by Moshe Aharoni on June 8, 2014N. Bugnariu, PhD, Physical Therapy Department, University of North...

doi: 10.2522/ptj.20100104Originally published online April 14, 2011

2011; 91:843-861.PHYS THER. Margot Cartizzone and Adam TeavRussell-Doreleyers, Kim Lambert, Alison D. Marshall, Laferrière, Angela Haines-Wangda, MarionStéphane Poitras, Jing Xian Li, Hillel M. Finestone, Lucie Kenny, Denis Prud'homme, Sydney Lineker, Mary Bell,McEwan, Steven P. Messier, Eric Doucet, Glen P. Francoeur, Sarah Milne, Laurianne Loew, JessicaBugnariu, Vivian A. Welch, Gino De Angelis, Lilliane

NicoletaEgan, Claire-Jehanne Dubouloz, Lynn Casimiro, Lucie Brosseau, George A. Wells, Peter Tugwell, MaryAdults Who Are Obese or OverweightGuidelines for the Management of Osteoarthritis in Ottawa Panel Evidence-Based Clinical Practice

http://ptjournal.apta.org/content/91/6/843found online at: The online version of this article, along with updated information and services, can be

Online-Only Material 3.DC1.html

http://ptjournal.apta.org/content/suppl/2011/05/25/91.6.84

Collections

Therapeutic Exercise Practice Guidelines

Patient/Client-Related Instruction Osteoarthritis Osteoarthritis

Obesity Evidence-Based Practice

in the following collection(s): This article, along with others on similar topics, appears

e-Letters

"Responses" in the online version of this article. "Submit a response" in the right-hand menu under

or click onhere To submit an e-Letter on this article, click

E-mail alerts to receive free e-mail alerts hereSign up

by Moshe Aharoni on June 8, 2014http://ptjournal.apta.org/Downloaded from by Moshe Aharoni on June 8, 2014http://ptjournal.apta.org/Downloaded from

http://http://ptjournal.apta.org/content/91/6/843http://ptjournal.apta.org/content/suppl/2011/05/25/91.6.843.DC1.htmlhttp://ptjournal.apta.org/content/suppl/2011/05/25/91.6.843.DC1.htmlhttp://ptjournal.apta.org/cgi/collection/evidencebased_practicehttp://ptjournal.apta.org/cgi/collection/obesityhttp://ptjournal.apta.org/cgi/collection/osteoarthritishttp://ptjournal.apta.org/cgi/collection/osteoarthritishttp://ptjournal.apta.org/cgi/collection/patient_clientrelated_instructionhttp://ptjournal.apta.org/cgi/collection/practice_guidelineshttp://ptjournal.apta.org/cgi/collection/therapeutic_exercisehttp://ptjournal.apta.org/letters/submit/ptjournal;91/6/843http://ptjournal.apta.org/subscriptions/etoc.xhtmlhttp://ptjournal.apta.org/http://ptjournal.apta.org/

Ottawa Panel Evidence-BasedClinical Practice Guidelines for theManagement of Osteoarthritis inAdults Who Are Obese or OverweightLucie Brosseau, George A. Wells, Peter Tugwell, Mary Egan, Claire-JehanneDubouloz, Lynn Casimiro, Nicoleta Bugnariu, Vivian A. Welch, Gino De Angelis,Lilliane Francoeur, Sarah Milne, Laurianne Loew, Jessica McEwan, Steven P.Messier, Eric Doucet, Glen P. Kenny, Denis Prud’homme, Sydney Lineker,Mary Bell, Stéphane Poitras, Jing Xian Li, Hillel M. Finestone, Lucie Laferrière,Angela Haines-Wangda, Marion Russell-Doreleyers, Kim Lambert, Alison D.Marshall, Margot Cartizzone, Adam Teav

Background and Purpose. The objective of this review was to construct anupdated evidence-based clinical practice guideline on the use of physical activity anddiet for the management of osteoarthritis (OA) in adults (�18 years of age) who areobese or overweight (body mass index �25 kg/m2).

Data Sources. Articles were extracted from the following databases: MEDLINE,EMBASE (Current Contents), SPORTDiscus, SUM, Scopus, CINAHL, AMED, BIOMED,PubMed, ERIC, the Cochrane Controlled Trials, and PEDro.

Study Selection. The Ottawa Panel and research assistance team strictly appliedthe inclusion and exclusion criteria from previous Ottawa Panel publications.

Data Extraction. An a priori literature search was conducted for articles relatedto obesity and OA of the lower extremities that were published from January 1, 1966,to November 30, 2010. Inclusion criteria and the methods to grade the recommen-dations were created by the Ottawa Panel.

Data Synthesis. Recommendations were graded based on the strength of evi-dence (A, B, C, C�, D, D�, or D�) as well as experimental design (I for randomizedcontrolled trials and II for nonrandomized studies). In agreement with previousOttawa Panel methods, Cochrane Collaboration methods were utilized for statisticalanalysis. Clinical significance was established by an improvement of �15% in theexperimental group compared with the control group. There were a total of 79recommendations from 9 articles. From these recommendations, there were 36positive recommendations: 21 grade A and 15 grade C�. There were no grade Brecommendations, and all recommendations were of clinical benefit.

Limitations. Further research is needed, as more than half of the trials were oflow methodological quality.

Conclusions. This review suggests that physical activity and diet programs arebeneficial, specifically for pain relief (9 grade A recommendations) and improvedfunctional status (6 grade A and 7 grade C� recommendations), for adults with OAwho are obese or overweight. The Ottawa Panel was able to demonstrate that whencomparing physical activity alone, diet alone, physical activity combined with diet,and control groups, the intervention including physical activity and diet produced themost beneficial results.

Ottawa Panel Members:Ottawa Methods Group:L. Brosseau, PhD, School of Rehabil-itation Sciences, Faculty of HealthSciences, University of Ottawa, andMontfort Hospital Research Insti-tute, Ottawa, Ontario, Canada.

G.A. Wells, PhD, Clinical Epidemiol-ogy Program, Ottawa HospitalResearch Institute, Ottawa Hospital,Civic Campus, Ottawa, Ontario,Canada, and Department of Epide-miology and Community Medicine,University of Ottawa.

P. Tugwell, MD, MSc, Departmentof Epidemiology and CommunityMedicine, University of Ottawa, andCentre for Global Health, Instituteof Population Health, Ottawa,Ontario, Canada.

M. Egan, PhD, School of Rehabilita-tion Sciences, Faculty of Health Sci-ences, University of Ottawa.

C.-J. Dubouloz, PhD, School ofRehabilitation Sciences, Facultyof Health Sciences, University ofOttawa.

L. Casimiro, PhD, Departmentof Academic Affairs, MontfortHospital.

N. Bugnariu, PhD, Physical TherapyDepartment, University of NorthTexas Health Science Center, FortWorth, Texas.

V.A. Welch, PhD, Centre forGlobal Health, Institute of Popula-tion Health.

G. De Angelis, BHSc, Departmentof Epidemiology and CommunityMedicine, and School of Rehabilita-tion Sciences, Faculty of Health Sci-ences, University of Ottawa.

L. Francoeur, BSc, Program in Nutri-tion Sciences, Faculty of Health Sci-ences, University of Ottawa.

S. Milne, MSc, School of Rehabilita-tion Sciences, Faculty of Health Sci-ences, University of Ottawa, andPhysiotherapy Department, Chil-dren’s Hospital of Eastern Ontario,Ottawa, Ontario, Canada.

L. Loew, MSc, School of Rehabilita-tion Sciences, Faculty of Health Sci-ences, University of Ottawa.

J. McEwan, MLIS, University ofOttawa Health Sciences Library.

Author information continues on nextpage.

Research Report

Post a Rapid Response tothis article at:ptjournal.apta.org

June 2011 Volume 91 Number 6 Physical Therapy f 843 by Moshe Aharoni on June 8, 2014http://ptjournal.apta.org/Downloaded from

http://ptjournal.apta.org/

Osteoarthritis (OA) is a degen-erative joint disease thatdirectly affects more than 27million Americans1 and is a primarycause of long-term disability.2 Inturn, disability is known to lead tophysical inactivity and muscle weak-ness, which further affect jointhealth, biomechanics, functional sta-tus, and quality of life (QoL), all thewhile leading to obesity.3 Recog-nized both as an important risk fac-tor for structural joint damage, espe-cially in weight-bearing joints,4,5 andas a global public health problem,6

obesity frequently is present amongphysically inactive people with OA.It increases the risk of developmentand progression of lower-limb OA7

and increases the risk of knee OAby 4 times compared with peoplewith a body mass index (BMI) of�30 kg/m2.8 In addition, obesityhas been shown to affect the jointmechanics9 and cause inflammationamong individuals with OA of theknee—an increased body masscan increase the mechanical stresson joints, causing disarrangementsor abnormalities and a response ininflammatory mediators, resultingin increased pain.10 An increase infat surrounding the quadriceps mus-cles also may increase disabilitydue to a decrease in lower-extremityperformance.11

Regular physical activity is veryimportant for maintaining musclestrength (force-generating capacity),joint structure, joint functioning, andbone health.4 Weight-bearing physi-cal activity has been shown toimprove muscle strength and self-reported measures of pain and phys-ical function among individuals withknee OA,12 improve injury preven-tion,13,14 improve balance control,15

and improve proprioception.16

Fortunately, obesity, physical inac-tivity, and muscle weakness aremodifiable risk factors of OA.17

Physical activity, including walking,

often is recommended for peoplewith OA, people with obesity, andpeople with a combination of bothconditions.5,18–21 Although there isincreasing evidence that suggestsphysical activity is associated withnumerous physical, functional, andQoL benefits, 27.8% of people withOA engage in physical activity on aregular basis compared with 31.0%of those without arthritis.22 The sci-entific literature, including a recentmeta-analysis, suggests that physicalactivity with or without diet hastherapeutic effects on pain, range ofmotion, muscle strength, and func-tional status.21,23–28 Recent guide-lines suggest physical activity or dietfor adults with OA,10,11 althoughthese guidelines were not focusedon a target population of adults whowere obese or overweight.

As summarized in previous OttawaPanel scientific articles, “Evidence-based clinical practice guidelines(EBCPGs) are precise statements onrecommended interventions that arebased on scientific literature andinclude a graded strength of evi-dence.”29 This EBCPG aims to usethe Ottawa Panel’s familiarity withEBCPGs29–33 to contribute to thefield of OA management by: (1) focus-ing on specific characteristics andneeds of individuals with OA whoare obese or overweight and (2) giv-ing practitioners (family physicians,kinesiologists, dietitians, physicaltherapists, physiatrists, rheumatolo-gists, and others) concise and up-to-

External Experts:S.P. Messier, PhD, Department of Health andExercise Science, J.B. Snow Biomechanics Lab-oratory, Wake Forest University, Winston-Salem, North Carolina.

E. Doucet, PhD, School of Human Kinetics,Faculty of Health Sciences, University ofOttawa.

G.P. Kenny, PhD, School of Human Kinetics,Faculty of Health Sciences, University ofOttawa.

D. Prud’homme, MD, MSc, School of HumanKinetics, Faculty of Health Sciences, Universityof Ottawa.

S. Lineker, PhD, The Arthritis Society, Toronto,Ontario, Canada.

M. Bell, MD, Continuing Education andKnowledge Transfer, University of Toronto.

S. Poitras, PhD, School of Rehabilitation Sci-ences, Faculty of Health Sciences, Universityof Ottawa.

J.X. Li, PhD, School of Human Kinetics, Facultyof Health Sciences, University of Ottawa.

H.M. Finestone, MD, SCO Health Services,Elisabeth Bruyère Health Centre, Ottawa,Ontario, Canada.

L. Laferrière, MHA, Directorate Force HealthProtection, Canadian Forces Health ServicesGroup Headquarters, National Defense,Ottawa, Ontario, Canada.

A. Haines-Wangda, MSc, The Ottawa Hospital,General Campus, Ottawa, Ontario, Canada.

M. Russell-Doreleyers, MSc, The Arthritis Soci-ety, Ottawa, Ontario, Canada.

K. Lambert, MSc, The Arthritis Society,Ottawa.

A consumer with osteoarthritis and obesity.

Assistant Manuscript Writers:A.D. Marshall, MA, School of RehabilitationSciences, Faculty of Health Sciences, Universityof Ottawa.

M. Cartizzone, MSc, School of RehabilitationSciences, Faculty of Health Sciences, Universityof Ottawa.

A. Teav, BSc, School of Human Kinetics, Fac-ulty of Health Sciences, University of Ottawa.

Address all correspondence and requests forreprints to: Lucie Brosseau, PhD, School ofRehabilitation Sciences, Faculty of HealthSciences, 451 Smyth Rd, University ofOttawa, Ottawa, Ontario, Canada K1H 8M5([email protected]).

[Ottawa Panel Evidence-Based Clinical Prac-tice Guidelines for the Management ofOsteoarthritis in Adults Who Are Obese orOverweight. Phys Ther. 2011;91:843–861.]

© 2011 American Physical Therapy Association

Available WithThis Article atptjournal.apta.org

• eAppendix: Details of IncludedStudies

• Audio Abstracts Podcast

This article was published ahead ofprint on April 14, 2011, atptjournal.apta.org.

Management of Osteoarthritis in Adults Who Are Obese or Overweight

844 f Physical Therapy Volume 91 Number 6 June 2011 by Moshe Aharoni on June 8, 2014http://ptjournal.apta.org/Downloaded from

mailto:[email protected]://ptjournal.apta.org/content/91/6/843/suppl/DC1http://ptjournal.apta.org/content/91/6/suppl/DC1http://ptjournal.apta.org/

date knowledge on physical inter-ventions, especially physical activitywith and without diet.

MethodThis project used the same meth-ods26 as those of a previous studyconducted by the Ottawa Panel ontherapeutic exercise for patientswith OA.21 Methodological qualitywas graded using the Jadad scale,34 a5-point measure that gives 2 pointsfor reliability and validity (ie, whetherthe study was double-blinded andwhether the double-blinded methodwas appropriate), 2 points for ran-domization (ie, whether the studyinvolved randomization and whetherthe randomization method was appro-priate), and 1 point for explana-tion of participant withdrawals anddropouts.

In conjunction with the methodsof previous Ottawa Panel publica-tions,29,33 the construction ofEBCPGs was developed using theAppraisal of Guidelines Researchand Evaluation (AGREE) criteria(www.agreecollaboration.org). TheOttawa Panel individual recommen-dations were graded as A, B, C, C�,

D, D�, or D� based on the strengthof evidence (Tab. 1). An alphabet-ical grading system was presentedaccording to the Ottawa Panel meth-ods,29 marked with an asterisk inTable 1, and an additional alpha-betical system recently adopted bythe Cochrane Collaboration (www.cochrane.org) has the correspond-ing levels in parenthesis.

Literature SearchA library scientist conducted anextensive a priori literature searchfor articles related to obesity and OAof the knee in June 2009. ApplyingCochrane Collaboration search tech-niques, the search included articlespublished from January 1, 1966, toNovember 31, 2010, and wereextracted from the following data-bases: MEDLINE, EMBASE (Cur-rent Contents), SPORTDiscus, SUM,Scopus, CINAHL, AMED, BIOMED,PubMed, ERIC, the Cochrane Con-trolled Trials, and PEDro. A handsearch of the reference lists of poten-tial case-control studies (CCSs) alsowas performed.

Study Inclusionand Exclusion CriteriaThe Ottawa Panel and research assis-tance team strictly applied the inclu-sion and exclusion criteria (Tab. 2)so that every included article met thespecific intervention, study design,participant, and outcome criteria.These precise selection methods aredescribed in previous Ottawa Panelpublications.29,33 These inclusionand exclusion criteria were approvedthrough Ottawa Panel consensus.

InterventionsStudies that applied physical activity,diet, or both for the management ofOA in adults who were obese oroverweight were included. Studieswere not included if interventionsincluded surgery, injections to thelower extremities, medication forweight loss, medication for manage-ment of OA symptoms, acupuncture,or multidisciplinary and function res-toration programs (Tab. 2).

Study DesignsStudies that were randomized con-trolled trials (RCTs), controlled clin-ical trials (CCTs), cohort studies, andhead-to-head studies (eg, diet versus

Table 1.Combined Grading Recommendationsa

Grade Clinical ImportanceStatistical

Significance Study Design

Grade A* (stronglyrecommended**)

�15% P�.05 RCT (single or meta-analysis)

Grade B* �15% P�.05 CCT or observational (single or meta-analysis)

Grade C�* (use suggested**) �15% Not significant RCT/CCT or observational (single or meta-analysis)

Grade C* (neutral**) �15% Not significant Any study design

Grade D* (neutral**) �15% (favors control) Not significant Any study design

Grade D�* (use notsuggested**)

�15% (favors control) Not significant RCT/CCT or observational (single or meta-analysis)

Grade D�* (strongly notrecommended**)

�15% (favors control) P�.05 (favors control) Well-designed RCT with �100 participants(if �100 participants, becomes a gradeD recommendation)

a Combined grading recommendations according to the Ottawa Panel29,33 for alphabetical grading system (indicated by asterisk) and to the CochraneCollaboration (www.cochrane.org) for international nominal grading system (indicated by double asterisk). RCT�randomized controlled trial, CCT�clinicalcontrolled trial.



www.agreecollaboration.orgwww.cochrane.orgwww.cochrane.orgwww.cochrane.orghttp://ptjournal.apta.org/

physical activity, as opposed to dietversus control) were included. Addi-tionally, only articles published inEnglish or French were included inorder to diminish time and transla-tion costs. Studies were excludedif they were uncontrolled cohortstudies, case studies, reviews, orguidelines; were conducted with nocomparison group; provided datawithout means and standard devia-

tions; reported a �20% dropout rate;or had a sample of fewer than 5patients per group (Tab. 2).

ParticipantsStudies must have had adult partici-pants (�18 years of age) who wereidentified as overweight (BMI �25kg/m2) or obese (BMI �30 kg/m2)and were affected by OA in thelower extremities.6 Studies were

excluded if participants had can-cer or other oncologic conditions,pulmonary conditions, cardiac con-ditions, dermatological conditions,neurological conditions, other rheu-matologic or musculoskeletal con-ditions, pediatric conditions (eg,juvenile arthritis), or psychiatric con-ditions (Tab. 2).

OutcomesSeveral types of outcomes were ofrelevance for this article (Tab. 2).Note that researchers might haveused different instruments to mea-sure the same outcome (eg, oneresearcher might use the LequesneIndex to measure the concept offunctional status, whereas anotherresearcher might use the WesternOntario and McMaster UniversitiesOsteoarthritis Index [WOMAC]).Outcomes of primary interest werefunctional status (Lequesne Index,WOMAC), physical function (Medi-cal Outcomes Study 36-Item Short-Form Health Survey questionnaire[SF-36] physical function score), self-efficacy (stair climbing test), painscale (visual analog scale [VAS]score), endurance (Six-Minute WalkTest), stiffness (WOMAC stiffnessscore), strength (hamstring andquadriceps muscle strength), torque(concentric knee extension), bodycomposition (waist circumference,body weight, BMI, lean body mass,and body fat mass), mental health(SF-36 mental health score), psycho-logical well-being (physical functionand body satisfaction measure), andmobility (walking speed) (Appen-dix). Articles were excluded if out-come measures were biomechanicalmeasures, biochemical measures, orserum markers (Tab. 2).

Study SelectionFollowing the literature search, 2reviewers from the research assis-tance team were trained by theOttawa Panel to systematically clas-sify the articles into inclusion andexclusion groups using the criteria

Table 2.Inclusion and Exclusion Criteriaa

Inclusion Exclusion

Interventions• Eligible control groups: placebo, untreated,

routine conventional therapy, activephysical therapy treatments, educationalpamphlets (lifestyle modification)

• Eligible interventions: physical activity(aerobic or strengthening, stretching),dietary modifications, behavioral support,electrotherapy

Interventions• Surgery or injections to lower-extremity joints• Medication for weight loss• Medication changes for management of OA

symptoms• Multidisciplinary, functional restoration programs

Study Designs• Randomized controlled trials• Clinical controlled trials• Head-to-head comparison of physical

activity and diet studies

Study Designs• Case series/case reports• Cohort studies• Case-control studies• Reviews and guidelines• Data without a mean and SD• Sample size of �5 participants per treatment

group• Studies with �25% dropout rate

Participants• Outpatients or inpatients• Diagnosis of OA (lower extremity)• Age groups �18 y• Mixed population (only if OA and RA)• BMI �25.0 kg/m2 for overweight and

�30.0 kg/m2

for obesity

Participants• Oncologic conditions• Pulmonary conditions• Neurologic conditions• Cardiac conditions• Dermatologic conditions• Pediatric conditions (juvenile arthritis)• Psychiatric conditions• No known pathology or impairments• Multiple conditions

Outcomes• Balance status• Body composition, girth, weight, BMI• Cardiopulmonary functions• Disease activity, progression• EMG activity• Functional status• Joint imaging• Mobility, flexibility, range of motion• Muscle force, endurance, power• Pain relief• Patient adherence• Patient satisfaction• Psychological well-being• Quality of life• Self-efficacy• Weight loss

Outcomes• Biochemical measures• Serum markers

a OA�osteoarthritis, RA�rheumatoid arthritis, SD�standard deviation, BMI�body mass index,EMG�electromyographic.




constructed by the Ottawa Panel(Tab. 2). Separately, the 2 review-ers read each article and com-piled lists of articles to be includedor excluded. When disagreementemerged, the primary investigator ofthe Ottawa Panel was consulted, anda consensus was reached on the ulti-mate placement of the article.

Data Extraction andMethodologicalQuality AssessmentThe 2 reviewers from the researchassistance team independentlyextracted data from the articlesretrieved from the literature searchwith the use of data extractionforms. Information of interestincluded study design, intervention,treatment groups, method, results,and quality scoring information. Thequality scoring information was usedto assess the articles according to theJadad scale.

Data AnalysisTo analyze the data, the OttawaPanel applied a Cochrane Collab-oration statistical analysis similar tothat of past Ottawa Panel publica-tions.29,33 The weighed mean differ-ence (WMD), the absolute benefit,and the percentage of relativechange between intervention andcontrol groups were calculated usingcontinuous data. Relative risks wereutilized to analyze dichotomous data(ie, data that can easily be separatedinto 2 or more categories). Calcula-tions were made with the ReviewManager (RevMan) computer soft-ware program.*

ResultsLiterature SearchFrom the November 2010 search,the library scientist found 114 arti-cles on the management of OA inadults who were obese or over-

weight. Of these articles, 63 wereidentified as potentially relevant.

The Ottawa Panel and the researchassistance team ultimately agreed,according to the inclusion andexclusion criteria (Tab. 2), on theselection of 10 articles that wereCCSs35–44 (see Appendix andeAppendix, available at ptjournal.apta.org). Initial disagreements onCCS selection were resolved throughOttawa Panel consensus.

Fifty-three studies were excludedfor various reasons. Studies wereexcluded if they showed no inter-vention,45–62 if they were not clinicaltrials,4,9,63–74 if they possessed irrele-vant outcome measures,75–78 if theywere CCSs,79 if they were cohortanalytic studies,80 if they containedgastroplasty interventions,81 if theydid not have any controls,82 ifthere were no data past baseline,83

if the dropout rate was �20%,84 ifthey were literature reviews,27,85

if they were case series,86 if therewas no description of weight loss,87

if the study population did not haveOA,88–90 if there were no compari-son groups,91 if there were no groupresults,92 or if they were concurrentstudies involving nonsteroidal anti-inflammatory drugs or medicinalweight loss interventions.93,94

Methodological QualityThe Jadad scale identified 5 arti-cles35,37,38,44,50 as being of highmethodological quality (�3) andthe remaining 5 trials as being oflow methodological quality,36,39,40,42,43

due mainly to the double-blindnesscriterion. Although they achieved alow score, these studies had goodmethods and were included in ourdatabase. The methodological qual-ity level, however, was mentionedwith each recommendation (Appen-dix) and detailed in the eAppendix.The Ottawa Panel agreed that it isdifficult to implement a double-blindness criterion for interventions

that include aerobic physical activ-ity (ie, no matter the alternativeproposed, the patients will knowwhether they participated in anaerobic physical activity program).More weight was given to whetherthe study method involved random-ization, rather than double blinding(see Appendix and eAppendix for asummary of the trials).

Effectiveness of PhysicalActivity or Diet for OA and ObesityDue to publication restrictions, onlythe CCSs35,37,38,50 with a methodo-logical quality of �3 according to theJadad scale34 will be presented in thefollowing section. For the same rea-son, only 2 figures among 50 wereselected (see Appendix and theeAppendix for an overview of theresults of all the included CCSs).

In regard to low-energy diet versusconventional high-protein diet (con-trol), an RCT by Christensen et al35

(N�96) (Tab. 3, Fig. 1; also seeAppendix and eAppendix) showedclinically important benefits withstatistical significance for improvedfunctional status (WOMAC totalscore) at end of treatment (8 weeks)(relative difference��24%) and forimproved physical function (WOMACfunction score) at end of treatment(relative difference��26%) (resultsnot shown). Clinical significancewithout statistical significance wasfound for improved pain relief(WOMAC pain score) at end of treat-ment (relative difference��15%)and for reduced stiffness (WOMACstiffness score) at end of treatment(relative difference��17%) (resultsnot shown).

For the intervention of exerciseand diet versus healthy lifestyle(control), an RCT by Focht et al41

(N�252) (see Appendix and eAppen-dix) exhibited clinically importantbenefits without statistical signifi-cance for improved self-efficacy onstairs (stair climbing) at end of treat-

* Copenhagen, Denmark: The Nordic Coch-rane Centre, The Cochrane Collaboration(www.Cochrane.org).



http://ptjournal.apta.org/content/91/6/843/suppl/DC1www.ptjournal.apta.orgwww.ptjournal.apta.orghttp://ptjournal.apta.org/content/91/6/843/suppl/DC1http://ptjournal.apta.org/content/91/6/843/suppl/DC1http://ptjournal.apta.org/content/91/6/843/suppl/DC1http://ptjournal.apta.org/content/91/6/843/suppl/DC1http://ptjournal.apta.org/content/91/6/843/suppl/DC1http://ptjournal.apta.org/content/91/6/843/suppl/DC1www.Cochrane.orghttp://ptjournal.apta.org/

ment (18 months) (relative differ-ence�17%). Moreover, it showedclinically important benefits with sta-tistical significance for improvedself-efficacy in walking (Six-MinuteWalk Test) at end of treatment (rela-tive difference�20%) (Tab. 4, Fig. 2).

A study by Messier et al37 (N�252),which measured the effects ofexercise and diet versus healthy life-style (control) (see Appendix andeAppendix), showed clinically im-portant benefits with statistical sig-nificance for improved endurance(Six-Minute Walk Test) at end oftreatment (6-month assessment)(relative difference�17%) and at endof treatment (18-month assessment)(relative difference�16%) (resultsnot shown). Clinically importantbenefits without statistical signifi-cance were demonstrated forimproved mobility (stair climbing) atend of treatment (6-month assess-ment) (relative difference�25%) andend of treatment (18-month assess-ment) (relative difference�23%)(results not shown).

Interestingly, the CCSs with highermethodological quality35,37,38,41 alsoobtained positive results (grades Aand C�) for several clinical out-comes when there were compari-sons between exercise only and thecontrol intervention and between

Table 3.Results for the Relative Difference for Low-Energy Diet Versus Conventional High-Protein Dieta

StudyTreatment

Group OutcomeNo. of

ParticipantsBaseline

MeanEnd-of-Study

MeanAbsoluteBenefit

RelativeDifferencein Change

FromBaseline WMD/95% CI

Christensen et al35

(2005)Low-energy diet Lean body mass

(kilograms)Higher is betterEnd of treatment

(8 wk)

40 50.6 47.6 �1.80 �4% WMD: 1.80CI low: 0.69CI high: 2.91

Conventionalhigh-proteindiet

40 51.1 49.9

a WMD�weighted mean difference, CI�confidence interval.

Favors ControlFavorsTreatment

–3.0 –2.5 –2.0 –1.5 –1.0 –0.5 0.0 0.5 1.0

Christensen et al,35

2005

End of treatment (8 wk)

Change infat-free

mass (kg)

Low-Energy Diet Versus ConventionalHigh-Protein Diet Control: Change in

Physical Fitness

Figure 1.Low-energy diet versus conventional high-protein diet (control): change in physicalfitness.



http://ptjournal.apta.org/content/91/6/843/suppl/DC1http://ptjournal.apta.org/

diet only and the control interven-tion. The results showed a largernumber of grades A and C� whenexercise and diet were combinedcompared with the control inter-vention (Appendix). It should benoted that Rejeski et al,38 in anRCT with similar comparativegroups,35,37,41 showed positive rec-ommendations for psychological well-being and QOL. There were nograde B recommendations.

A grade A recommendation (Tab. 1),according to the Ottawa Panelgrading system, was obtained inthose RCTs that were found tobe both clinically and statisticallysignificant. The following outcomesreceived a grade A recommendation:psychological well-being,38 func-tional status,35,36,40 strength,36,43

torque,43 mobility,38,40,42 walkingendurance,39 self-efficacy during

stair climbing,41 self-efficacy in walk-ing,41 pain relief,36,40,42 and bodycomposition.36,40

A grade C� recommendation(Tab. 1) was given by the OttawaPanel for an RCT or CCT with clini-cal but not statistical significance.Grade C� recommendations werefound for psychological well-being,38 body composition,35,36,43

functional status,35 pain relief,35,40

transfer pain frequency,39 transferpain intensity,39 stiffness relief,35

self-efficacy during stair climbing,41

self-efficacy in walking,41 mobility,37

and torque.45

A grade C recommendation (Tab. 1)was granted by the Ottawa Panelwhen neither clinical significancenor statistical significance was found.These outcomes included mentalstatus,38 functional status,35,36,38,39

psychological well-being,38 bodycomposition,35,36,43 walking endur-ance,36,37,39,40 mobility,37,39,42 self-efficacy on stairs,41 self-efficacy inwalking,41 pain relief,37,42 ambula-tion pain frequency,39 transfer painintensity,39 transfer pain frequency,39

strength,36 and torque.43

A grade D recommendation (Tab. 1)was given for outcomes that demon-strated clinical importance favoringthe control group of less than 15%.Ottawa Panel grade D recommenda-tions were given for: pain relief,37,39

functional status,39 stiffness,40 bodycomposition,35,36,40 strength,36 andmental status.38

A grade D� recommendation(Tab. 1) was given to outcomes dem-onstrating clinical importance favor-ing control of �15% for the control

Table 4.Results for the Relative Difference for Exercise and Diet Versus Healthy Lifestyle (Control)a

StudyTreatment

Group OutcomeNo. of

ParticipantsBaseline

MeanEnd-of-Study

MeanAbsoluteBenefit

RelativeDifferencein Change

FromBaseline WMD/95% CI

Focht et al41

(2005)Exercise and diet Self-efficacy (stair

climbing)Higher is betterEnd of treatment

(18 mo)

58 63.84 77.25 11.37 17% WMD: �4.97CI low: �13.98CI high: 4.04

Healthy lifestyle 67 70.24 72.28

Focht et al41

(2005)Exercise and diet Self-efficacy (Six-

Minute WalkTest)

Higher is betterEnd of treatment

(18 mo)

58 66.44 84.95 13.49 20% WMD: �12.05CI low: �21.97CI high: �2.13


Messier et al37

(2004)Exercise and diet WOMAC pain score

Lower is betterEnd of treatment

(6 mo)

63 7.27 5.47 �0.74 �10% WMD: �0.72CI low: �2.01CI high: 0.57


Messier et al37

(2004)Exercise and diet WOMAC pain score

Lower is betterEnd of treatment

(18 mo)

58 7.27 5.07 �0.97 �13% WMD: �0.95CI low: �2.22CI high: 0.32


a WMD�weighted mean difference, CI�confidence interval, WOMAC�Western Ontario and McMaster Universities Osteoarthritis Index.




group. Only one outcome, painrelief,39 received a D� grade.

Lastly, grade D� recommendations(Tab. 1) were given for outcomesthat showed both clinical and statis-tical significance favoring the controlgroup. According to the OttawaPanel grading system, psychologicalwell-being38 received a D� grade.

DiscussionThe Ottawa Panel was able todemonstrate that, when comparingphysical activity alone, diet alone,

physical activity combined with diet,and control interventions, the inter-vention including physical activityand diet produced the most bene-ficial results (a total of 86 positiverecommendations: 22 grade A and23 grade C�), specifically in clinicaloutcomes such as pain relief,strength, functional status, and QoL.

However, the Ottawa Panel recom-mendations regarding weight loss(�5%) and body compositionchanges in study participants withOA who were obese or overweight

were not as convincing. Although5% weight loss combined with phys-ical activity provided improvementin clinical outcomes, the weight losswas not sufficient to slow the diseaseprogression.37

Christensen et al35 showed that alow-energy diet was more effectivein improving functional status andphysical function than a conven-tional high-protein diet, and a strongnegative correlation was shownbetween increases in physical func-tion and decreases in percentage ofbody fat among an elderly popula-tion. Focht et al41 showed that phys-ical activity and diet can improveself-efficacy in walking, strength, painrelief, and functional status. Theintervention period was a sufficientlength (18 months), and the studyhad a sufficient sample size (N�316). Messier et al37 used a physicalactivity and diet intervention for asufficient length of time (6 and 18months) with a large and adequatesample size (N�316). The results ofthis study showed that the combina-tion of physical activity and dietimproved mobility, pain relief, andendurance. The results also showedthat intensive weight loss improvedfunctional status, pain relief, mobil-ity, endurance, and torque. Rejeskiet al38 had adequate diet and physicalactivity intervention periods (6 and18 months) with a large sample size(N�316). This study was unique, asit assessed QoL as an outcome mea-sure using the SF-36. The results ofthis study were interesting, as theyfavored the control group.

Lim et al44 demonstrated severaldifferences with various exercisemodes among individuals with OA.Aquatic exercise including endur-ance, strength, and aerobic trainingwas shown to be more effective inreducing pain compared with ahome-based exercise program con-sisting of strengthening exercises.44

Land-based exercise performed in

–25 –20 –15 –10 –5 0 5 10

FavorsExercise +Diet

FavorsControl

Focht et al,41 2005

Focht et al,41 2005

End of treatment (18 mo)

Self-efficacy (6MWT)

End of treatment (18 mo)

Self-efficacy (stair climbing)

Exercise and Diet Versus Healthy Lifestyle:Self-Efficacy

Figure 2.Exercise and diet versus healthy lifestyle (control): self-efficacy. 6MWT�Six-MinuteWalk Test.




a gym that included strength, aero-bic, stretching, and range-of-motiontraining was shown to be more effec-tive in reducing pain compared witha home-based exercise program con-sisting of strengthening exercises(Appendix).44

The significant pain relief despitenonsubstantial weight loss amongstudy participants with OA whowere obese or overweight can beattributed to the beneficial effectsof exercise on the musculoskeletalsystem.36–38 The significant improve-ments in functional outcomes,despite a relatively low percentageof weight loss, among study partici-pants with OA who were obese oroverweight may be attributed to thebeneficial effects of exercise on thecardiovascular and neuromuscularsystems. Regular physical activityinduced improvements in endur-ance, strength, and balance, whichresulted in improved abilities to per-form activities of daily living andimproved participation in leisureactivities.

The significant improvements inQoL among study participants withOA who were obese or overweightmay be a reflection of mental healthand social benefits associated withthe participation in physical activi-ties. Physical activity promotes psy-chological well-being by reducingfeelings of fatigue, depression, andanxiety and improving self-esteem,confidence, concentration, and men-tal awareness. The social benefitsof participation in physical activityinclude a reduction in the sense ofisolation and loneliness, improvedsocial networks and social capital,and increased community connect-edness and cohesion.

The recent conceptual frameworkunderlying the Intensive Diet andExercise for Arthritis (IDEA) studyprotocol17 proposed that weightloss, which would reduce joint load

and inflammation, may improvemobility, body composition, andstrength and result in reduced painand an enhanced functional out-come and QoL in individuals withOA who are obese. A multidis-ciplinary team is recommendedto investigate the complex healthcomplications related to obesity.95

Weight loss can be facilitated bydietitians, and physical therapistsplay a key role in enrolling indi-viduals who are obese or over-weight in lower-extremity musclestrengthening and balance exerciseprograms74,96–98 after significantweight loss9 in order to maintainand develop lean muscle mass,91

to improve the biomechanics andstability of the knee, and to indirectlyslow down the progression ofknee OA.99–101 These biomechanicalimprovements can potentiallyreduce pain and enhance functionalstatus in individuals with OA whoare obese.

Behavioral strategies such as patienteducation, health counseling, realis-tic and achievable goal setting,telephone contacts, 7-day physicalactivity and dietary logs, social/peer support, self-management, self-efficacy, and positive feedback,either alone or in numerous com-binations, have been studied in pop-ulations without OA and shouldbe explored in depth in individualswith OA who are obese or over-weight.102,103 However, relapse afterweight loss in individuals with nospecific disability is rampant evenwith behavioral interventions.104

This may be an additional challenge,especially in older individuals whoare disabled.

Although the Ottawa Panel memberscannot strongly support the posi-tive recommendations that emergedfrom the CCSs with lower methodo-logical quality,36,39,40,42,43 these CCSsprovide good research hypothesesthat can be further explored for

future RCTs related to: (1) intensivediet40,42 (2) combined pain ther-apy,43 (3) longer follow-up,36,37 and(4) more-quantitative outcomes tomeasure weight loss36 (Appendix).

LimitationsAlthough the Ottawa Panel guide-lines were developed using rigor-ous quantitative methods,36 a poten-tial publication bias may haveoccurred because only those articleswritten in English and French wereincluded. This selection processinevitably misrepresents the amountof research that has been conductedon OA and obesity globally. Addition-ally, the Ottawa Panel is made up ofclinical experts from North Americaas opposed to being more interna-tionally focused with clinical prac-titioners and researchers outsideof North America, such as thosepeople who make up the Osteo-arthritis Research Society Inter-national (OARSI) Panel. The currentOttawa Panel recommendations dem-onstrated herein differ from the2008 recommendations produced bythe OARSI18 in that the Ottawa Panelexamined the efficacy of physicalactivity and diet programs in themanagement of OA in adults whowere obese or overweight and didnot include surgical or pharmaco-logical interventions.

Of the included studies, 2 showedlimitations: the study by Christensenel al35 had a small study sample(N�71) and a short interventionlength (8 weeks), and the study byMiller et al40 focused solely on theoutcomes of an intensive weight lossintervention. The intervention dura-tion was sufficient (6 months); how-ever, the study sample size was quitesmall (N�87). The D grades (ie, clin-ical importance is �15% and favorscontrol) in functional status and painrelief are somewhat misleading. Atotal of 3 D grades were given inhead-to-head studies (eg, diet versusphysical activity) in which physical




activity alone was favored over acombination of diet and physicalactivity.17,36

The methodological quality (ie, thetype and duration) of diet or physicalactivity programs must be takeninto consideration (Appendix). It isnot certain why the physical activitygroups fared better for these out-comes compared with the combineddiet and physical activity groups.We hypothesize that: (1) physicalactivity improved functional mea-sures, (2) diet improved self-reportmeasures, and (3) diet and physicalactivity improved both functionaland self-report measures.

Although radiographic evidence isprimarily used for the diagnosis ofOA, there may be difficulties withthe diagnosis if there is a lack ofeither physical or laboratory find-ings, causing an inconsistencybetween symptoms and the resultsof the radiographic examinations.105

In these situations, classic clinicalcriteria for knee OA and gradingscales are used to support and clarifythe diagnosis.105

The Jadad scale was selected overother quality measure alternatives,such as the PEDro scale, to remainconsistent with Cochrane Collabora-tion methods. Olivo et al106 foundthat the Jadad scale exhibits the bestpsychometric quality for assessingthe methodological quality of CCTs,even though single- and double-blindstudies are difficult to conduct107

in OA and obesity research. How-ever, sound methodological researchin this area is both necessary andimportant. To the knowledge of theOttawa Panel, there is no conclusiveevidence on the most appropriatemethodological scale to apply forOA and obesity research. The use ofBMI, waist circumference, and bodyweight as valid indicators of suc-cessful weight loss in individualswith OA is debatable because these

measurements do not discriminatebetween lean and fat body mass.108

Body composition measurements aremore valid.29

Further research should examineother alternative therapies for weightreduction that were not included inthis review. Toda109 found somephysical benefits of mazindol dietpills for weight reduction in anelderly population (ie, 45–69 yearsof age) in Japan. Such results promptfurther discussion on the use of dietpills and whether their regular usagewould be beneficial or practical(eg, cost) for specific populations ofpeople with knee OA (ie, elderlypeople, people who are obese oroverweight). Although both strength-ening and aerobic exercises areeffective for OA,110,111 another inter-esting topic that warrants furtherresearch and that would helpaddress more-specific activity pre-scription is the differences in aerobicversus resistance training activity.

Three groups of studies—Messier etal37 (2004) and Rejeski et al38 (2002);Messier et al39 (2000), Miller et al40

(2006), and Messier et al17 (2009);and Focht et al41 (2005), Wang et al43

(2007), Focht et al50 (2002), andFocht et al51 (2004)—come from thesame laboratory, which may mis-represent the depth of availableresearch and may bias the conclu-sions that can be derived from a lim-ited database. Adults with OA whoare obese or overweight are a chal-lenging population to study becauseit is not known exactly how manyof the physical impairments aredue to weight problems versus thejoint condition.35 Because individu-als with OA generally are older, theyalso have many associated medicalconditions. There also is evidencefrom a relatively large randomizedsubject pool (N�316) that partici-pants with comparatively less painat baseline demonstrated more ben-efits at follow-up.41 More research is

needed on how programs for manag-ing OA for individuals with obesitycan be optimally adjusted to maxi-mize benefits for various levels ofknee OA disability.

Implications for PracticeThe Ottawa Panel found importantevidence to support the use ofdiet or physical activity programsfor the overall management of OAof the knee. Results of positive rec-ommendations (grades A and C�)from included studies with highmethodological quality (Jadad scalescore 3)20,37,38,41,44 indicate that dietor physical activity programs werepromising for short-term (6-month)pain relief (2 grade A recommen-dations, 6 grade C� recommenda-tions); for long-term follow-up(18-month) pain relief (3 grade C�recommendations); and for theimprovement of torque (2 grade C�recommendations), functional sta-tus (2 grade A recommendations,2 grade C� recommendations),self-efficacy (2 grade A recommen-dations, 2 grade C� recommenda-tions), endurance (2 grade A recom-mendations), mobility (1 grade Arecommendation, 3 grade C� rec-ommendations), and psychologicalwell-being (2 grade A recommen-dations, 1 grade C� recommenda-tion). Groups that received bothphysical activity and diet producedthe best results compared with phys-ical activity–only groups, diet-onlygroups, and control groups.

The Ottawa Panel recommendsreducing weight prior to the imple-mentation of weight-bearing exer-cise to maintain joint integrity andto avoid joint disease and dysfunc-tion. The Ottawa Panel also recom-mends the inclusion of diet orphysical activity programs in themanagement of OA among individu-als who are obese or overweight.Because physical therapists are notexperts in diet, it is suggested thatthey work with an interdisciplinary




team including dietitians. Additionalresearch is needed on the specificcomponents of the interventionsthat produced clinical benefits forthe older adult populations so thattreatment can become more stan-dardized and easier to prescribe bypractitioners and clinicians. Lastly,more knowledge is needed on howOA management programs can bebetter tailored to increase long-term adherence, especially amongpeople with OA who are obese oroverweight.

The Ottawa EBCPGs Development Groupis indebted to the following people for theirtechnical support and help with data extrac-tion and data analysis: Dr Carole Oatis,Dr Norma MacIntyre, Dr France Légaré,Ms Jessie McGowan, Ms Rachel Goard,Ms Isabelle Lajoie, Ms Jennifer O’Neil,Ms Kathryn Easy, Mr Mohamed Abdi,Ms Laura Trafford, Mr George Gray, andMs Nathalie Jean.

This study was financially supported byThe Arthritis Society (Canada), the OntarioMinistry of Health and Long-Term Care(Canada), The University of Ottawa, Facultyof Health Sciences, and the Ministry ofHuman Resources, Summer Students Pro-gram (Canada).

This article was submitted March 22, 2010,and was accepted January 9, 2011.

DOI: 10.2522/ptj.20100104

References1 Lawrence RC, Helmick CG, Arnett FC,

et al. Estimates of the prevalence ofarthritis and selected musculoskeletaldisorders in the United States. ArthritisRheum. 1998;41:778–799.

2 Walker JM, Helewa A. Physical Therapyin Arthritis. Philadelphia, PA: WB Saun-ders Co; 2004.

3 Catenacci VA, Hill JO, Wyatt HR. Theobesity epidemic. Clin Chest Med. 2009;30:415–444, vii.

4 Wearing SC, Hennig EM, Byrne NM, et al.Musculoskeletal disorders associatedwith obesity: a biomechanical perspec-tive. Obes Rev. 2006;7:239–250.

5 Brady TJ, Sniezek JE. Implementing theNational Arthritis Action Plan: newpopulation-based approaches to increas-ing physical activity among people witharthritis. Arthritis Rheum. 2003;49:471–476.

6 Obesity, preventing and managing theglobal epidemic: report of a WHO con-sultation. World Health Organ Tech RepSer. 2000;894:i–xii,1–253.

7 Janken EA, Collins A, Kosak AT. Over-view of the relationship between painand obesity: what do we know, where dowe go next? J Rehabil Res Dev. 2007;44:245–262.

8 Anderson JJ, Felson DT. Factors associ-ated with osteoarthritis of knee in thefirst national health and nutrition exami-nation survey (HANES I): evidence for anassociation with overweight, race, andphysical demands of work. Am J Epide-miol. 1988;128:179–189.

9 Hartz AJ, Fischer ME, Bril G, et al. Theassociation of obesity with joint painand osteoarthritis in the HANES data.J Chronic Dis. 1986;39:311–319.

10 Griffen TM, Guilak F. The role ofmechanical loading in the onset and pro-gression of osteoarthritis. Exerc Sport SciRev. 2005;33:195–200.

11 Zhai G, Blizzard L, Srikanth V, et al. Cor-relates of knee pain in older adults: Tas-manian older adult cohort study. Arthri-tis Rheum. 2006;15:264–271.

12 Lange AK, Vanwanseele B, FiataroneSingh MA. Strength training for treatmentof osteoarthritis of the knee: a system-atic review. Arthritis Rheum. 2008;59:1488–1494.

13 Shaw JM, Snow CM. Weighted vest exer-cise improves indices of fall risk in olderwomen. J Gerontol A Biol Sci Med Sci.1998;53:M53–M58.

14 Kim S, Lockhart T. Effects of 8 weeks ofbalance or weight training for the inde-pendently living elderly on the outcomesof induced slips. Int J Rehabil Res. 2010;33:49–55.

15 Handrigan G, Hue O, Simoneau M, et al.Weight loss and muscular strength affectstatic balance control. Int J Obes (Lond).2010;34:936–942.

16 Jan MH, Lin CH, Lin YF, et al. Effectsof weight-bearing versus nonweight-bearing exercise on function, walkingspeed, and position sense in participantswith knee osteoarthritis: a randomizedcontrolled trial. Arch Phys Med Rehabil.2009;90:897–904.

17 Messier SP, Legault C, Mihalko S, et al.The intensive diet and exercise for arthri-tis (IDEA) trial: design and rationale.BMC Musculoskelet Disord. 2009;10:93.

18 Zhang W, Moskowitz RW, Nuki G, et al.OARSI recommendations for the manage-ment of hip and knee osteoarthritis, partII: OARSI evidence-based, expert consen-sus guidelines. Osteoarthritis Cartilage.2008;16:137–162.

19 American College of Rheumatology Sub-committee on Osteoarthritis Guidelines.Recommendations for the medical man-agement of osteoarthritis of the hip andknee: 2000 update. Arthritis Rheum.2000;43:1905–1915.

20 Work group recommendations; 2002exercise and physical activity confer-ence, St Louis, Missouri; session V: evi-dence of benefit of exercise and physicalactivity in arthritis. Arthritis Rheum.2003;49:453–454.

21 Ottawa Panel evidence-based clinicalpractice guidelines for therapeutic exer-cises and manual therapy in the treat-ment of osteoarthritis. Phys Ther. 2005;85:907–971.

22 Hootman JM, Macera CA, Ham SA, et al.Physical activity levels among the generalUS adult population and in adults withand without arthritis. Arthritis Rheum.2003;49:129–135.

23 Guidelines for the Management of Painin Osteoarthritis, Rheumatoid Arthritis,and Juvenile Chronic Arthritis. Glen-view, IL: American Pain Society; 2002.Clinical Practice Guideline.

24 British Medical Journal Clinical Evi-dence: A Compendium of the Best Avail-able Evidence for Effective Health Care.9th ed. London, United Kingdom: BMJBooks; 2004.

25 Ontario Treatment Guidelines for Osteo-arthritis, Rheumatoid Arthritis, andAcute Musculoskeletal Injury. Toronto,Ontario, Canada: Ontario MusculoskeletalTherapeutics Review Panel; 2000.

26 Philadelphia Panel evidence-based clini-cal practice guidelines on selected reha-bilitation interventions for low backpain. Phys Ther. 2001;81:1641–1674.

27 Pencharz JN, Grigoriadis E, Jansz GF,Bombardier C. A critical appraisal of clin-ical practice guidelines for the treatmentof lower-limb osteoarthritis. ArthritisRes. 2002;4:36–44.

28 Roddy E, Zhang W, Doherty M. Aero-bic walking or strengthening exercisefor osteoarthritis of the knee: a system-atic review. Ann Rheum Dis. 2005;64:544–548.

29 Ottawa Panel Evidence-Based ClinicalPractice Guidelines for Aerobic FitnessExercises in the Management of Fibromy-algia: Part 1. Phys Ther. 2008;88:857–871.

30 Ottawa Panel Evidence-Based ClinicalPractice Guidelines for Therapeutic Exer-cises in the Management of RheumatoidArthritis in Adults. Phys Ther. 2004;84:934–981.

31 Ottawa Panel Evidence-Based ClinicalPractice Guidelines for Electrotherapyand Thermotherapy Interventions in theManagement of Rheumatoid Arthritis inAdults. Phys Ther. 2004;84:1016–1043.

32 Ottawa Panel; Khadilikar A, Phillips K,et al. Ottawa Panel evidence-based clini-cal practice guidelines for post-strokerehabilitation. Top Stroke Rehabil. 2006;13:1–269.

33 Ottawa Panel Evidence-Based ClinicalPractice Guidelines for StrengtheningExercises in the Management of Fibro-myalgia: Part 2. Phys Ther. 2008;88:873–886.

34 Jadad AR, Moore RA, Carroll D, et al.Assessing the quality of reports of ran-domized clinical trials: is blinding neces-sary? Control Clin Trials. 1996;17:1–12.




35 Christensen R, Astrup A, Bliddal H.Weight loss: the treatment of choice forknee osteoarthritis: a randomized trial.Osteoarthritis Cartilage. 2005;13:20–27.

36 Ghroubi S, Elleuch H, Kaffel N, et al. Con-tribution of exercise and diet in the man-agement of knee osteoarthritis in theobese [article in French]. Ann ReadaptMed Phys. 2008;51:663–670.

37 Messier SP, Loeser RF, Miller GD, et al.Exercise and dietary weight loss in over-weight and obese older adults with kneeosteoarthritis: the arthritis, diet, andactivity promotional trial. ArthritisRheum. 2004;50:1501–1510.

38 Rejeski WJ, Focht BC, Messier SP. Obese,older adults with knee osteoarthritis:weight loss, exercise, and quality of life.Health Psychol. 2002;21:419–426.

39 Messier SP, Loeser RF, Mitchell MN, et al.Exercise and weight loss in obese olderadults with knee osteoarthritis: a prelim-inary study. J Am Geriatr Soc. 2000;48:1062–1072.

40 Miller GD, Nicklas BJ, Davis C, et al.Intensive weight loss program improvesphysical function in older obese adultswith knee osteoarthritis. Obesity (SilverSpring). 2006;14:1219–1230.

41 Focht BC, Rejeski WJ, Ambrosius WT,et al. Exercise, self-efficacy, and mobilityperformance in overweight and obeseolder adults with knee osteoarthritis.Arthritis Rheum. 2005;53:659–665.

42 Huang MH, Chen CH, Chen TW, et al.The effects of weight reduction on therehabilitation of patients with kneeosteoarthritis and obesity. Arthritis CareRes. 2000;13:398–405.

43 Wang X, Miller GD, Messier SP, NicklasBJ. Knee strength maintained despite lossof lean body mass during weight loss inolder obese adults with knee osteoarthri-tis. J Gerontol A Biol Sci Med Sci. 2007;62:866–871.

44 Lim JY, Tchai E, Jang SN. Effectiveness ofaquatic exercise for obese patients withknee osteoarthritis: a randomized con-trolled trial. PMR. 2010l2:723–731.

45 Recommendations for the medical man-agement of osteoarthritis of the hip andknee; 2000 update: American College ofRheumatology Subcommittee on Osteo-arthritis guidelines. Arthritis Rheum.2000;43:1905–1915.

46 Browning RC, Kram R. Effects of obesityon the biomechanics of walking at differ-ent speeds. Med Sci Sports Exerc. 2007;39:1632–1641.

47 Davis MA, Ettinger WH, Neuhaus JM.Obesity and osteoarthritis of the knee:evidence from the National Health andNutrition Examination Survey (NHANES I).Semin Arthritis Rheum. 1990;20(3 suppl1):34–41.

48 DeVita P, Hortobagyi T. Obesity is notassociated with increased knee jointtorque and power during level walking.J Biomech. 2003;36:1355–1362.

49 Felson DT, Goggins J, Niu J, et al. Theeffect of body weight on progressionof knee osteoarthritis is dependent onalignment. Arthritis Rheum. 2004;50:3904–3909.

50 Focht BC, Ewing V, Gauvin L, Rejeski WJ.The unique and transient impact of acuteexercise on pain perception in olderoverweight or obese adults with kneeosteoarthritis. Ann Behav Med. 2002;24:201–210.

51 Focht BC, Gauvin L, Rejeski WJ. The con-tribution of daily experiences and acuteexercise to fluctuations in daily feelingstates among older, obese adults withknee osteoarthritis. J Behav Med. 2004;27:101–121.

52 Foran JR, Mont MA, Etienne G, et al. Theoutcome of total knee arthroplasty inobese patients. J Bone Joint Surg Am.2004;86-A:1609–1615.

53 Gillespie GN, Porteous AJ. Obesity andknee arthroplasty. Knee. 2007;14:81–86.

54 Holmberg S, Thelin A, Thelin N. Kneeosteoarthritis and body mass index: apopulation-based case-control study.Scand J Rheumatol. 2005;34:59–64.

55 Jadelis K, Miller ME, Ettinger WH Jr,Messier SP. Strength, balance, and themodifying effects of obesity and kneepain: results from the ObservationalArthritis Study in Seniors (OASIS). J AmGeriatr Soc. 2001;49:884–891.

56 Jinks C, Jordan K, Croft P. Disabling kneepain—another consequence of obesity:results from a prospective cohort study.BMC Public Health. 2006;6:258.

57 Messier SP. Osteoarthitis of the knee andassociated factors of age and obesity:effects on gait. Med Sci Sports Exerc.1994;26:1446–1452.

58 Perrot S, Poiraudeau S, Kabir M, et al.Active or passive pain coping strategiesin hip and knee osteoarthritis: results ofa national survey of 4,719 patients in aprimary care setting. Arthritis Rheum.2008;59:1555–1562.

59 Reijman M, Pols HA, Bergink AP, et al.Body mass index associated with onsetand progression of osteoarthritis of theknee but not of the hip: the Rotterdamstudy. Ann Rheum Dis. 2007;66:158–162.

60 Sutbeyaz ST, Sezer N, Koseoglu BF, et al.Influence of knee osteoarthritis on exer-cise capacity and quality of life in obeseadults. Obesity (Silver Spring). 2007;15:2071–2076.

61 Zhai G, Cicuttini F, Ding C, et al. Corre-lates of knee pain in younger subjects.Clin Rheumatol. 2007;26:75–80.

62 Nebel MB, Sims EL, Keefe FJ, et al. Therelationship of self-reported pain andfunctional impairment to gait mechanicsin overweight and obese persons withknee osteoarthritis. Arch Phys Med Reha-bil. 2009;90:1874–1879

63 Weight loss for knee arthritis. Bandolier.2005;133:4–5.

64 Bennell KL, Hunt MA, Wrigley TV, et al.Role of muscle in the genesis and man-agement of knee osteoarthritis. RheumDis Clin North Am. 2008;34:731–754.

65 Christensen R, Bartels EM, Astrup A,Bliddal H. Effect of weight reduction inobese patients diagnosed with kneeosteoarthritis: a systematic review andmeta-analysis. Ann Rheum Dis. 2007;66:433–439.

66 Ciliska, D. Review: moderate weight lossreduces functional disability but does notreduce pain in obese patients with kneeosteoarthritis. Evid Based Nurs. 2008;11:17.

67 Davis MA, Ettinger WH, Neuhaus JM,Hauck WW. Sex differences in osteo-arthritis of the knee: the role of obesity.Am J Epidemiol. 1988;127:1019–1030.

68 DeHaan MN, Guzman J, Bayley MT, BellMJ. Knee osteoarthritis clinical practiceguidelines: how are we doing? J Rheu-matol. 2007;34:2099–2105.

69 Felson DT, Anderson JJ, Naimark A, et al.Obesity and knee osteoarthritis: the Fra-mingham study. Ann Intern Med. 1988;109:18–24.

70 Fransen M. Dietary weight loss and exer-cise for obese adults with knee osteo-arthritis: modest weight loss targets,mild exercise, modest effects. ArthritisRheum. 2004;50:1366–1369.

71 Felson DT, Lawrence RC, Dieppe PA,et al. Osteoarthritis, new insights; part 1:the disease and its risk factors. AnnIntern Med. 2000;133:635–646.

72 Messier SP. Obesity and osteoarthritis:disease genesis and nonpharmacologicweight management. Rheum Dis ClinNorth Am. 2008;34:713–729.

73 Sturmer T, Gunther KP, Brenner H. Obe-sity, overweight and patterns of osteo-arthritis: the Ulm Osteoarthritis Study.J Clin Epidemiol. 2000;53:307–313.

74 Zhang Y, Jordan JM. Epidemiology ofosteoarthritis. Rheum Dis Clin NorthAm. 2008;34:515–529.

75 Chua SD Jr, Messier SP, Legault C, et al.Effect of an exercise and dietary interven-tion on serum biomarkers in overweightand obese adults with osteoarthritis ofthe knee. Osteoarthritis Cartilage. 2008;16:1047–1053.

76 Miller GD, Nicklas BJ, Davis CC, et al. Isserum leptin related to physical functionand is it modifiable through weight lossand exercise in older adults with kneeosteoarthritis? Int J Obes Relat MetabDisord. 2004;28:1383–1390.

77 Nicklas BJ, Ambrosius W, Messier SP,et al. Diet-induced weight loss, exercise,and chronic inflammation in older, obeseadults: a randomized controlled clinicaltrial. Am J Clin Nutr. 2004;79:544–551.

78 Sevick MA, Miller GD, Loeser RF, et al.Cost-effectiveness of exercise and diet inoverweight and obese adults with kneeosteoarthritis. Med Sci Sports & Exerc.2009;41:1167–1174.

79 Coggon D, Reading I, Croft P, et al. Kneeosteoarthritis and obesity. Int J ObesRelat Metab Disord. 2001;25:622–627.

80 Felson DT, Zhang Y, Anthony JM, et al.Weight loss reduces the risk for symp-tomatic knee osteoarthritis in women:the Framingham study. Ann Intern Med.1992;116:535–539.




81 McGoey BV, Deitel M, Saplys RJ, KlimanME. Effect of weight loss on musculosk-eletal pain in the morbidly obese. J BoneJoint Surg Br. 1990;72:322–323.

82 Messier SP, Gutekunst DJ, Davis C,DeVita P. Weight loss reduces knee-jointloads in overweight and obese adultswith knee osteoarthritis. Arthritis Rheum.2005;52:2026–2032.

83 Miller GD, Rejeski WJ, Williamson JD,et al. The Arthritis, Diet and Activity Pro-motion Trial (ADAPT): design, rationale,and baseline results. Control Clin Trials.2003;24:462–480.

84 Jenkinson CM, Doherty M, Avery AJ,et al. Effects of dietary intervention andquadriceps strengthening exercises onpain and function in overweight peoplewith knee pain: randomised controlledtrial [erratum in: BMJ. 2010;340:c2088].BMJ. 2009;339:b3170.

85 Witham MD, Avenell A. Interventions toachieve long-term weight loss in obeseolder people: a systematic review and meta-analysis. Age Ageing. 2010;39:176–184.

86 Marks R. Obesity profiles with kneeosteoarthritis: correlation with pain, dis-ability, disease progression. Obesity (Sil-ver Spring). 2007;15:1867–1874.

87 Peterson MG, Kovar-Toledano PA, OtisJC, et al. Effect of a walking program ongait characteristics in patients with osteo-arthritis. Arthritis Care Res. 1993;6:11–16.

88 Sköldstam L, Brudin L, Hagfors L, Johans-son G. Weight reduction is not a majorreason for improvement in rheumatoidarthritis from lacto-vegetarian, vegan orMediterranean diets. Nutr J. 2005;4:15.

89 Miller ER III, Erlinger TP, Young DR,et al. Results of the Diet, Exercise, andWeight Loss Intervention Trial (DEW-IT).Hypertension. 2002;40:612–618.

90 Foy CG, Lewis CE, Hairston KG, et al.Intensive lifestyle intervention improvesphysical function among obese adultswith knee pain: findings from the LookAHEAD trial [erratum in: Obesity (SilverSpring). 2011;19:233]. Obesity (SilverSpring). 2011;19:83–93.

91 van Gool CH, Penninx BW, Kempen GI,et al. Effects of exercise adherence onphysical function among overweightolder adults with knee osteoarthritis.Arthritis Rheum. 2005;53:24–32.

92 van Gool CH, Penninx BW, Kempen GI,et al. Determinants of high and low atten-dance to diet and exercise interventionsamong overweight and obese olderadults: results from the arthritis, diet, andactivity promotion trial. Contemp ClinTrials. 2006;27:227–237.

93 Toda Y. A comparison of the efficacy ofconservative therapies for obese patientswith osteoarthritis of the knee [article inJapanese]. Ryumachi. 2002;42:795–800.

94 Toda Y, Toda T, Takemura S, et al.Change in body fat, but not body weightor metabolic correlates of obesity, isrelated to symptomatic relief of obesepatients with knee osteoarthritis after aweight control program. J Rheumatol.1998;25:2181–2186.

95 O’Neil PM, Rieder S. Utility and validityof the eating behaviour inventory in clin-ical obesity research: a review of the lit-erature. Obes Rev. 2005;6:209–216.

96 Thorp LE, Wimmer MA, Foucher KC,et al. The biomechanical effects offocused muscle training on medial kneeloads in OA of the knee: a pilot, proof ofconcept study. J Musculoskelet Neuro-nal Interact. 2010;10;166–173.

97 Shaw BS, Shaw I, Brown GA. Comparisonof resistance and concurrent resistanceand endurance training regimes in thedevelopment of strength. J StrengthCond Res. 2009;23:2507–2514.

98 Kraemer WJ, Volek JS, Clark KL, et al.Influence of exercise training on physio-logical and performance changes withweight loss in men. Med Sci Sports Exerc.1999;31:1320–1329.

99 Mikesky AE, Mazzuca SA, Brandt KD,et al. Effects of strength training on theincidence and progression of knee osteo-arthritis. Arthritis Rheum. 2006;55;690–699.

100 McGinty G, Irrgang JJ, Pezzullo D. Bio-mechanical considerations for rehabilita-tion of the knee. Clin Biomech (Bristol,Avon). 2000;15;160–166.

101 Sharma L, Chang A. Overweight: advanc-ing our understanding of its impact onthe knee and the hip. Ann Rheum Dis.2007;66:141–142.

102 Wadden TA. Butryn ML. Behavioral treat-ment of obesity. Endocrinol Metab ClinNorth Am. 2003;32:981–1003.

103 Lang A, Froelicher ES. Management ofoverweight and obesity in adults: behav-ioral intervention for long-term weightloss and maintenance. Eur J CardiovascNurs. 2006;5:102–114.

104 Anderson JW, Konz EC, Frederich RC,Wood CL. Long-term weight-loss mainte-nance: a meta-analysis of US studies. Am JClin Nutr. 2001;74:579–584.

105 Kalunian K. UpToDate Online 18.2:Diagnosis and classification of osteoar-thritis. Waltham, MA: Wellesley; 2007:1.Available at: http://www.uptodate.com/contents/diagnosis-and-classification-of-osteoarthritis.

106 Olivo SA, Macedo LG, Gadotti IC, et al.Scales to assess the quality of randomizedcontrolled trials: a systematic review.Phys Ther. 2008;88:156–175.

107 Iversen MD, Petersson IF. Design issuesand priorities in team and nonpharmaco-logical arthritis care research. J Rheuma-tol. 2006;33:1904–1907.

108 Sowers MF, Yosef M, Yamada D, et al.BMI vs. body composition and radio-graphically defined osteoarthritis of theknee in women: a 4-year follow-upstudy. Osteoarthritis Cartilage. 2008;16:367–372.

109 Toda Y. The effect of energy restriction,walking, and exercise on lower extrem-ity lean body mass in obese women withosteoarthritis of the knee. J Orthop Sci.2001;6:148–154.

110 Pelland L, Brosseau L, Wells G, et al.Efficacy of strengthening exercise forosteoarthritis; part 1: a meta-analysis.Phys Ther Rev. 2004;9:77–108.

111 Brosseau L, Pelland L, Wells G, et al. Effi-cacy of aerobic exercises for osteoarthri-tis; part 2: a meta-analysis. Phys Ther Rev.2004;9:125–145.



http://www.uptodate.com/contents/diagnosis-and-classification-of-osteoarthritishttp://www.uptodate.com/contents/diagnosis-and-classification-of-osteoarthritishttp://www.uptodate.com/contents/diagnosis-and-classification-of-osteoarthritishttp://ptjournal.apta.org/

Appendix.Ottawa Panel Recommendationsa

1. Physical activity (aerobic training and strength training) versus control, level I (1 RCT, N�56, low quality).36

Grade A for pain relief (VAS score), functional status (WOMAC score), and strength (quadriceps and hamstring musclestrength) at end of treatment (8 weeks) (clinically important benefit demonstrated). Grade C� for body composition (waistcircumference) at end of treatment (8 weeks) (clinically important benefit demonstrated without statistical significance).Grade C for functional status (Lequesne Index), walking endurance (6MWT), and body composition (body weight, BMI[kg/m2], and lean body mass [kg]) at end of treatment (8 weeks) (no benefit demonstrated). Grade D for body composition(lean body mass [%]) at end of treatment (8 weeks) (no benefit demonstrated but favoring control).

2. Physical activity (aerobic training and strength training) and diet (caloric restriction of 25%–30%, 1,500–3,000 cal/d) versus control, level I (1 RCT, N�56, low quality).36 Grade A for pain relief (VAS score), functional status(WOMAC score, Lequesne Index), body composition (waist circumference), and strength (quadriceps and hamstring musclestrength) at end of treatment (8 weeks) (clinically important benefit demonstrated). Grade C for body composition (leanbody mass, fat body mass, body weight [kg], and BMI [kg/m2]) and walking endurance (6MWT) at end of treatment (8 weeks)(no benefit demonstrated).

3. Diet (caloric restriction of 25%–30%, 1,500–3,000 cal/d) versus control, level I (1 RCT, N�56, low quality).36

Grade A for pain relief (VAS score) at end of treatment (8 weeks) (clinically important benefit demonstrated). Grade C forfunctional status (WOMAC score, Lequesne Index), walking endurance (6MWT), body composition (body weight, BMI[kg/m2], waist circumference, fat body mass, and lean body mass), and strength (quadriceps muscle strength) at end oftreatment (8 weeks) (no benefit demonstrated). Grade D for strength (hamstring muscle strength) at end of treatment (8weeks) (no benefit demonstrated but favoring control).

4. Physical activity (aerobic training and strength training) versus diet (caloric restriction of 25%–30%, 1,500–3,000 cal/d), level I (1 RCT, N�56, low quality).36 Grade C� for improved strength (quadriceps and hamstring musclestrength) at end of treatment (8 weeks) (clinically important benefit demonstrated without statistical significance). Grade Cfor pain relief (VAS score), functional status (WOMAC score, Lequesne Index), walking endurance (6MWT), and bodycomposition (body weight, BMI [kg/m2], lean body mass, waist circumference, and fat body mass) at end of treatment(8 weeks) (no benefit demonstrated).

5. Physical activity (aerobic training and strength training) versus physical activity (aerobic training and strengthtraining) and diet (caloric restriction of 25%–30%, 1,500–3,000 cal/d), level I (1 RCT, N�56, low quality).36

Grade A for functional status favoring diet and physical activity (Lequesne Index) at end of treatment (8 weeks) (clinicallyimportant benefit demonstrated). Grade C� for pain relief favoring diet and physical activity (VAS score) at end of treatment(8 weeks) (clinically important benefit demonstrated without statistical significance). Grade C for functional status (WOMACscore), walking endurance (6MWT), body composition (body weight, BMI [kg/m2], waist circumference, lean body mass, fatbody mass), and strength (quadriceps and hamstring muscle strength) at end of treatment (8 weeks) (no benefit demon-strated).

6. Diet (caloric restriction of 25%–30%, 1,500–3,000 cal/d) versus physical activity (aerobic training and strengthtraining) and diet (caloric restriction of 25%–30%, 1,500–3,000 cal/d), level I (1 RCT, N�56, low quality).36

Grade A for pain relief favoring diet and physical activity (VAS score), functional status favoring diet and physical activity(WOMAC score), and functional status favoring diet and physical activity (Lequesne Index) at end of treatment (8 weeks)(clinically important benefit demonstrated). Grade C� for strength favoring diet and strength (quadriceps and hamstringmuscle strength) at end of treatment (8 weeks) (clinically important benefit demonstrated without statistical significance).Grade C for walking endurance (6MWT) and body composition (body weight, BMI [kg/m2], waist circumference, lean bodymass, and fat body mass) at end of treatment (8 weeks) (no benefit demonstrated).

7. Intensive physical activity (aerobic training, strength training ) and intensive diet (deficit of 1,000 kcal/d, 20%protein, 25% fat, 55% carbohydrate diet) versus control, level I, (2 RCTs, N�174, low quality).40,43 Grade A forfunctional status (WOMAC sum score),40 pain relief (WOMAC pain score),40 physical function (WOMAC function score),40

mobility (stair climbing time),40 and torque (concentric knee extensors) at end of treatment (6 months)43 (clinicallyimportant benefit demonstrated). Grade C� for force (concentric knee extension)43 and torque (eccentric knee extension/lean mass)43 at end of treatment (6 months) (clinically important benefit demonstrated without statistical significance).Grade C for walking endurance (6MWT),40 force (eccentric knee extension),43 and body composition (body weight, BMI[kg/m2], waist circumference, and fat body mass)40 at end of treatment (6 months) (no benefit demonstrated). Grade D forstiffness (WOMAC stiffness score)40 and body composition (lean body mass)40 at end of treatment (6 months) (no benefitdemonstrated, but results favored the control group).

(Continued)




Appendix.Continued

8. Diet (behavior modification, lower caloric intake based on group dynamics literature and social-cognitivetheory, average weight loss of 5%) versus control, level I (1 RCT, N�316, low quality).38 Grade C for quality of life(SF-36 mental health and physical function scores) at average between the 6-month follow-up and the 18-month assessmentat end of treatment (no benefit demonstrated). Grade D� for psychological well-being (physical function and bodysatisfaction measure) at average between the 6-month follow-up and the 18-month assessment at end of treatment (clinicallyimportant benefit favoring control demonstrated with statistical significance).

9. Physical activity (aerobic [50%–75% HRR] and resistance training) versus diet (behavior modification, lowercaloric intake based on group dynamics literature and social-cognitive theory, average weight loss of 5%) andphysical activity (aerobic [50%–75% HRR] and resistance training), level I (1 RCT, N�316, low quality).38 Grade Afor psychological well-being (physical function and body satisfaction measure) at average between the 6-month follow-up andthe 18-month assessment at end of treatment (clinically important benefit demonstrated favoring physical activity). Grade Cfor quality of life (SF-36 mental health and physical function scores) at average between the 6-month follow-up and the18-month assessment at end of treatment (no benefit demonstrated).

10. Diet (behavior modification, lower caloric intake based on group dynamics literature and social-cognitivetheory, average weight loss of 5%) and physical activity (aerobic [50%–75% HRR] and resistance training) versuscontrol, level 1 (1 RCT, N�316, low quality).38 Grade C for quality of life (SF-36 physical function score) at averagebetween the 6-month follow-up and the 18-month assessment at end of treatment (no benefit demonstrated). Grade D forquality of life (SF-36 mental health score) at average between the 6-month follow-up and the 18-month assessment at end oftreatment (no benefit demonstrated, but results favored the control group). Grade D� for psychological well-being (physicalfunction and body satisfaction measure) at average between the 6-month follow-up and the 18-month assessment at end oftreatment (clinically important benefit favoring control demonstrated with statistical significance).

11. Diet (behavior modification, lower caloric intake based on group dynamics literature and social-cognitivetheory, average weight loss of 5%) versus physical activity (aerobic [50%–75% HRR] and resistance training),level I (1 RCT, N�316, low quality).38 Grade C� for psychological well-being (body satisfaction measure–physicalfunction) at average between the 6-month follow-up and the 18-month assessment at end of treatment (clinically importantbenefit demonstrated without statistical significance favoring diet only). Grade C� for psychological well-being (bodysatisfaction measure–appearance) at average between the 6-month follow-up and the 18-month assessment at end oftreatment (clinically important benefit demonstrated without statistical significance favoring physical activity). Grade C for qualityof life (SF-36 mental health and physical function scores) at average between the 6-month follow-up and the 18-month assessmentat end of treatment (no benefit demonstrated).

12. Diet (behavior modification, lower caloric intake based on group dynamics literature and social-cognitivetheory, average weight loss of 5%) versus diet (aerobic [50%–75% HRR] and resistance training) and physicalactivity (aerobic [50%–75% HRR] and resistance training), level I (1 RCT, N�316, low quality).38 Grade Afor improved psychological well-being (body satisfaction measure–physical function) at average between the 6-monthfollow-up and the 18-month assessment at end of treatment (clinically important benefit demonstrated favoring diet).Grade C for psychological well-being (body satisfaction measure–appearance) and quality of life (SF-36 physical func-tion score) at average between the 6-month follow-up and the 18-month assessment at end of treatment (no benefitdemonstrated).

13. Physical activity (aerobic [50%–75% HRR] and resistance training) versus diet (behavior modification, lowercaloric intake based on group dynamics literature and social-cognitive theory, average weight loss of 5%) andphysical activity (aerobic [50%–75% HRR] and resistance training), level I (1 RCT, N�316, low quality).38 Grade Afor psychological well-being (physical function and body satisfaction measure) at average between the 6-month follow-up andthe 18-month assessment at end of treatment (clinically important benefit demonstrated favoring physical activity). Grade Cfor mental status (SF-36 mental health and physical function scores) at average between the 6-month follow-up and the18-month assessment at end of treatment (no benefit demonstrated).

(Continued)




Appendix.Continued

14. Diet (behavior modification, lower caloric intake based on group dynamics literature and social-cognitivetheory, average weight loss of 5%) and physical activity (aerobic [50%–75% HRR] and resistance training) versuscontrol, level I (1 RCT, N�316, low quality).38 Grade C for quality of life (SF-36 physical function score) at averagebetween the 6-month follow-up and the 18-month assessment at end of treatment (no benefit demonstrated). Grade D forquality of life (SF-36 mental health score) at average between the 6-month follow-up and the 18-month assessment at end oftreatment (no benefit demonstrated but favoring control). Grade D� for psychological well-being (physical function andbody satisfaction measure–appearance) at average between the 6-month follow-up and the 18-month assessment at end oftreatment (clinically important benefit favoring control demonstrated with statistical significance).

15. Diet (behavior modification, lower caloric intake based on group dynamics literature and social-cognitivetheory, average weight loss of 5%) versus physical activity (aerobic [50%–75% HRR] and resistance training),level I (1 RCT, N�316, low quality).38 Grade C� for psychological well-being (body satisfaction measure–physicalfunction) at average between the 6-month follow-up and the 18-month assessment at end of treatment (clinically importantbenefit demonstrated without statistical significance favoring diet). Grade C� for psychological well-being (body satisfactionmeasure–appearance) at average between the 6-month follow-up and the 18-month assessment at end of treatment (clinicallyimportant benefit demonstrated without statistical significance favoring physical activity). Grade C for quality of life (SF-36mental health and physical function scores) at average between the 6-month follow-up and the 18-month assessment at endof treatment (no benefit demonstrated).

16. Diet (behavior modification, lower caloric intake based on group dynamics literature and social-cognitivetheory, average weight loss of 5%) versus diet (behavior modification, lower caloric intake based on groupdynamics literature and social-cognitive theory, average weight loss of 5%) and physical activity (aerobic[50%–75% HRR] and resistance training), level I (1 RCT, N�316, low quality).38 Grade A for psychological well-being(body satisfaction measure–physical function) at average between the 6-month follow-up and the 18-month assessment at endof treatment (clinically important benefit demonstrated favoring diet). Grade C for psychological well-being (body satisfac-tion measure–appearance) and quality of life (SF-36 mental health and physical function scores) at average between the6-month follow-up and the 18-month assessment at end of treatment (no benefit demonstrated).

17. Low energy diet versus conventional high protein diet (control), level I (1 RCT, N�96, high quality).35 Grade Afor functional status (WOMAC total score) and physical function (WOMAC function score) at end of treatment (8 weeks)(clinically important benefit demonstrated). Grade C� for pain relief (WOMAC pain score) and stiffness (WOMAC stiffnessscore) at end of treatment (8 weeks) (clinically important benefit demonstrated without statistical significance). Grade C forbody composition (body weight [kg], fat body mass [kg], fat body mass [%]) and functional status (Lequesne index score[0–26]) at end of treatment (8 weeks) (no benefit demonstrated). Grade D for body composition (lean body mass [kg]) atend of treatment (8 weeks) (no benefit demonstrated but favoring control).

18. Physical activity (strength training and aerobic training) and diet (patient education and cognitive-behavioralmodification strategies) versus physical activity (strength training and aerobic training), level I (3 RCTs,N�656)37,39,41 (1 RCT37 is high quality, 2 RCTs39,41 are low quality). Grade C� for pain relief (transfer pain frequencyand transfer pain intensity)39 at end of treatment (3 months), pain relief (WOMAC pain score)37 at end of treatment (6 and18 months), and self-efficacy on stairs (stair climb)41 at end of treatment (18 months) (clinically important benefitdemonstrated without statistical significance favoring physical activity and diet). Grade C for pain relief (ambulation painfrequency)39 at end of treatment (3 months and 6 months), pain relief (transfer pain intensity and transfer pain frequency)39

at end of treatment (6 months), walking endurance (6MWT)37,39 at end of treatment (3 months,39 6 months,37,39 and18 months37), mobility (stair climbing)37,39 at end of treatment (3 months,39 6 months,37,39 and 18 months37), functionalstatus (transfer self-reported function) at end of treatment (3 months),39 functional status (ambulation self-reported function,transfer self-reported function, summary self-reported function) at end of treatment (6 months)39 and self-efficacy in walking(6MWT) at end of treatment (18 months)41 (no benefit demonstrated). Grade D for pain relief (ambulation pain intensity)39

and functional status (ambulation self-reported function, summary self-reported function) at end of treatment (3 months)39

(no benefit demonstrated, but results favored physical activity). Grade D� for pain relief (ambulation pain intensity) atend of treatment (6 months)39 (clinically important benefit demonstrated favoring physical activity without statisticalsignificance).

(Continued)




Appendix.Continued

19. Physical activity (strength training and aerobic training) and diet (behavior modification, lower caloric intakebased on group dynamics literature and social-cognitive theory, average weight loss of 5%) versus control, levelI (2 RCTs, N�632)37,41 (1 RCT37 is high quality, 1 RCT41 is low quality). Grade A for walking endurance (6MWT)37 atend of treatment (6 months and 18 months) and self-efficacy (stair climbing and 6MWT)41 at end of treatment (18 months)(clinically important benefit demonstrated). Grade C� for mobility (stair climbing)37 at end of treatment (6 months and18 months) (clinically important benefit demonstrated without statistical significance). Grade C for pain relief (WOMAC painscore)37 at end of treatment (6 months and 18 months) (no benefit demonstrated).

20. Physical activity (strength training and aerobic training) and diet (behavior modification, lower caloric intakebased on group dynamics literature and social-cognitive theory, average weight loss of 5%) versus diet (behaviormodification, lower caloric intake based on group dynamics literature and social-cognitive theory, averageweight loss of 5%), level I (2 RCTs, N�632)37,41 (1 RCT37 is high quality, 1 RCT41 is low quality). Grade A forself-efficacy in walking (6MWT)41 at end of treatment (18 months) (clinically important benefit demonstrated favoringphysical activity and diet). Grade C� for mobility (stair climbing

by Moshe Aharoni on June 8, 2014N. Bugnariu, PhD, Physical Therapy Department, University of North...

Documents

Transcript of by Moshe Aharoni on June 8, 2014N. Bugnariu, PhD, Physical Therapy Department, University of North...