from a modified nominal group study. Br J Cancer. 2007;96:875– 881. 15 Caron-Flinterman JF, Broerse JE, Teerling J, Bunders JF. Patients’ priorities concerning health research: the case of asthma and COPD research in the Netherlands. Health Expect. 2005;8:253–263. 16 Welfare MR, Colligan J, Molyneux S, et al. The identification of topics for research that are important to people with ulcerative colitis. Eur J Gastroenterol Hepatol. 2006;18:939 –944. 17 Ong BN, Hooper H. Involving users in low back pain research. Health Expect. 2003; 6:332–341. 18 Tallon D, Chard J, Dieppe P. Exploring the priorities of patients with osteoarthritis of the knee. Arthritis Care Res. 2000;13: 312–319. 19 Vliet Vlieland TP. CARE: International Conference on Multidisciplinary Care in Rheumatoid Arthritis. International Journal of Advances in Rheumatology. 2002; 1:34 –36. 20 Petersson IF, Bremander AB, Klareskog L, Stenstrom C. Who cares about team care? Ann Rheum Dis. 2005;64:644. 21 Li LC, Bombardier C. Setting priorities in arthritis care: Care III Conference. J Rheumatol. 2006;33:1891–1894. 22 Jamtvedt G, Dahm KT, Christie A, et al. Physical therapy interventions for patients with osteoarthritis of the knee: an overview of systematic reviews. Phys Ther. 2008;88:123–136. 23 Dagfinrud H, Kvien TK, Hagen KB. Physiotherapy interventions for ankylosing spondylitis. Cochrane Database Syst Rev. 2008;1:CD002822. 24 Gaudin P, Leguen-Guegan S, Allenet B, et al. Is dynamic exercise beneficial in patients with rheumatoid arthritis? Joint Bone Spine. 2008;75:11–17. 25 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. 26 Hewlett S, Cockshott Z, Byron M, et al. Patients’ perceptions of fatigue in rheumatoid arthritis: overwhelming, uncontrollable, ignored. Arthritis Rheum. 2005;53: 697–702. 27 Kirwan J, Heiberg T, Hewlett S, et al. Outcomes from the Patient Perspective Workshop at OMERACT 6. J Rheumatol. 2003; 30:868 – 872. 28 Tuntland H, Kjeken I, Nordheim L, et al. Assistive technology for rheumatoid arthritis. Cochrane Database Syst Rev. 2009;4: CD006729. 29 Gossec L, Pavy S, Pham T, et al. Nonpharmacological treatments in early rheumatoid arthritis: clinical practice guidelines based on published evidence and expert opinion. Joint Bone Spine. 2006;73: 396 – 402.

Part of the manuscript was presented orally at the CARE V International Conference, April

458

f

Physical Therapy

Volume 90

Number 3

March 2010

Patient-Centered Research 30 Vitetta L, Cicuttini F, Sali A. Alternative therapies for musculoskeletal conditions. Best Pract Res Clin Rheumatol. 2008;22: 499 –522. 31 Sidiropoulos PI, Hatemi G, Song IH, et al. Evidence-based recommendations for the management of ankylosing spondylitis: systematic literature search of the 3E Initiative in Rheumatology involving a broad panel of experts and practising rheumatologists. Rheumatology (Oxford). 2008; 47:355–361. 32 Carville SF, Arendt-Nielsen S, Bliddal H, et al. EULAR evidence-based recommendations for the management of fibromyalgia syndrome. Ann Rheum Dis. 2008;67: 536 –541. 33 Zhang W, Doherty M, Leeb BF, et al. EULAR evidence-based recommendations for the management of hand osteoarthritis: report of a task force of the EULAR Standing Committee for International Clinical Studies Including Therapeutics (ESCISIT). Ann Rheum Dis. 2007;66: 377–388. 34 Zhang W, Doherty M, Arden N, et al. 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. 35 Jordan KM, Arden NK, Doherty M, et al. EULAR recommendations 2003: an evidence-based approach to the management of knee osteoarthritis: report of a task force of the Standing Committee for International Clinical Studies Including Therapeutic Trials (ESCISIT). Ann Rheum Dis 2003;62:1145–1155. 36 Allaire S, Wolfe F, Niu J, et al. Evaluation of the effect of anti-tumor necrosis factor agent use on rheumatoid arthritis work disability: the jury is still out. Arthritis Rheum. 2008;59:1082–1089. 37 Puolakka K, Kautiainen H, Mottonen T, et al. Early suppression of disease activity is essential for maintenance of work capacity in patients with recent-onset rheumatoid arthritis: five-year experience from the FIN-RACo trial. Arthritis Rheum. 2005;52:36 – 41. 38 Puolakka K, Kautiainen H, Mottonen T, et al. Cost of Finnish statutory inpatient rehabilitation and its impact on functional and work capacity of patients with early rheumatoid arthritis: experience from the FIN-RACo trial. Scand J Rheumatol. 2007; 36:270 –257. 39 de Buck PD, Le CS, Van Den Hout WB, et al. Randomized comparison of a multidisciplinary job-retention vocational rehabilitation program with usual outpatient care in patients with chronic arthritis at risk for job loss. Arthritis Rheum. 2005; 53:682– 690. 40 de Buck PD, Schoones JW, Allaire SH, Vliet Vlieland TP. Vocational rehabilitation in patients with chronic rheumatic diseases: a systematic literature review. Semin Arthritis. Rheum 2002;32:196 –203. 41 Verheggen FW, Nieman FH, Reerink E, Kok GJ. Patient satisfaction with clinical trial participation. Int J Qual Health Care. 1998;10:319 –330.

March 2010

42 Hussain-Gambles M. South Asian patients’ views and experiences of clinical trial participation. Fam Pract. 2004;21:636 – 642. 43 Heaven B, Murtagh M, Rapley T, et al. Patients or research subjects? A qualitative study of participation in a randomised controlled trial of a complex intervention. Patient Educ Couns. 2006;62:260 –270. 44 Verheggen FW, Nieman F, Jonkers R. Determinants of patient participation in clinical studies requiring informed consent: why patients enter a clinical trial. Patient Educ Couns. 1998;35:111–125. 45 Morris N, Balmer B. Are you sitting comfortably? Perspectives of the researchers and the researched on “being comfortable.” Account Res. 2006;13:111–133. 46 Morris N, Balmer B. Volunteer human subjects’ understandings of their participation in a biomedical research experiment. Soc Sci Med. 2006;62:998 –1008. 47 McDonald M, Townsend A, Cox SM, et al. Trust in health research relationships: accounts of human subjects. J Empir Res Hum Res Ethics. 2008;3:35– 47. 48 Dixon-Woods M, Tarrant C. Why do people cooperate with medical research? Findings from three studies. Soc Sci Med. 2009;68:2215–2222. 49 Kass NE, Sugarman J, Faden R, SchochSpana M. Trust: the fragile foundation of contemporary biomedical research. Hastings Cent Rep. 1996;26:25–29. 50 Kjeken I. Participation, Involvement and Functional Assessment in Rheumatology Care [thesis]. Oslo, Norway: University of Oslo; 2006. 51 International Ethical Guidelines for Biomedical Research Involving Human Subjects. Council for International Organizations of Medical Sciences. Available at: www.cioms.ch / frame_guidelines_nov_ 2002.htm. Accessed September 17, 2009. 52 Banks TM. Misusing informed consent: a critique of limitations on research subjects’ access to genetic research results. Sask Law Rev. 2000;63:539 –580. 53 Shalowitz DI, Miller FG. The search for clarity in communicating research results to study participants. J Med Ethics. 2008; 34:e17. 54 Dixon-Woods M, Jackson C, Windridge KC, Kenyon S. Receiving a summary of the results of a trial: qualitative study of participants’ views. BMJ. 2006;332:206 –210. 55 Shalowitz DI, Miller FG. Disclosing individual results of clinical research: implications of respect for participants. JAMA. 2005;294:737–740. 56 Fernandez CV, Kodish E, Weijer C. Informing study participants of research results: an ethical imperative. IRB. 2003;25:12–19. 57 Partridge AH, Winer EP. Informing clinical trial participants about study results. JAMA. 2002;288:363–365. 58 Schulz CJ, Riddle MP, Valdimirsdottir HB, et al. Impact on survivors of retinoblastoma when informed of study results on risk of second cancers. Med Pediatr Oncol. 2003;41:36 – 43.

59 Macneil SD, Fernandez CV. Informing research participants of research results: analysis of Canadian university based research ethics board policies. J Med Ethics. 2006;32:49 –54. 60 Rigby H, Fernandez CV. Providing research results to study participants: support versus practice of researchers presenting at the American Society of Hematology annual meeting. Blood. 2005; 106:1199 –1202. 61 Patridge AH, Hackett N, Blood E, et al. Oncology physician and nurse practices and attitudes regarding offering clinical trial results to study participants. J Natl Cancer Inst. 2004;96:629 – 632. 62 Fernandez CV, Kodish E, Taweel S, et al. Disclosure of the right of research participants to receive research results: an analysis of consent forms in the Children’s Oncology Group. Cancer. 2003;97: 2904 –2909. 63 Dorsey ER, Beck CA, Adams M, et al. Communicating clinical trial results to research participants. Arch Neurol. 2008;65: 1590 –1595. 64 Revmarapporten 2008: Om revmatiske leddsykdommer (The Rheuma Report 2008). Norwegian Rheumatism Association; 2009. Available at: www.revmatiker. no/Plager/Revmarapporten. Accessed September 17, 2009. 65 Fullilove MT, Green LL, HernandezCordero LJ, Fullilove RE. Obvious and notso-obvious strategies to disseminate research. Health Promot Pract. 2006;7: 306 –311. 66 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. 67 Oliver S, Clarke-Jones L, Rees R, et al. Involving consumers in research and development agenda setting for the NHS: developing an evidence-based approach. Health Technol Assess. 2004;8:1–148, III– IV. 68 O’Donnell M, Entwistle V. Consumer involvement in decisions about what healthrelated research is funded. Health Policy. 2004;70:281–290. 69 Royle J, Steel R, Hanley B, Bradburn J. Getting Involved in Research: A Guide for Consumers. London, United Kingom: Consumers in NHS Research Support Unit; 2001. 70 Abma TA, Nierse CJ, Widdershoven GAM. Patients as partners in responsive research: methodological notions for collaborations in mixed research teams. Qual Health Res. 2009;19:401– 415. 71 Boote J, Barber R, Cooper C. Principles and indicators of successful consumer involvement in NHS research: results of a Delphi study and subgroup analysis. Health Policy. 2006;75:280 –297. 72 Telford R, Boote JD, Cooper CL. What does it mean to involve consumers successfully in NHS research? A consensus study. Health Expect. 2004;7:209 –220.

Volume 90

Number 3

Physical Therapy f

459

Patient-Centered Research Appendix 1. Outline of Surveys Administered by the 3 Scandinavian Rheumatism Associations All 3 questionnaires contained questions regarding participants’ diagnosis, sex, and disease duration. The Norwegian questionnaire also included 2 questions regarding education and work status. In the Danish and Norwegian samples, statistical comparisons between groups were made concerning diagnoses (including diagnoses representing ⱖ10% of the total sample), using chi-square analyses. Comparisons concerning level of education and work status were conducted in the Norwegian sample. Denmark Three employees of the Danish Rheumatism Association developed the questionnaire. Topics were listed according to their experience of what members thought were important areas to research, based on questions raised by members in telephone calls and meetings. In the first part of the questionnaire, the participants were asked to indicate the 3 out of 17 listed topics they prioritized as the most important areas for future research, followed by an open question where the participants could suggest other topics. The second part of the questionnaire concerned participants’ experiences with and willingness to participate in research and contained an open question asking for suggestions on how to improve patients’ involvement in research. The questionnaire was available at the Danish Rheumatism Association’s Web site over a period of 8 months in 2007. A total of 451 people with a mean age of 61 years answered the questionnaire, of which 83% were women. Norway The Norwegian questionnaire was a revised version of the Danish questionnaire and was developed by a group of 2 researchers, 1 clinician, and 1 consumer with rheumatoid arthritis. In the questionnaire, participants first were asked to check the 3 out of 9 listed disease consequences they thought were the most important areas for future research, and then they were asked to indicate the 3 out of 9 treatments they would prioritize as the most important research areas. Some questions were added concerning members’ willingness to participate in research projects, and each section in the questionnaire had open-ended questions, allowing the participants to comment on the different issues. The questionnaire was mailed to 1,000 randomly selected members of the Norwegian Rheumatism Association in the autumn of 2007. A total of 516 people with a mean age of 56.9 years returned the questionnaire, of which 75% were women. Sweden The Swedish questionnaire was developed by 2 employees of the Swedish Rheumatism Association with the help of a researcher and contained questions concerning respondents’ use of the association’s Web site, magazine, and yearly supplement with summaries of relevant research and their participation in open meetings for members and families. In a last question, respondents were asked to list their 3 favorite sources for updating their knowledge about relevant research. The questionnaire was mailed to 300 randomly selected members of the Swedish Rheumatism Association in the spring of 2008, of which 158 completed the survey. The participants had a median age of 61 years, and 83% were women.

Appendix 2. Ways Patient Research Partners Can Contribute in Research Projects3,5,6,66 – 69 Patient research partners can contribute in research by: • identifying and prioritizing research topics • reviewing research protocols • developing specific research questions • identifying important outcomes • developing interview guides • choosing feasible tests and instruments • discussing inclusion and exclusion criteria • deciding appropriate time intervals from baseline to control • giving input concerning factors that can facilitate or hamper participation • writing or commenting on patient information sheets • serving as “pilot” participants • recruiting participants • conducting interviews with study participants • preparing or leading a focus group • asking people to complete questionnaires or surveys • monitoring research • analyzing and interpreting results • disseminating results • evaluating research • participating in decisions about research funding • providing training for professionals about involving consumers in research

460

f

Physical Therapy

Volume 90

Number 3

March 2010

Letters to the Editor On “Exposure to low amounts of ultrasound energy…” Alexander LD, Gilman DRD, Brown DR, et al. Phys Ther. 2010;90:14–25. I must protest the title of this systematic review.1 A review of this systematic review reveals that there is nothing in the review that indicates whether or not low amounts of ultrasound energy improve soft tissue shoulder pathology. In fact, the authors clearly state, “Our results suggest that the effectiveness of ultrasound on soft tissue pathologies has not yet been evaluated using optimal treatment parameters, and, therefore, it is premature to conclude through systematic review of existing literature that this treatment dose ‘is not effective.’” Given that, the reader has to wonder how and why this title came about. Why didn’t the title reflect the finding that there were some beneficial effects from higher exposure to ultrasound energy? Why didn’t the title reflect that the current research was not sufficient to tell clinicians anything worthwhile regarding the use of ultrasound? Truthfully, the article told us far more about the inefficacy of the research than it did about the inefficacy of ultrasound. Why was this article even published, given what little clinical benefit it offers? Why were most of the studies the authors reviewed ever published? They may tell future researchers other ways not to do research, but they tell clinicians very little, if anything, about how to treat patients who have shoulder pathology. I am concerned that the use of this title is irresponsible. As third-party payers review the literature, not having the requisite

March 2010

Letter 3.10.indd 461

knowledge and ability to truly analyze the content, they will rely on a title like that of this article to deny coverage. We cannot afford to give these people such ammunition. My biggest concern, though, is the negative effect that this title and the article in general may have on the evidence-based practice movement. We hear many complaints from researchers and academicians bemoaning the fact that so many clinicians do not embrace evidence-based practice. I suggest that it is, at least in part, because so many studies and articles like this one are published. Because they lack credibility and defy some of the conventional wisdom that is gained in introspective clinical practice, they shed a bad light on the evidence-based practice movement. Clinicians subsequently become suspicious of all of the “evidence” and either disregard it or stop reviewing it altogether. That is a terrible injustice, as it sets us backward rather than moves us forward. Understand that I am not writing in support of ultrasound. I am writing in support of responsible evidence-based practice. The jury is still out as to the efficacy of ultrasound. Clinically, use of ultrasound seems to help make mobilization easier. Is that a good clinical observation, or merely a “We’ve always done it that way, so don’t tell us it doesn’t work” lament? So far, the research has not been able to answer that question, so it is still up to clinicians to try to decide whether it works for them or not. The fact that so many clinicians still choose to use ultrasound for pain modulation and for mobilization preparation—whereas they

abandoned many other modalities through the years—probably tells us a lot. It might be the best available evidence we have right now. Maybe someday the research will prove them right, and maybe it will prove them wrong. In the meantime, we have to be very careful not to misinterpret or misrepresent what evidence we do have. We have to use our research just as responsibly as we are supposed to use all of our other practice tools. Philip P. Tygiel P.P. Tygiel, PT, is Physical Therapist, Tygiel Physical Therapy, 6606 E Carondelet, Tucson, AZ 85710. Address all correspondence to Mr Tygiel at: [email protected]. This letter was posted as a Rapid Response on January 7, 2010, at ptjournal.apta.org.

Reference 1 Alexander LD, Gilman DRD, Brown DR, et al. Exposure to low amounts of ultrasound energy does not improve soft tissue shoulder pathology: a systematic review. Phys Ther. 2010;90:14–25. [DOI: 10.2522/ptj.2010.90.3.461.1]

Author Response I thank the responder for his comments1 on our article.2 I believe this commentary helps to highlight a key point made in the article and perhaps challenges the editorial process associated with PTJ. I regret that Tygiel feels this article will have little clinical benefit to others. It represents a culmination of a tremendous amount of work, and obviously we believe that it makes an important contribution to the physical therapy profession. It supports a change in practice, which is to reduce the use of short treatments of low-intensity pulsed ultrasound that deliver extremely low amounts of ultrasound energy. In addition to changing clini-

Volume 90 Number 3 Physical Therapy ■ 461

2/12/10 4:51 PM

Letters to the Editor On “Exposure to low amounts of ultrasound energy…” Alexander LD, Gilman DRD, Brown DR, et al. Phys Ther. 2010;90:14–25. I must protest the title of this systematic review.1 A review of this systematic review reveals that there is nothing in the review that indicates whether or not low amounts of ultrasound energy improve soft tissue shoulder pathology. In fact, the authors clearly state, “Our results suggest that the effectiveness of ultrasound on soft tissue pathologies has not yet been evaluated using optimal treatment parameters, and, therefore, it is premature to conclude through systematic review of existing literature that this treatment dose ‘is not effective.’” Given that, the reader has to wonder how and why this title came about. Why didn’t the title reflect the finding that there were some beneficial effects from higher exposure to ultrasound energy? Why didn’t the title reflect that the current research was not sufficient to tell clinicians anything worthwhile regarding the use of ultrasound? Truthfully, the article told us far more about the inefficacy of the research than it did about the inefficacy of ultrasound. Why was this article even published, given what little clinical benefit it offers? Why were most of the studies the authors reviewed ever published? They may tell future researchers other ways not to do research, but they tell clinicians very little, if anything, about how to treat patients who have shoulder pathology. I am concerned that the use of this title is irresponsible. As third-party payers review the literature, not having the requisite

March 2010

Letter 3.10.indd 461

knowledge and ability to truly analyze the content, they will rely on a title like that of this article to deny coverage. We cannot afford to give these people such ammunition. My biggest concern, though, is the negative effect that this title and the article in general may have on the evidence-based practice movement. We hear many complaints from researchers and academicians bemoaning the fact that so many clinicians do not embrace evidence-based practice. I suggest that it is, at least in part, because so many studies and articles like this one are published. Because they lack credibility and defy some of the conventional wisdom that is gained in introspective clinical practice, they shed a bad light on the evidence-based practice movement. Clinicians subsequently become suspicious of all of the “evidence” and either disregard it or stop reviewing it altogether. That is a terrible injustice, as it sets us backward rather than moves us forward. Understand that I am not writing in support of ultrasound. I am writing in support of responsible evidence-based practice. The jury is still out as to the efficacy of ultrasound. Clinically, use of ultrasound seems to help make mobilization easier. Is that a good clinical observation, or merely a “We’ve always done it that way, so don’t tell us it doesn’t work” lament? So far, the research has not been able to answer that question, so it is still up to clinicians to try to decide whether it works for them or not. The fact that so many clinicians still choose to use ultrasound for pain modulation and for mobilization preparation—whereas they

abandoned many other modalities through the years—probably tells us a lot. It might be the best available evidence we have right now. Maybe someday the research will prove them right, and maybe it will prove them wrong. In the meantime, we have to be very careful not to misinterpret or misrepresent what evidence we do have. We have to use our research just as responsibly as we are supposed to use all of our other practice tools. Philip P. Tygiel P.P. Tygiel, PT, is Physical Therapist, Tygiel Physical Therapy, 6606 E Carondelet, Tucson, AZ 85710. Address all correspondence to Mr Tygiel at: [email protected]. This letter was posted as a Rapid Response on January 7, 2010, at ptjournal.apta.org.

Reference 1 Alexander LD, Gilman DRD, Brown DR, et al. Exposure to low amounts of ultrasound energy does not improve soft tissue shoulder pathology: a systematic review. Phys Ther. 2010;90:14–25. [DOI: 10.2522/ptj.2010.90.3.461.1]

Author Response I thank the responder for his comments1 on our article.2 I believe this commentary helps to highlight a key point made in the article and perhaps challenges the editorial process associated with PTJ. I regret that Tygiel feels this article will have little clinical benefit to others. It represents a culmination of a tremendous amount of work, and obviously we believe that it makes an important contribution to the physical therapy profession. It supports a change in practice, which is to reduce the use of short treatments of low-intensity pulsed ultrasound that deliver extremely low amounts of ultrasound energy. In addition to changing clini-

Volume 90 Number 3 Physical Therapy ■ 461

2/12/10 4:51 PM

Letters to the Editor cal practice, we must change how clinical trials are conducted and what is considered when conducting systematic reviews. If readers read the full text of our article, they will see that we agree with Tygiel’s point that current research is not helping us determine whether ultrasound will aid the treatment of people with shoulder disorders. In particular, the amount of energy delivered in the treatment arm of clinical trials must be “adequate” if we are to evaluate the effectiveness of any physical therapy intervention. With the popularity of systematic reviews and meta-analyses used to evaluate clinical treatments, the appropriateness of treatment parameters (not just methodological quality) must be considered when drawing conclusions about the treatment outcomes. Although the title may not have been best, I do believe the conclusions in the abstract and the results and discussion sections of the article accurately reflect the current literature. Unfortunately, there is not enough evidence to support

a suggestion that higher doses of ultrasound are effective in treating people with shoulder disorders. My hope is that third-party payers will venture beyond the title and at least read the conclusion section of the abstract. The title used in this article was not one submitted by the authors of this article and was not considered by the manuscript reviewers. It was in the final stages of editorial production that the title was changed to the current version so that it would be consistent with PTJ’s move to promote the use of active, more informative article titles. In fact, a revised title suggested by the editor in chief was “There is not strong evidence to support the use of therapeutic ultrasound for soft tissue shoulder pathology: a systematic review.” I suspect this title would have fueled to an even greater extent the valid concerns raised about the negative impact on the evidence-based practice movement. Perhaps this discussion will stimulate a review of this editorial practice of PTJ.

Pamela E. Houghton P.E. Houghton, HBSc, BScPT, PhD, is Associate Professor, School of Physical Therapy, University of Western Ontario, Room 1458, Elborn College, London, Ontario, Canada N6G 1H1. Address all correspondence to Dr Houghton at: [email protected]. This letter was posted as a Rapid Response on January 11, 2010, at ptjournal.apta.org.

References 1 Tygiel PP. Letter to the editor on “Exposure to low amounts of ultrasound energy does not improve soft tissue shoulder pathology: a systematic review. Phys Ther. 2010;90:461. 2 Alexander LD, Gilman DRD, Brown DR, et al. Exposure to low amounts of ultrasound energy does not improve soft tissue shoulder pathology: a systematic review. Phys Ther. 2010;90:14–25. [DOI: 10.2522/ptj.2010.90.3.461.2]

Correction Steffen T, et al. “Test-retest reliability and minimal detectable change on balance and ambulation tests...” Phys Ther. 2008;88:733–746. In the article titled “Test-Retest Reliability and Minimal Detectable Change on Balance and Ambulation Tests, the 36-Item Short-Form Health Survey, and the Unified Parkinson Disease Rating Scale in People With Parkinsonism” by Teresa Steffen and Megan Seney in the June 2008 issue of PTJ, the authors report an error in the literature review on page 737 (column 1, second paragraph). The sentences in that paragraph should have read (changes in bold): “The MDC95 values for comfortable and fast gait speeds were 0.20 and 0.30 m/s, respectively, for people with osteoporosis60 and 0.12 and 0.22 m/s for people with hip fracture.21 Overall, the MDC95 values for both comfortable and fast gait speeds appeared to be about 0.25 m/s or less for populations tested to date.” The latter sentence originally included the phrase “except for patients with hip fracture, with an MDC95 value of approximately 0.50 m/s,” which was an error. DOI: 10.2522/ptj.20070214.cx

462 ■ Physical Therapy Volume 90 Number 3

Letter 3.10.indd 462

March 2010

2/17/10 2:57 PM

Letters to the Editor cal practice, we must change how clinical trials are conducted and what is considered when conducting systematic reviews. If readers read the full text of our article, they will see that we agree with Tygiel’s point that current research is not helping us determine whether ultrasound will aid the treatment of people with shoulder disorders. In particular, the amount of energy delivered in the treatment arm of clinical trials must be “adequate” if we are to evaluate the effectiveness of any physical therapy intervention. With the popularity of systematic reviews and meta-analyses used to evaluate clinical treatments, the appropriateness of treatment parameters (not just methodological quality) must be considered when drawing conclusions about the treatment outcomes. Although the title may not have been best, I do believe the conclusions in the abstract and the results and discussion sections of the article accurately reflect the current literature. Unfortunately, there is not enough evidence to support

a suggestion that higher doses of ultrasound are effective in treating people with shoulder disorders. My hope is that third-party payers will venture beyond the title and at least read the conclusion section of the abstract. The title used in this article was not one submitted by the authors of this article and was not considered by the manuscript reviewers. It was in the final stages of editorial production that the title was changed to the current version so that it would be consistent with PTJ’s move to promote the use of active, more informative article titles. In fact, a revised title suggested by the editor in chief was “There is not strong evidence to support the use of therapeutic ultrasound for soft tissue shoulder pathology: a systematic review.” I suspect this title would have fueled to an even greater extent the valid concerns raised about the negative impact on the evidence-based practice movement. Perhaps this discussion will stimulate a review of this editorial practice of PTJ.

Pamela E. Houghton P.E. Houghton, HBSc, BScPT, PhD, is Associate Professor, School of Physical Therapy, University of Western Ontario, Room 1458, Elborn College, London, Ontario, Canada N6G 1H1. Address all correspondence to Dr Houghton at: [email protected]. This letter was posted as a Rapid Response on January 11, 2010, at ptjournal.apta.org.

References 1 Tygiel PP. Letter to the editor on “Exposure to low amounts of ultrasound energy does not improve soft tissue shoulder pathology: a systematic review. Phys Ther. 2010;90:461. 2 Alexander LD, Gilman DRD, Brown DR, et al. Exposure to low amounts of ultrasound energy does not improve soft tissue shoulder pathology: a systematic review. Phys Ther. 2010;90:14–25. [DOI: 10.2522/ptj.2010.90.3.461.2]

Correction Steffen T, et al. “Test-retest reliability and minimal detectable change on balance and ambulation tests...” Phys Ther. 2008;88:733–746. In the article titled “Test-Retest Reliability and Minimal Detectable Change on Balance and Ambulation Tests, the 36-Item Short-Form Health Survey, and the Unified Parkinson Disease Rating Scale in People With Parkinsonism” by Teresa Steffen and Megan Seney in the June 2008 issue of PTJ, the authors report an error in the literature review on page 737 (column 1, second paragraph). The sentences in that paragraph should have read (changes in bold): “The MDC95 values for comfortable and fast gait speeds were 0.20 and 0.30 m/s, respectively, for people with osteoporosis60 and 0.12 and 0.22 m/s for people with hip fracture.21 Overall, the MDC95 values for both comfortable and fast gait speeds appeared to be about 0.25 m/s or less for populations tested to date.” The latter sentence originally included the phrase “except for patients with hip fracture, with an MDC95 value of approximately 0.50 m/s,” which was an error. DOI: 10.2522/ptj.20070214.cx

462 ■ Physical Therapy Volume 90 Number 3

Letter 3.10.indd 462

March 2010

2/17/10 2:57 PM

Association Business In Memoriam The deaths of these members were reported to APTA between January 1, 2009, and December 31, 2009. Robert P. Albright Eunice W. Badoux Carol Bankson William T. Bell Jody R. Billingsley Cynthia T. Blanton Eugene H. Bodnar Lulu Boerner Gloria M. Brawley Lester Brower Robert I. Burks Catherine Dalcourt Julia A. Dazen Anne C. Debbaillion Carolyn L. Dieffenbach Dennis F. Driver Lucy S. Driver Walter Dworschak Helen L. Emerson Richard Erhard Martin Feldman

March 2010

Assoc Bus_3.10R.indd 463

Barbara L. Glasow Lenora V. Grizzell Elizabeth A. Hamilton Ina Helweg Joseph Henson Irja R. Hofschire Leonard Hultgren, Jr Daniel Winn Humphrey Wayne E. Kirker Faith MacLennan Alice L. McCleary Trevor Blake McCrory Wayne McKinney Max E. Mickey Jane Murdock Mary C. Noble Donna E. Peddicord Kathryn E. Phillips Pamela M. Pope I. Elayne Resignola Martine K. Rini

William Russell Sarah H. Seybold Scott T. Shuey Gerald W. Shumway Louise M. Shute Hazel C. Siegel M. Irene Snider Edward P. Taylor Mary Took Cheryl L. Turoff Martin R. Vassallo Beverly G. Villwock Dorothy F. Waldo Lynn A. Wallace William M. Wilde R. Lorraine Willer Elizabeth Winn John D. Workman, III Cherie A. Yadao

Volume 90 Number 3 Physical Therapy ■ 463

2/12/10 4:51 PM

Scholarships, Fellowships, and Grants News from the Foundation for Physical Therapy Recent Publications by Foundation-Funded Researchers “Endurance Time is Joint-Specific: A Modelling and Meta-Analysis Investigation,” by Frey Law LA and Avin KG, is featured in Ergonomics (2010;53[1]:109–129). Laura Frey Law, PT, PhD, received a McMillan Doctoral Scholarship in 2000, a Promotion of Doctoral Studies (PODS) I scholarship in 2001, and a PODS II in 2002. Keith Avin, PT, DPT is a 2008 recipient of a Kendall Scholarship as well as a current 2009 PODS I awardee.

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, or 5 entries will be placed for a $40 donation. If you missed your chance to enter at our booth at CSM, 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.

“Low Back Pain in Adolescents: A Comparison of Clinical Outcomes in Sports Participants and Nonparticipants,” by Fritz JM and Clifford SN, was published in the Journal of Athletic Training (2010; 45[1]:61–66). Julie Fritz, PT, PhD, ATC, won a 2002 Research Grant, and Shannon Clifford, PT, PhD, was a 2002 McMillan Doctoral Scholarship winner. “An Expanding Delivery Model Helps Clinicians Access the Literature,” by Guy Simoneau, PT, DPT, ATC, was featured in Journal of Orthopaedic & Sports Physical Therapy (2010;40[1]:1–3). Simoneau won a Foundation Research Grant in 1996. “6-Hz Primed Low-Frequency rTMS to Contralesional M1 in Two Cases with Middle Cerebral Artery Stroke,” by Carey JR, Anderson DC, Gillick BT, Whitford M, and Pascual-Leone A, was published online in Neuroscience Letters on December 16, 2009. James Carey, PT, PhD, received a Foundation award in 1986; Maureen Whitford, PT, received a 2004 McMillan Doctoral Scholarship; and Bernadette Gillick is a 2009 PODS II awardee. “Lower Energy Cost of Skeletal Muscle Contractions in Older Humans,” by Tevald MA, Foulis SA, Lanza IR, and Kent-Braun JA, was published online in American Journal of Physiology—Regulatory, Integrative and Comparative Physiology on December 23, 2009. Michael Tevald, PT, PhD, was awarded a New Investigator Fellowship Training Initiative (NIFTI) in 2007.

464 ■ Physical Therapy Volume 90 Number 3

Foundation 03.10.indd 464

“Neurophysiologic and Rehabilitation Insights from the Split-Belt and Other Locomotor Adaptation Paradigms,” by Reisman DS, Bastian AJ, and Morton SM, was published online ahead of print in Physical Therapy on December 18, 2009, and appears in the February 2010 issue (2010;90[2]:187–195). Darcy Reisman, PT, PhD, is a 1999 McMillan Doctoral Scholarship recipient. Amy Bastian, PT, PhD, received Foundation awards in 1994 and 1995. Susanne Morton, PT, PhD, was the winner of a 2001 PODS I, a 2002 PODS II, and a 2009 Research Grant.

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 awarding 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 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. [DOI: 10.2522/ptj.2010.90.3.464]

March 2010

2/12/10 4:52 PM

Scholarships, Fellowships, and Grants

Thank You for a Great CSM 2010! Thank you to all who stopped by the Foundation’s booth at CSM 2010 or participated in events that benefited the Foundation. The staff enjoyed meeting new faces and catching up with familiar ones. Special thanks to the Home Health Section for hosting the Coffee, along with event sponsor and Foundation Partner in Research Gentiva Health Services. What a way to wake up and start off Friday at CSM—with a nice, fresh cup of joe! Thank you also to

the Sports Physical Therapy Section for hosting the Beach Party Redux & Silent Auction again this year. Attendees enjoyed this evening of entertainment and look forward to next year’s Sports Section Gala. Both of these exciting events benefited the Foundation and will help the Foundation continue to fund outstanding future physical therapist researchers. The Foundation thanks all who attended the forum “Foundation Funding for Post-Professional Doctoral Study: Options and

Guidelines.” The participants were highly motivated to learn tips and advice from Scientific Review Committee (SRC) volunteers. It is no surprise that the quality of applications being submitted to the Foundation continues to rise. Heartfelt thanks are due to the SRC volunteers leading the roundtable discussions. Finally, the Foundation thanks the Section on Research, especially Section Chair Carole Tucker, for sponsoring this session once again.

New in Open Door: Medcom Video Training Programs Collection ProQuest now provides access to Medcom’s video training collection of 75 health care topics important to practitioners. Topics include: safe lifts, falls prevention, wound care, body mechanics, pressure ulcers, documentation, patient safety, HIPAA, medical errors, and more. Each topic contains a series of short presentations (1-5 minutes). View one part of the series, or all, as your time allows. Watch the videos inside the ProQuest interface, or download the files into QuickTime or .mp4 formats.

How do you access the Medcom Video Training Programs Collection? Go to Open Door, access ProQuest, click on the “Browse” tab (third from the left), and select “Video Training Programs” under the “Competency and Training Resources” section. A list of the available video topics appears. Access videos from this page. Bookmark www.apta.org/opendoor for online access to vital clinical research, whenever and wherever you need it. Visit often for full-text access to research and articles from more than a thousand leading clinical and academic publications on topics critical to clinical practice.

Questions? E-mail [email protected] or call 800/999-2782 (ext 8534). Open Door is an APTA members–only benefit.

March 2010

Foundation 03.10.indd 465

Volume 90 Number 3 Physical Therapy ■ 465

2/12/10 4:53 PM