UBC Faculty Research and Publications

A physiotherapist-delivered integrated exercise and pain coping skills training intervention for individuals… Bennell, Kim L; Ahamed, Yasmin; Bryant, Christina; Jull, Gwendolen; Hunt, Michael A; Kenardy, Justin; Forbes, Andrew; Harris, Anthony; Nicholas, Michael; Metcalf, Ben; Egerton, Thorlene; Keefe, Francis J Jul 24, 2012

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12891_2012_Article_1531.pdf [ 376.64kB ]
JSON: 52383-1.0223786.json
JSON-LD: 52383-1.0223786-ld.json
RDF/XML (Pretty): 52383-1.0223786-rdf.xml
RDF/JSON: 52383-1.0223786-rdf.json
Turtle: 52383-1.0223786-turtle.txt
N-Triples: 52383-1.0223786-rdf-ntriples.txt
Original Record: 52383-1.0223786-source.json
Full Text

Full Text

STUDY PROTOCOL Open AccessA physiotherapist-delivered integrated exerciseand pain coping skills training intervention forindividuals with knee osteoarthritis: a randomisedcontrolled trial protocolKim L Bennell1*, Yasmin Ahamed1, Christina Bryant2, Gwendolen Jull3, Michael A Hunt4, Justin Kenardy5,Andrew Forbes6, Anthony Harris7, Michael Nicholas8, Ben Metcalf1, Thorlene Egerton1 and Francis J Keefe9AbstractBackground: Knee osteoarthritis (OA) is a prevalent chronic musculoskeletal condition with no cure. Pain is theprimary symptom and results from a complex interaction between structural changes, physical impairments andpsychological factors. Much evidence supports the use of strengthening exercises to improve pain and physicalfunction in this patient population. There is also a growing body of research examining the effects of psychologist-delivered pain coping skills training (PCST) particularly in other chronic pain conditions. Though typically providedseparately, there are symptom, resource and personnel advantages of exercise and PCST being delivered togetherby a single healthcare professional. Physiotherapists are a logical choice to be trained to deliver a PCST interventionas they already have expertise in administering exercise for knee OA and are cognisant of the need for abiopsychosocial approach to management. No studies to date have examined the effects of an integrated exerciseand PCST program delivered solely by physiotherapists in this population. The primary aim of this multisiterandomised controlled trial is to investigate whether an integrated 12-week PCST and exercise treatment programdelivered by physiotherapists is more efficacious than either program alone in treating pain and physical function inindividuals with knee OA.Methods/design: This will be an assessor-blinded, 3-arm randomised controlled trial of a 12-week interventioninvolving 10 physiotherapy visits together with home practice. Participants with symptomatic and radiographicknee OA will be recruited from the community in two cities in Australia and randomized into one of three groups:exercise alone, PCST alone, or integrated PCST and exercise. Randomisation will be stratified by city (Melbourne orBrisbane) and gender. Primary outcomes are overall average pain in the past week measured by a Visual AnalogueScale and physical function measured by the Western Ontario and McMaster Universities Osteoarthritis Indexsubscale. Secondary outcomes include global rating of change, muscle strength, functional performance, physicalactivity levels, health related quality of life and psychological factors. Measurements will be taken at baseline andimmediately following the intervention (12 weeks) as well as at 32 weeks and 52 weeks to examine maintenance ofany intervention effects. Specific assessment of adherence to the treatment program will also be made at weeks 22and 42. Relative cost-effectiveness will be determined from health service usage and outcome data.Discussion: The findings from this randomised controlled trial will provide evidence for the efficacy of anintegrated PCST and exercise program delivered by physiotherapists in the management of painful and functionallylimiting knee OA compared to either program alone.Trial registration: Australian New Zealand Clinical Trials Registry reference number: ACTRN12610000533099* Correspondence: k.bennell@unimelb.edu.au1Centre for Health, Exercise & Sports Medicine, Department of Physiotherapy,University of Melbourne, Melbourne, VIC, AustraliaFull list of author information is available at the end of the article© 2012 Bennell et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the CreativeCommons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, andreproduction in any medium, provided the original work is properly cited.Bennell et al. BMC Musculoskeletal Disorders 2012, 13:129http://www.biomedcentral.com/1471-2474/13/129BackgroundKnee osteoarthritis (OA) is a prevalent chronic musculo-skeletal condition [1] associated with pain, physical andpsychological dysfunction, and reduced quality of life inaffected individuals [2,3]. In addition to the personalburden of knee OA, there are substantial direct and in-direct health care costs making knee OA a major publichealth problem [4]. Given the extent of the problem andthe fact that the prevalence of OA will escalate with theageing population and increases in obesity rates [5], ef-fective treatment strategies are required. In particularstrategies that promote long-term self-management areimportant given the chronicity of the condition.Pain is the primary symptom of knee OA and peoplewith higher levels of pain have lower levels of physicalfunction, greater functional decline and reduced qualityof life [6]. Individuals with painful knee OA experiencedifficulty performing basic daily activities such as shop-ping, performing household chores, stair climbing aswell as engaging in social and outdoor activities [7]. Fur-thermore, knee pain related to OA is one of the stron-gest predictors of employment status and productivity[8]. Reducing pain is therefore a relevant and importanttreatment aim for this patient group.The experience of pain is influenced by a multitude ofstructural, physical, and psychosocial factors [6,9,10].Whilst stimulation of nociceptors in the capsule, sub-chondral bone, ligaments and other joint tissues contrib-ute to the perception of pain, structural damage in kneeOA is in fact not well correlated with pain severity [11].Instead, physical and psychological impairments that arecommonly found in this patient population are more im-portant predictors of pain. Muscle weakness, particularlyof the quadriceps, is associated with higher levels of painand greater declines in physical dysfunction [10,12]. Psy-chological impairments including pain catastrophising[13], poor pain coping strategies [14], anxiety [15], de-pression or depressed mood [15,16] and social isolation[16] are also related to increased pain levels in peoplewith knee OA. Furthermore, bi-directional relationshipsexist between pain and physical and psychologicalimpairments whereby pain can influence, and in turn beinfluenced by these factors [17], leading to a downwardcascade in physical and mental functioning. Given theimportance of both physical and psychological impair-ments, it would seem logical that treatments should ad-dress both aspects in order to maximise patientoutcomes.Clinical guidelines advocate conservative non-drugstrategies for the treatment of knee OA [18,19]. In par-ticular, exercise including muscle strengthening isrecommended [20] and is supported by considerable re-search evidence [21,22]. A recent systematic reviewincluded 18 randomised controlled trials, the majority ofwhich involved home-based programs of quadriceps orlower limb muscle strengthening. The results showedsignificant improvements in pain and self-reported phys-ical function with muscle strengthening exercise [22].However, despite consistent findings of short-termimprovements with exercise, reduced adherence to exer-cise programs limits long-term effectiveness [21,23].Thus, interventions that facilitate long-term exercise ad-herence are needed.Whether strengthening exercise also improves psycho-logical parameters in people with knee OA is less clearas few studies include adequate assessment of such para-meters [24-26]. There is some evidence from rando-mised controlled trials that strengthening exercise isassociated with reductions in depressive symptoms [27]in those with knee OA [28] and in other chronic condi-tions [29,30]. Such exercise has also been shown to im-prove self-efficacy, fatigue symptoms, and sleep qualityin depressed older adults [31]. Based on this limited evi-dence, it appears that strengthening exercise may im-prove some aspects of psychological functioning in thosewith knee OA but further research is needed.As psychological factors are related to pain, psycho-logical interventions are worthy of attention. Of the psy-chological interventions that have been considered, painmanagement, based on the principles of cognitive behav-ioural therapy, is the most extensively researched inter-vention in chronic pain conditions and has the strongestevidence base. One such approach, Pain Coping SkillsTraining (PCST) focuses on self-management and is wellrecognized as an effective cognitive behavioural treat-ment for disease-related pain conditions [29,30,32,33].However, a meta-analysis identified only two clinicaltrials involving PCST in those with knee OA. Both trialsshowed improvements in pain and physical functionover a 12 week intervention period [24,34,35] with bene-fits appearing to diminish over time. Given the potentialbenefits of PCST and yet the limited evidence in peoplewith knee OA, further research is needed to investigateits efficacy.In accordance with a biopsychosocial approach to themanagement of chronic pain [36] both physical and psy-chological impairments should be addressed in peoplewith knee OA. The studies that have tested integratedexercise and psychological treatments in a variety ofother chronic conditions including cancer [32,33], lowback pain [29] and fibromyalgia [30] have shown positiveresults. Only one study has examined the effects of acombined program of PCST and exercise on pain inthose with knee OA [24]. This study compared a 12 weekintervention of spouse-assisted PCST alone, exercisetraining alone, spouse-assisted PCST and exercise train-ing or standard care in 72 participants with knee OA[24]. The results showed that the combined programBennell et al. BMC Musculoskeletal Disorders 2012, 13:129 Page 2 of 17http://www.biomedcentral.com/1471-2474/13/129improved physical fitness, strength, pain coping and self-efficacy. In addition, those with improvements in self-efficacy were more likely to improve in psychologicalfunctioning. However, the combined intervention requiredparticipants to attend a total of four hours per week oftherapy delivered by two different health professionals,in addition to home practice. From a practical perspec-tive this requires a considerable time commitment frompatients and involves substantial treatment costs thatwould not necessarily be sustainable in everyday practice.Research investigating other methods of delivering com-bined treatments is required.PCST is generally delivered by a psychologist specialis-ing in pain management. However, it may be beneficialto utilise a single health care professional such as aphysiotherapist to deliver an integrated physical and psy-chological intervention [37,38]. Potential advantages ofusing a single therapist include better integration of theintervention components, increased availability of PCSTtreatment to those who may not have access to a psych-ologist, reduced time and cost for patients and reducedoverall costs to the health care system. Although phy-siotherapists do not have formal training in pain man-agement they are a logical choice to be trained in thedelivery of PCST given their expertise in administeringphysical treatments to treat pain and their understand-ing of the biopsychosocial approach. No studies to datehave examined the effects of an integrated exercise andformal PCST program delivered by physiotherapists inthis patient population.This project primarily aims to compare the efficacy ofa 12-week integrated PCST and exercise program deliv-ered by physiotherapists to treat pain and physical func-tion in individuals with knee OA compared to PCST orexercise programs alone. The secondary aims of thestudy are to compare the efficacy of these programs onfunctional performance, psychological parameters, qual-ity of life, muscle strength, physical activity levels andcost-effectiveness and to examine longer-term outcomesover a 9-month follow up period.Primary hypothesisH1: A 12-week integrated intervention of exercise andPCST will be more efficacious in improving pain andself-reported physical function than a 12-week interven-tion of either PCST or exercise alone immediately fol-lowing the intervention.Secondary hypothesesH2: An integrated intervention of exercise and PCSTwill be more efficacious in improving pain and self-reported physical function than either PCST or exercisealone at 32 weeks and 52 weeks.H3: An integrated intervention will be more effica-cious in improving psychological function, functionalperformance, quality of life, physical activity levels andperceived response to treatment than either PCST or ex-ercise alone immediately following the intervention andat 32 weeks and 52 weeks.H4: Exercise will lead to greater improvements inmuscle strength than PCST; PCST will lead to greaterimprovements in psychological parameters than exercise;and an integrated intervention will lead to greaterimprovements in both strength and psychological para-meters at measured time points.H5: Adherence to exercise during the 9-month un-supervised follow-up period will be greater with an inte-grated intervention than with an intervention of exercisealone.H6: An integrated intervention will be more cost-ef-fective than an intervention of exercise or PCST alonewhen costs are compared and related to the effects ofthe intervention at 52 weeks.H7: Specific patient baseline characteristics will mod-erate or predict treatment effects while pre- to post-treatment changes in the targeted cognitions, behavioursand physical impairments will mediate the effects ofPCST and exercise on subsequent patient pain anddisability.Methods/designTrial designThis will be an assessor-blinded, 3-arm randomisedcontrolled trial of a 12-week intervention involving 10physiotherapy visits together with home practice. Mea-surements will be taken at baseline and immediatelyfollowing the intervention (12 weeks) as well as at32 weeks and 52 weeks to examine maintenance of anyintervention effects. Specific assessment of adherence tothe treatment program will also be made at weeks 22 and42. The study will be conducted at two sites, Melbourneand Brisbane, Australia to facilitate generalizability of theresults and to ensure timely recruitment. The protocol willconform to CONSORT guidelines for reporting non-pharmacological interventions [39] and has been regis-tered with the Australia and New Zealand Clinical TrialsRegistry prior to study commencement.ParticipantsWe will recruit participants with painful knee OA fromthe community in the Melbourne and Brisbane metro-politan regions. A number of recruitment strategies willbe used including (i) advertising through local clubs,community centers, newspapers, Australian HealthManagement, Arthritis Australia and University web-sites, University staff newsletters, radio, and Facebook;(ii) placing brochures and study posters in medical andBennell et al. BMC Musculoskeletal Disorders 2012, 13:129 Page 3 of 17http://www.biomedcentral.com/1471-2474/13/129physiotherapy clinics; (iii) conducting presentationsabout knee OA in the local community, and (iv) usingour database of people who have been recruited fromthe community for prior studies and have given consentfor future contact.To be eligible, participants must fulfill the followingcriteria:i. Aged ≥ 50 years;ii. Knee OA fulfilling American Collegeof Rheumatology classification criteria [40]of knee pain on most days of the past monthand tibiofemoral osteophytes on x-ray(Kellgren and Lawrence≥Grade 2) [41];iii. Knee pain for ≥ 3 months;iv. Overall average knee pain in the last week ≥ 40 ona 100 mm Visual Analogue Scale (VAS);v. Western Ontario and McMasterUniversities (WOMAC) OsteoarthritisIndex physical function score of ≥ 25 indicatingat least a moderate level of difficulty in performingactivities of daily living.The exclusion criteria are:i. Knee surgery including arthroscopy within the past6 months;ii. Awaiting or planning any back or lower limbsurgery within the next 12 months;iii. Current or past (within 3 months) oral orintra-articular corticosteroid use;iv. Systemic arthritic conditions such as rheumatoidarthritis;v. Physiotherapy, chiropractic or acupuncturetreatment or exercises specifically for the kneewithin the past 6 months;vi. Walking >30 min continuously daily orparticipating in a regular (more than twice a week)exercise program;vii. Past participation in a PCST program;viii. Inability to walk unaided as this is necessary forsome of the physical testing;ix. Medical condition precluding safe exercise such asuncontrolled hypertension or heart condition;x. Self-reported psychiatric history such asschizophrenia;xi. Self-reported diagnosis of current clinicaldepression;xii. Neurological condition such as Parkinson’s disease,Multiple sclerosis or stroke;xiii. Inadequate written and spoken English;xiv. Unable to comply with the protocol such asinability to attend therapy sessions or attendassessment appointments at the University.ProcedureThe procedure is outlined in Figure 1. Preliminaryscreening will be conducted over the telephone by a re-search assistant not involved in outcome assessment.Volunteers who are deemed potentially eligible willundergo a semiflexed posteroanterior x-ray of their pain-ful knee (the more symptomatic knee in cases of bilat-eral eligible knee pain) at one of six trial radiologycentres unless they can provide their own such filmsfrom within the previous 12 months. X-ray grading willbe performed by two trained researchers at each site andany disagreement will be resolved through discussion orwhere necessary, by a third rater. A screening record willbe maintained to document the criteria eliminatingthose found to be ineligible. Participants will attend theUniversity of Melbourne or the University of Queens-land for baseline testing, following which they will berandomised into one of three intervention groups: (i) ex-ercise; (ii) PCST; or (iii) integrated exercise and PCST.Each intervention will last for 12 weeks and will involve10 individual visits to a project physiotherapist togetherwith home exercise and/or pain coping skills practice.Following the intervention, participants will continuetheir home exercise and/or pain coping skills practiceunsupervised for nine months. Participants will be re-assessed at week 12 (immediately following the interven-tion), at week 32 (by postal questionnaires) and at week52 (at the University). Additional questionnaires relatingto home practice adherence will also be collected atweeks 22 and 42. Participants will also wear a pedometerfor 7 consecutive days at baseline, 12 weeks and52 weeks. All participants will be asked to refrain fromseeking other forms of treatment during the trial. How-ever, due to ethical considerations, analgesia and non-steroidal anti-inflammatory drugs will be permitted asrequired.Ethics approval has been obtained from the Universityof Melbourne Human Research Ethics Committee(HREC: 1033341) and Radiation Safety Human Servicesand from the University of Queensland Medical Re-search Ethics Committee (MREC: 2010000340) and Ra-diation Protection Advisor. All participants will providewritten informed consent.BlindingThe outcome assessors at both sites will be blind togroup allocation and will not be involved in providingthe interventions nor will they visit any of the physio-therapy treatment centers. Participants will be requestednot to disclose details about their treatment to the out-come assessors. The physiotherapists as well as the psy-chologists and researchers managing the study at bothsites are by necessity unblinded. The statistician will beBennell et al. BMC Musculoskeletal Disorders 2012, 13:129 Page 4 of 17http://www.biomedcentral.com/1471-2474/13/129blind to group allocation until completion of the statis-tical analyses.Randomisation and allocation concealmentThe randomisation schedule will be prepared by thestudy biostatistician using a computer generated randomnumbers table. Randomisation will be conducted by ran-dom permuted blocks of size from 4 to 6 stratified bysite (Melbourne or Brisbane) and gender (male or fe-male). To conceal randomisation, an independent staffmember will prepare consecutively numbered, sealed,opaque envelopes for each site. The envelopes will bekept in a locked location at each site accessible only byan unblinded researcher. Each envelope will be openedin sequence once the participant has completed thebaseline measurements by a person not involved in par-ticipant recruitment. An unblinded researcher will thenschedule the participants’ first appointment with theirtreating physiotherapist.PhysiotherapistsTwenty-one experienced physiotherapists, 11 in Mel-bourne (six delivering both PCST only and integratedinterventions, five delivering the exercise only interven-tion) and 10 in Brisbane (five delivering both PCST onlyand integrated interventions, five delivering the exerciseonly intervention), with at least five years of musculo-skeletal clinical experience and located at various metro-politan private practices will be trained to deliver theinterventions. Two of the PCST physiotherapists inFigure 1 Diagram of study protocol.Bennell et al. BMC Musculoskeletal Disorders 2012, 13:129 Page 5 of 17http://www.biomedcentral.com/1471-2474/13/129Melbourne have prior experience with the delivery ofPCST interventions and regularly use a similar form ofthis treatment in their practices. To ensure no carry overeffects from training in PCST, the physiotherapists deli-vering the exercise only intervention will not be trainedto provide the PCST. This large number of treating phy-siotherapists is necessary for practical reasons and to im-prove the generalisability of the results.PsychologistsFour clinical psychologists (two at each of the Melbourneand Brisbane sites) will be responsible for the ongoingPCST training and monitoring of the physiotherapistswho are involved in the delivery of PCST. These includeone senior psychology researcher at each site who willoversee and guide one site psychologist. The two site psy-chologists will be responsible for the ongoing training andmonitoring of the physiotherapists. Both have more thanfive years of clinical experience and specialise in painmanagement.Training of physiotherapistsThe physiotherapist training to deliver the PCST interven-tion will involve an initial 4-day workshop facilitated by apsychologist and pain management specialist (FK) whodeveloped the PCST program. The physiotherapists willthen participate in regular tutorials and role-playing withthe site psychologists as well as individual practice with anindependent person acting as a patient. The physiothera-pists will be accredited to deliver the 10 PCST treatmentsessions once audio-recordings of each practice session arereviewed by the site psychologist and meet pre-specifiedcriteria for content and quality of delivery.All 21 physiotherapists will be trained to deliver theexercise program. Training will comprise a 4-hour prac-tical workshop conducted by musculoskeletal phy-siotherapists. The physiotherapists will be provided witha detailed study manual and a DVD of the trainingworkshop.InterventionsParticipants in each intervention group will attend 10 in-dividual treatment sessions with a project physiotherap-ist. The timing of the sessions is approximately once perweek. This reflects a realistic dosage in clinical practiceand is indicative of the timeframe needed for exerciseand PCST effects. Participants will choose the physio-therapist to attend based on the geographical availabilityof therapists trained to deliver their assigned interven-tion. Treatment will commence within one week of thebaseline assessment.Exercise interventionThis will be a home-based exercise program designed tostrengthen lower limb muscles and incorporates exercisescommonly used in clinical practice. It is based on clinicaltrials showing that such exercise programs improve painand function [42]. Six exercises aimed to strengthen thequadriceps, hamstrings, and hip abductor muscles will betaught to participants by the physiotherapist (Table 1). Ateach of the treatment sessions, the physiotherapist willmonitor proper performance and exercise intensity andprogress the exercises, if necessary. Intensity will be deter-mined by the participant’s ability to complete 10 repetitionsfor a given exercise and by perceived difficulty (using amodified Borg scale for resistance training [43]). The par-ticipant will be asked to perform the prescribed home exer-cises four times weekly aiming for a dosage of 3 sets of 10repetitions. Weights and resistance elastic bands as well ashandouts describing the exercises will be provided. Eachphysiotherapy treatment session will be 25 mins in length(Figure 1, Table 2).Pain Coping Skills Training (PCST) interventionThe PCST program has been designed to specifically en-hance the participants’ ability to employ behavioral andcognitive pain coping strategies aimed at increasing self-efficacy and decreasing pain and pain catastrophising.The program involves 10 weekly modules (Table 3) andTable 1 Description of the exercise programExercise Description and progressionKnee extensor strengthening Seated knee extensions with ankle weights. Ankle weights progressed.Hip abductor strengthening Level 1: Side lying with ankle weights. Ankle weights progressed.Level 2: Standing hip abduction with elastic band around ankles. Elastic band resistance progressed.Weight-bearing knee/hipextensor strengtheningLevel 1: Partial wall squats (option to add elastic band around knees to incorporate hip abductor muscles).Level 2: Sit-to-stand (option to add elastic band around knees to incorporate hip abductor muscles).Level 3: Split sit-to-stand (or split partial wall squats) – most weight bearing on affected side.Knee flexor strengthening Seated knee flexion against elastic band resistance. Elastic band resistance progressed.Step Ups/Step Down Progress by increasing the height of the step then adding weight (i.e., back pack or hand weights).All exercises must be progressed during the program. Dosage is 3 × 10 repetitions with 5 second holds for all exercises with the exception of level 2 and 3 sit tostand exercises which have 3 second holds.Bennell et al. BMC Musculoskeletal Disorders 2012, 13:129 Page 6 of 17http://www.biomedcentral.com/1471-2474/13/129is similar to those used in previous studies in knee OA[24,34]. Interactive sessions will emphasise the partici-pants’ understanding of the neuromechanical processesof pain which underscores the role that PCST can playas well as provide training in a number of pain copingskills (activity-rest cycling, pleasant activity scheduling,problem solving, identifying negative thoughts, challen-ging negative thoughts, developing coping thoughts,pleasant imagery, counting backwards, and auditorystimulation) and in practical ways of applying newlydeveloped coping skills. The participant will be askedto perform the prescribed home PCST practice dailywith the dosage dependent on the actual skill taughtduring the particular week. Each physiotherapy treat-ment session will be 45 mins in length (Figure 1,Table 2).Integrated exercise and PCST interventionThis intervention will integrate both the exercise andPCST programs described above within a single inte-grated treatment session. While the therapist will de-liver the same PCST program, examples relating toexercise and activity can be used as material for theskills being taught to a greater extent than in thePCST alone group. Participants will be encouraged toincorporate the learned PCST skills into their home ex-ercise program and specific training will be providedon how learned skills can be applied during the exer-cise performance at the same clinic visit. Each physio-therapy treatment session will be 70 min in lengthwhich includes 25 mins of exercise and 45 mins ofPCST (Figure 1, Table 2).Quality assurance and intervention integrityPhysiotherapist adherence to the PCST protocol andquality of delivery will be monitored and enhanced byregular (approximately fortnightly) team meetingswith the site psychologist. Audio-recordings of eachphysiotherapy treatment session will be reviewed bythe site psychologist with feedback provided to thephysiotherapist. For the two treatment groups involv-ing PCST, the site psychologists will formally rate ap-proximately 10% of sessions using randomly selectedtreatment audiotapes. The Melbourne psychologistwill rate those from Brisbane and vice versa followinga period of training to ensure consistency of ratings.The physiotherapist delivery of the PCST treatmentswill be rated on two aspects: 1) adherence to theprotocol using a yes/no format and calculated to givea percentage score; and 2) quality of the treatmentusing a 1–5 scale (1 being poor, 2 fair, 3 satisfactory,4 very good and 5 excellent) for each of the followingtherapist behaviours: establishes/maintains rapport;remains on schedule with the protocol or makes ap-propriate adjustments when indicated; applies PCSTprotocol to participant’s situation and current chal-lenges; encourages participant’s active involvement inthe session; uses time effectively/appropriate pacing;demonstrates good interpersonal skills; demonstratesprofessionalism and clinical judgment; overall effect-iveness/skill of the therapist.In addition, an unblinded research assistant will ran-domly observe selected treatment sessions for all treat-ment arms at each physiotherapy clinic and providefeedback to the therapist. Group-specific quality assur-ance checklists will be completed by the researchTable 2 Overview of pain coping skills training (PCST) and exercise componentsPain coping skills training ExerciseSession 1 Session 1• Introduction and discussion of patient assessment form • Introduction and discussion of patient assessment form• Patient education about knee OA and treatment • Patient education about knee OA and treatment• Teach patient weekly home PCST practice • Teach patient home exercises• Home practice prescribed daily • Home exercises prescribed 4 times/week• Home practice prescribed daily • Discuss patient log book and attendanceSession 2-10 Session 2-10• Review of previous week • Review of previous week• Teach patient new pain coping skill • Check and progress patient home exercises• Check and progress patient home practice • Home exercises prescribed 4times/week• Home practice prescribed daily • Check patient log book• Check patient log book and set goals for the weekFollow-up period Follow-up period• Home practice prescribed as required • Home exercises prescribed 3 times/weekBennell et al. BMC Musculoskeletal Disorders 2012, 13:129 Page 7 of 17http://www.biomedcentral.com/1471-2474/13/129assistant during the session. The checklist containsitems pertaining to the protocol such as therapist timespent solely with the participant, treatment notes com-pleted, review of home practice and handouts provided,through to more complex items such as therapist obser-vation of participant practicing the exercises, therapistuse of the rating perceived exertion scale to review in-tensity of exercises and progression of weights ifrequired and for those in the PCST treatment groups,new concepts introduced and explained. Lastly, partici-pants will complete a questionnaire about their experi-ence with the physiotherapist including whether theywould recommend the physiotherapist to someone theyknew.Follow-up periodDuring the 9-month unsupervised follow-up period, par-ticipants in the exercise only and integrated groups willbe requested to continue their home strengthening pro-gram but this will be reduced to three times per week.Participants in the PCST and integrated groups will berequested to continue their PCST home practices asneeded (Figure 1, Table 2). At weeks 22, 32, 42 and 52participants will be requested to complete questionnairesposted to them pertaining to adherence to the homeprogram. In addition, the treating physiotherapist willtelephone participants at weeks 22, 38 and 46 to discussprogress with the aim of enhancing adherence to thehome program. After the 9-month period (week 52),Table 3 Description of the Pain Coping Skills Training (PCST) InterventionPCST session Content Home practice dosageSession 1: Progressive Muscle Relaxation (PMR) - Introduce gait control theory 2 PMR practices per day- Provide rationale for pain coping skills training- Train participant in PMRSession 2: Mini-Practices - Review PMR 10 or more mini-practices per day-Train participants on mini-practicesSession 3: Activity-Rest Cycling - Review PMR and mini-practices Use technique twice per week- Introduce activity-rest cyclingSession 4: Pleasant ActivityScheduling- Describe how pleasant activityscheduling can be used to controland decrease pain3 pleasant activities per week- Set pleasant activity goals with participant- Discuss how to use mini-practices andactivity-rest cycling in achieving pleasantactivity goals.Session 5: Identifying Negative Thoughts,Thought Records- Present cognitive model (ABC Model-howan event leads to automatic thoughts andresult in certain consequences)Record situations and thoughts daily- Teach participant how to usethought records to monitor negative thoughtsSession 6: Challenging Negative Thoughts,Calming Self-Statements- Work with participant to challengenegative thoughtsPractice developing alternative copingthoughts daily- Develop calming self-statementsSession 7: Problem Solving I,Pleasant Imagery andDistraction Techniques I- Training in problem solving Problem solving activity: 1 per day- Training in pleasant imagery- Training in counting backwards Pleasant imagery: 2 per daySession 8: Distraction Techniques II,Review of Skills- Train use of focal points andauditory stimulation as distraction methods3 distraction techniques per week- Review skills from previous weeksSession 9: Problem Solving II (Applying PainCoping Skills in Problem Situations)- Identify problem situations Record situations and thoughts daily- Develop coping plansSession 10: Coping Skills Maintenance,Early Warning Signs/Developing a Coping Plan- Review principles of relapse prevention- Identify early warning signs of reduced coping- Develop coping plans to address lapses in copingAll components of the PCST program are mandatory. Home practice from each session is carried forward into the remainder of the program.Bennell et al. BMC Musculoskeletal Disorders 2012, 13:129 Page 8 of 17http://www.biomedcentral.com/1471-2474/13/129participants will attend a follow up testing session at theUniversity laboratory.MeasurementsBaseline descriptive data will be obtained by question-naire and will include age, sex, duration of knee OAsymptoms, previous treatment, surgery and medicationuse for knee OA, employment status, marital status,education level and previous health problems. Radio-graphic disease severity will be assessed from x-ray usingthe Kellgren and Lawrence grading system [41]. Mea-sures of height and weight will be taken and body massindex calculated. A summary of all measures collected inthe trial are shown in Table 4.Primary outcome measuresOutcome measures have been selected based on thoserecommended for clinical trials of OA [44]. The primaryoutcomes are change in self-reported pain and physicalfunction at 12 weeks.a) Pain: average knee pain in the past week will beself-assessed using a 100 mm Visual AnalogueScale (VAS) with terminal descriptors of “no pain”and “worst pain possible”. The VAS painmeasurement has demonstrated reliability in OA[44].b) Physical function: this will be self-assessed usingthe Western Ontario and McMaster UniversitiesOsteoarthritis Index (WOMAC) Likert version 3.1.The physical function subscale has 17 items with afive point Likert response (0 indicating no physicaldysfunction, 5 indicating severe physicaldysfunction) giving a total score out of 68. TheWOMAC is a disease-specific instrument whosevalidity, reliability, and responsiveness have beendemonstrated in an extensive range of OAstudies [45].Secondary outcome measuresOther pain measures Pain will also be self-assessedusing the pain subscale of the Western Ontario andMcMaster Universities Osteoarthritis Index (WOMAC)Likert version 3.1. The pain subscale has 5 items with afive point Likert response (0 indicating no pain, 5 indi-cating severe pain) giving a total score out of 20. It is avalid and reliable measure that has been used extensivelyin OA studies [45].Global rating of change Participants will rate their per-ceived overall change in their knee and their change inpain and in physical function with treatment comparedto baseline on seven-point ordinal scales (1-much worse,to 7-much better). Scales of this kind are frequently usedas an external criterion for comparison with changes inscores of other outcomes. Measuring patient-perceivedimprovement using a rating of change scale has beenshown to be a clinically relevant and stable concept forinterpreting truly meaningful improvements from the in-dividual perspective [46].Strength Maximal normalized isometric quadriceps andhamstring muscle strength (peak torque; Nm/kg) at 90degrees of knee flexion will be assessed in sitting usingcustom-designed apparatus comprising a force trans-ducer (Sparker Instruments, Wenzhou, China) attachedto a chair. Lever arm length will be measured as the dis-tance from the knee joint line to the mid-point on theankle cuff. After two submaximal warm-up trials, a totalof three maximal three second trials will be performedseparated by a 30 second rest period. The peak value willbe used for analysis. Maximal normalized isometric hipabduction strength (peak torque, Nm/kg) will be mea-sured in supine using a handheld dynamometer(Nicholas MMT, Lafayette Instruments, Lafayette, IN)[47]. After two, sub-maximal warm-up trials to familiar-ise participants with the testing procedure, participantswill perform three maximal contraction trials each ofthree seconds duration, separated by 30 seconds of rest.The mean of the two trials will be used for analysis.Functional performance The 30-second sit-to-standtest provides a direct, objective measure of physicalfunction [48]. After a practice trial, the number of timesparticipants can rise to a full standing position from sit-ting and return to sitting, with arms crossed and heldagainst the chest, in 30 seconds will be counted. Agreater number indicates better performance.Walking performance will be assessed by calculatingwalking velocity (m/sec) as participants walk 20 metersin their usual footwear with the instructions “walk asquickly as you can without overexerting yourself” [49].This will be performed twice and the times averagedwith higher velocity values indicating better walkingperformance.Dynamic standing balance will be assessed by the steptest. Participants are requested to stand on their mostpainful leg in front of a 15 cm high step. They arerequired to place their contralateral foot up and downon the step as many times as they can in 15 seconds. Apractice trial involving 3–4 steps will be performed priorto the test trial. A greater number of steps indicatesgreater balance and function [50].The timed up and go (TUG) test evaluates walkingspeed and mobility [51]. Participants are instructed tostand up from a standard height chair and walk attheir normal pace around a marker 3 meters awayBennell et al. BMC Musculoskeletal Disorders 2012, 13:129 Page 9 of 17http://www.biomedcentral.com/1471-2474/13/129before returning to the chair and sitting down again. Atotal of two trials will be performed and the best resulttaken as the final score. Faster times indicate greaterperformance.Physical activity level Habitual physical activity willbe measured in two ways, one using a questionnaireand the second using a pedometer. The Physical Ac-tivity Scale for the Elderly (PASE) is a 10-itemTable 4 Summary of measures to be collectedPrimary outcome measures Data collection instrument Collection pointsAverage overall pain in past week 100 mm VAS 0, 12, 32, 52 weeksSelf reported physical function inpast 48 hoursPhysical function subscale WOMACOsteoarthritis Index 3.1 Likert version0, 12, 32, 52 weeksSecondary outcome measuresPain Pain subscale WOMAC Osteoarthritis Index 3.1 Likert version 0, 12, 32, 52 weeksGlobal rating of change Overall, for pain and for function – 7 point scale 12, 32, 52 weeksPerceived response to treatment – 7 point ordinal scale 12, 32, 52 weeksMuscle strength Isometric quadriceps and hamstrings in sitting using a force transducer 0, 12, 52 weeksIsometric hip abductors – Hand held dynamometer in supine 0, 12, 52 weeksFunctional performance Timed 20 m walk 0, 12, 52 weeks30 second sit-to-stand 0, 12, 52 weeksTimed Up and Go 0, 12, 52 weeksStep test 0, 12, 52 weeksPhysical activity levels Physical Activity Scale for the Elderly (PASE) 0, 12, 32, 52 weeksPedometer worn for 7 days 0, 12, 52 weeksHealth-related quality of life Assessment of Quality of Life Instrument version 2 (AQoL II) 0, 12, 32, 52 weeksSelf-reported psychological measures Arthritis Impact Measurement Scale 2 0, 12, 32, 52 weeksArthritis Self-Efficacy Scale 0, 12, 32, 52 weeksArthritis Self-Efficacy for Pain communication Scale 0, 12, 32, 52 weeksPain Self-Efficacy Scale 0, 12, 32, 52 weeksPain Catastrophising Scale 0, 12, 32, 52 weeksCoping Strategies questionnaire 0, 12, 32, 52 weeksDepression, Anxiety & Stress subscale 0, 12, 32, 52 weeksHolding Back Scale 0, 12, 32, 52 weeksPatient Health Questionnaire-9 0, 12, 32, 52 weeksSelf Efficacy for Exercise Scale 0 weeksBarriers to Exercise Scale 0 weeksBenefits of Exercise Scale 0 weeksBarriers and enablers to home exercise 32, 52 weeksOther measuresTreatment credibility Treatment Credibility Scale 1, 12 weeksTreatment session attendance Therapist treatment records During interventionHome practice during treatment Participant log book – number of days/times completed Daily during interventionTherapist rating of participant adherence using 11-point numeric rating scale 12 weeksSelf-rated using 11-point numeric rating scale 22, 32, 42, 52 weeksHome practice during follow up Questionnaire-number of days performed exercises/pain coping skills in past week 22, 32, 42, 52 weeksQuestionnaire - usefulness of pain coping skills 32, 52 weeksAdverse events Participant logbook Daily during interventionQuestionnaire 32, 52 weeksHealthcare usage and related costs Questionnaire and health system 0, 12, 32, 52 weeksBennell et al. BMC Musculoskeletal Disorders 2012, 13:129 Page 10 of 17http://www.biomedcentral.com/1471-2474/13/129questionnaire that will be used to measure both thefrequency and type of recreational and occupationalphysical activity undertaken by participants over theprevious week. Higher scores indicate greater levelsof physical activity. The PASE was developed andvalidated in samples of older adults ≥ 55 years ofage [52].A pedometer (HJ-005 Omron Healthcare, Japan) will beworn at the waist for seven consecutive days on threeoccasions (baseline, week 12 and week 52) to record thenumber of steps taken per day. Participants arerequested to wear the pedometer full time during theirwaking hours. Pedometers have been found to be a sim-ple and inexpensive means to estimate physical activitylevels [53].Health-related quality of life This will be assessedusing the Assessment of Quality of Life instrument ver-sion 2 (AQoL II) which has 20 questions covering sixdimensions including independent living, social relation-ships, physical senses, coping, pain and psychologicalwellbeing [54]. The AQoL II is a multi-attribute instru-ment with strong psychometric properties. It produces asingle utility index that ranges from −0.04 (worst pos-sible health-related quality of life) to 1.00 (full health-related quality of life) [54].Psychological measures The Arthritis Impact Meas-urement Scale (AIMS2) is a disease-specific self-reported instrument designed to assess the healthstatus of those with rheumatic conditions [55]. It iscomprised of 12 subscales, of which three will beused in this study. Two are psychological subscalesrelating to levels of mood (5 questions) and tension(6 questions) over the past month while the thirdpertains to thoughts of overall arthritis impact (1question). Questions are rated on a 5-point Guttmanscale and total scores are summed with a range from0 to 60 with higher scores indicating greater disabil-ity. The AIMS2 has high-internal consistency, test-re-test reliability and validity and is moderately sensitive tochange [55].The Arthritis Self-Efficacy Scale is a 20-item ques-tionnaire with three subscales that assess self-efficacyfor control of pain management (5 questions), forphysical function (9 questions) and for other arthritissymptoms (6 questions). Questions are rated on anumerical rating scale from “1” (very uncertain), to“10” (very certain). The mean value of each of theitems in a subscale provides an overall score foreach subscale. The range of scores for each of thesubscales is from 1 to 10. Higher scores indicatehigh levels of perceived self-efficacy. Previous studiessupport the reliability and validity of this scale inthose with arthritis [56].The Arthritis Self-Efficacy Scale for Pain Commu-nication is a modified version of the Arthritis Self-Efficacy Scale [56]. It is comprised of 5-items asses-sing the participants’ level of confidence in commu-nicating their pain to their spouse/partner and theirconfidence that they will receive help, support andunderstanding from them [57]. Items are rated on ascale of “10” (very uncertain) to ”100” (very certain).Scores range from 10 to 100 and summary scoresare determined by calculating the mean ratingacross all 5 items. This scale has been found tohave good internal consistency (Cronbach α=0.94)[57].The Pain Self-Efficacy Questionnaire is a 10-itemscale measuring both the individuals’ expectation andconfidence that they can perform a particular task. Itcovers a range of functional and non-functional taskssuch as housework, socialising and coping with painwithout medication. Questions are rated on a 7-pointLikert scale ranging from “0” (not confident at all) to“6” (completely confident). The scores are summedto give a total score with a range of 0–60. Indivi-duals with scores <20 are considered to have lowpain self-efficacy whereas those with scores >40 areconsidered to have high pain self-efficacy [58]. Thisquestionnaire is a valid measure with high internalconsistency and test-retest reliability [59].Pain catastrophising will be measured using the13-item Pain Catastrophising this scale is dividedinto three subscales that measure tendencies to ru-minate about pain (4 questions), magnify pain (3questions), and feel helpless about pain (6 ques-tions). Each item is rated on a 5-point scale rangingfrom “0” (not at all) to “4” (all the time). The totalscore is a sum of all items from each subscale withhigher scores indicating greater levels of catastro-phising. The scale has high internal consistency andis associated with heightened pain, psychological dis-tress, and physical disability [60].The Coping Strategies Questionnaire (CSQ) [61] willbe administered to assess both cognitive and behaviouralpain coping strategies. This 48-item self-reported ques-tionnaire comprises a cognitive pain coping strategiescomponent with 44 items covering 6 subscales: divertingattention, catastrophising, reinterpreting pain sensations,ignoring sensations, coping self-statements and prayingand hoping. The behavioural pain coping strategies com-ponent assesses the effectiveness of the strategies aboveto control and decrease pain. It is comprised of 4 itemscontaining 2 subscales: increasing pain behaviour and in-creasing behavioural activities. Each of the 48 items isscored on a 7-point Likert scale from “0” (never do that)Bennell et al. BMC Musculoskeletal Disorders 2012, 13:129 Page 11 of 17http://www.biomedcentral.com/1471-2474/13/129to “6” (always do that). Participants also rate their overalleffectiveness of the coping strategies used, how muchcontrol they feel they have over their pain and howmuch they feel they are able to reduce their pain. Rat-ings are made on a 7-point Likert scale ranging from “0”(no control/can decrease pain somewhat) to “6”(complete control/can decrease it completely) [61]. Totalscores are obtained by the sum of each of the valuesfrom questions pertaining to a particular subscale foreach of the cognitive or behavioural components. Higherscore values indicate greater pain coping abilities. TheCSQ is a commonly used instrument in both clinicaland research settings [62]. It has demonstrated sensitiv-ity to change from treatment in chronic pain samples aswell as good construct validity as well as good internalconsistency [61,62].The Depression, Anxiety and Stress Subscale (DASS)measures three related negative emotional states of de-pression, anxiety and stress [63]. The 21-item subscalewill be used instead of the full version as it is more prac-tical for research purposes. This version has 7 questionsfor each of the three subscales taken directly from theDASS questionnaire. Questions consist of statementspertaining to the past week and are rated on a 4-pointscale ranging from “0” (did not apply to me) to “3” (ap-plied to me very much, or most of the time). Scoresfrom each subscale are summed and multiplied by two.Thus, subscale scores range from 0–42 with higherscores indicating greater levels of distress. It has high in-ternal consistency and construct validity [63,64].The Holding Back Scale (HBS) is a modified version ofthe Emotional Self-Disclosure Scale (ESDS) developedby Snell et al. [65]. The HBS assesses the extent to whichparticipants discuss their OA disease-related thoughtsand feelings with their spouse/partner [66]. High levelsof holding back have been found to be significantly asso-ciated with increased psychological distress and poor re-lationship functioning in cancer patients [66-68]. It iscomprised of 11-items relating to OA-related fear andconcerns, pain, body appearance, financial and job-related concerns. Those without a spouse or significantother will be advised to think of someone they are closeto such as a child or friend when completing the ques-tionnaire. Questions were rated on a 6-point Likert scalefrom “0” (not at all) to “5” (a lot). The sum of scoresrange from 0–55 for a total score with higher scoresrepresenting a greater willingness to discuss relevantemotions with a spouse or significant other [65]. TheHBS is a valid and reliable measure which has high in-ternal consistency in cancer patients [69].The Patient Health Questionnaire-9 (PHQ9) is a shortversion of the Patient Health Questionnaire (PHQ) thatscreens for psychological disorders. The PHQ-9 is a 9-item depression scale that scores each of the 9 DSM-IVcriteria. It can both establish the depressive disorder aswell as determine symptom severity. Questions arerated on a 4-point scale ranging from “0” (not at all) to“3” (nearly every day). The sum of scores with a rangeof 0–27 determines the severity measure. Scores≤ 15represent none to moderate severity of depression andthose ≥ 15 represent moderately severe to severe depres-sion. It is commonly used in clinical settings and is areliable and valid measure of depression severity [70].Three constructs for physical activity beliefs will beassessed at baseline by reliable questionnaires. The Self-Efficacy for Physical Activity Scale is a 5-item scale thatevaluates confidence in ability to participate regularly inphysical activities during a variety of situations and feel-ings with higher scores indicating greater self-efficacyfor physical activity [71]. The Benefits of Physical Activ-ity Scale is a 14-item scale that examines whether parti-cipants are aware of the benefits of physical activity interms of physical, psychological and social constructswith higher scores indicating a perception of more bene-fits [71]. The Barriers to Physical Activity Scale is a 23-item self-report questionnaire that identifies the extentto which specific conditions are considered to make par-ticipation in physical activities difficult [71]. Higherscores indicate a perception of more barriers to physicalactivity and have been correlated to less exercise partici-pation [71]. In addition, a customized questionnaire re-lating to barriers and enablers to the homestrengthening exercise program will be administered tothe exercise groups.Other measuresParticipants will rate their thoughts about the treatmentcredibility and their treatment expectations after the firstand last treatment sessions (Weeks 1 and 12) using a 4-item scale that has been previously described [72]. Thefirst three questions are rated on a 11-point numericalrating scale from “0” (not at all confident) to “10” (abso-lutely confident) and pertain to the participants’ confi-dence in the treatment to manage and relieve their painas well as their confidence in recommending the treat-ment to a friend in a similar condition. The last questionassesses how logical the participant thought the treat-ment was and is rated on a 7-point numerical ratingscale ranging from “0” (not logical at all) to “6” (verylogical).Participant adherence to the treatment program in allthree groups will be obtained by recording the numberof physiotherapy sessions attended (out of a maximumnumber of ten). Participants will maintain a daily log-book to record the number of home exercises and/orpain coping skills practice completed during the 12-weektreatment phase. The physiotherapist will also indicateon an 11 point numeric rating scale, their perceivedBennell et al. BMC Musculoskeletal Disorders 2012, 13:129 Page 12 of 17http://www.biomedcentral.com/1471-2474/13/129rating of the participant’s overall adherence to the treat-ment program with “0” (not at all) to “10” (completely asinstructed) [73].Home exercise adherence during the follow up periodwill be measured in the exercise only and the integratedexercise and PCST groups via a self-report question-naire. This will be administered on four occasions (22,32, 42 and 52 weeks). Participants will be asked howmany times in the previous week they performed thehome exercises. These will be summed over the fouroccasions (maximum of 12 exercise days). To assess theamount of home pain coping skills practice performedduring the follow-up period, participants in the PCSTalone group or the integrated group will be mailed ashort questionnaire at weeks 22, 32, 42 and 52 whichasks how many times they have performed each of thedifferent skills practice in the previous week. Participantsin all groups will provide a self-rating of their adherenceto their specific home program using a similar 11 pointnumeric rating scale with “0” (not at all) to “10” (com-pletely as instructed) at weeks 22, 32, 42 and 52. Fur-thermore, the usefulness of the various pain coping skillswill be determined at weeks 32 and 52 in the PCSTgroups by a 5-point likert scale from “1” (not useful) to“5” (very useful).Adverse events and the use of co-intervention will berecorded in the logbook during the treatment periodand via questionnaire at weeks 12, 32, and 52 in the fol-low-up phase. Adverse events will be defined as anyproblem from the exercises and/or pain coping skillsthat lasted for more than two days and caused them toseek treatment.Information on health care costs and direct non-healthcare costs over the entire study period (52 weeks) will becollected by questionnaire. Direct health care costs willinclude costs of physiotherapy attendance, additionalhealth provider visits (doctors, specialists, other healthcare professionals), investigative procedures, purchase ofprescription and over the counter medication, home careand hospitalization. Direct non-health care resourceswill include number of lost days from work.Sample sizeThe two primary endpoints are knee pain measured onthe VAS and WOMAC physical function score. Theminimum clinically important difference to be detectedin OA trials is a change in pain of 18 mm (on 100 mmVAS) [44] and a change of six physical functionWOMAC units (out of 68) [73]. Based on our previousdata, we assume a common between-participant stand-ard deviation of change of 30 mm for pain and 12 unitsfor WOMAC physical function. These statistics indicatea smaller standardized effect size of interest (Cohen’s d)of 0.5 for the WOMAC measure than the d of 0.6 forpain. Since primary analyses will adjust for baseline ofthe outcome measure by ANCOVA, the sample size todetect the above effect sizes with 80% power taking intoaccount a pre-post correlation of 0.50 is 48 patients perarm for WOMAC function and 33 per arm for VASpain. We obtained the pre-post correlation from datafrom Lim et al., in which correlations of 0.58 and 0.77were observed for pain and function respectively [74],and therefore our value of 0.50 is expected to be conser-vative. A further issue is that comparisons of the exercisealone arm with each of the PCST arms involve separatephysiotherapists per arm and therefore clustering effectsby physiotherapists need to be accounted for in the sam-ple size and analysis. Prior data indicate that the intra-physiotherapist correlation for pain and function is likelyto be at most 0.050 [74]. Assuming physiotherapists willeach treat on average 7 patients (hence clustering designeffect = 1 + 6*0.050 = 1.30), a total sample size of 63patients per arm is required. Assuming 10% dropoutincreases the sample size to 70 per arm, or 10 phy-siotherapists treating 7 patients each. Slight loss ofpower is expected due to imbalances in number ofpatients per physiotherapist; however with access to 11physiotherapists in the PCST arms this should be negated.The comparison of Exercise and PCST versus PCSTalone arms will be performed ‘within-physiotherapist’and hence does not suffer from these clustering effects.With 70 patients per arm the power is 93% to detect theWOMAC effect size and 98% for pain.Statistical analysisComparisons will be performed using an intention-to-treat analysis using all randomized participants. Thisanalysis will include all participants including those whohave missing data and those who are not fully adherentto the protocol. Some attrition is anticipated despite thefact that we will implement procedures to minimize lossto follow-up and participant withdrawal, and maximizeadherence. We will employ multiple imputation methodsto handle missing data in the analyses. Standard diag-nostic plots will check model assumptions. Effect sizeswill be calculated with an effect size of 0.2 being small,0.5 medium and 0.8 large. All tests will be two-tailedand carried out at the 5% significance level.Demographic and clinical characteristics as well asbaseline data will be presented to assess the baselinecomparability of the intervention groups. These variableswill also be examined for those participants who with-draw from the study. Descriptive statistics will be pre-sented for each group as mean change (standarddeviation, 95% confidence intervals) in the two primaryoutcomes from baseline to 12 weeks. Between-groupBennell et al. BMC Musculoskeletal Disorders 2012, 13:129 Page 13 of 17http://www.biomedcentral.com/1471-2474/13/129mean differences and 95% confidence intervals will beestimated with a mixed effects linear regression modelin which physiotherapists are treated as random effectsand baseline scores of the primary outcome are enteredas the covariate [75], together with adjustment for thestratification variables of site and gender. Secondary out-comes will be as above for normally distributed mea-sures, or will use binary or proportional odds randomeffects regression models for binary or ordinal outcomesas appropriate. Standard diagnostic plots will checkmodel assumptions. All tests will be two-tailed and car-ried out at the 5% significance level with no statisticaladjustment for multiple testing.Economic evaluationThe economic evaluation will assess the incrementalcost of the integrated intervention compared with PCSTand with exercise. In each case the incremental cost willbe compared to the incremental benefits of treatment interms of a clinically significant improvement in pain, aclinically significant improvement in function, and thedifference in quality adjusted life years (QALYs). The in-cremental QALYs will be measured by the betweengroup difference in the mean AQoL score over12 months. A social perspective on costs will be takenand will include resource use incurred both by healthservices and by the participant irrespective of paymentsource. Health care costs will be calculated from theutilisation data and average unit costs for each item. Wewill not include the costs of training the physiotherapistsin the delivery of PCST in the primary analysis. The in-clusion of time/productivity gains is controversial andthe cost effectiveness ratios will be calculated with andwithout these “indirect costs”. Confidence intervals forincremental cost effectiveness will be calculated directlyusing non parametric bootstrapping. In addition we willcalculate a cost effectiveness acceptability curve basedfor a range of hypothetical money values of outcomes[76]. This will be done using individual cost and out-come data over the 12 months or, if adjustments for im-balance at baseline and clustering of patients byphysiotherapist are necessary, analysed using a mixedlinear regression model [77].TimelinesThis study has been funded by Australian HealthManagement and ethics approval was obtained at theUniversity of Melbourne in March 2010 and at theUniversity of Queensland in June 2010. Radiation Safetyapproval to conduct weight bearing knee joint x-rayswas approved at the Universities of Melbourne andQueensland in June 2010. Recruitment and training ofthe physiotherapists occurred after this time period. Theanticipated timelines for the project at both sites are asfollows:August 2010 Baseline testing commencesMarch 2012 Recruitment completeJune 2012 All participants complete immediate postintervention testing (Week 12)March 2013 All participants complete 9 monthfollow-up (Week 52)DiscussionOsteoarthritis is a complex chronic disease with manyfactors that contribute to the pain and disability. Thisreinforces the need for a biopsychosocial approach whendeveloping treatments aimed at improving the self-management of this disease. Previous studies of musclestrengthening exercises have shown significant improve-ments in pain, physical function and performance mea-sures [5,25,78]. However, individuals with knee OA alsorequire treatment that addresses the psychological impair-ments that are commonly found [79]. An individuals’coping mechanism for how they deal with pain is ofcrucial importance [80]. For instance, those who developmaladaptive coping behaviours such as limiting activitydue to fear of increased joint damage and/or increasedpain, avoiding social obligations, pain catastrophisingand worry can lead to reports of increased pain and func-tional decline [80-83]. Depression, anxiety and stress arealso significant predictors of upcoming pain symptoms[3,31,84]. Overwhelmingly, the research suggests thatpain catastrophising and self-efficacy cognitive variablesare important predictors of pain and function [13,29,85].Keefe et al., provide support for the use of PCST inimproving psychological functioning in this patientpopulation [24,34,35,81]. Thus, given that both exerciseand PCST have been shown to be effective for symp-tomatic relief in knee OA, we contend that an inte-grated PCST and exercise program is likely to be moreefficacious than either intervention alone.The study design has several major strengths. First,considerable attention has been paid to quality controlof the PCST intervention. The project physiotherapistswill undergo rigorous training and accreditation to de-liver the PCST intervention. All sessions will be audiorecorded and then reviewed by the site psychologist.Regular meetings will be held throughout the study toprovide ongoing practice and supervision for the phy-siotherapists. PCST treatment fidelity will be formallydetermined via rating of a random sample of the PCSTtreatments. Second, the outcome measures used arevalid and reliable, cover a range of clinically importantconstructs and include those recommended for clinicaltrials of OA [44]. These include self-report measures ofpain, function, quality of life and global response toBennell et al. BMC Musculoskeletal Disorders 2012, 13:129 Page 14 of 17http://www.biomedcentral.com/1471-2474/13/129treatment. A range of other measures is also included toencompass functional performance, strength, psycho-logical functioning and physical activity levels. Inaddition, a health economic assessment is included giventhe need to justify the cost-effectiveness of treatments inthe current economic climate. Third, the study has beendesigned with attention to methodological features suchas randomization, concealed allocation and blinded out-come assessment. The sample size has been calculatedto detect minimal clinically important differences be-tween groups, taking into account the clustering effectsof the physiotherapists.ConclusionThis study uses a randomized controlled trial design toinvestigate the efficacy of an integrated exercise andPCST program delivered by physiotherapists in improv-ing pain and physical function in those with knee OAcompared with exercise or PCST alone. The novel find-ings will enable evidence-based recommendations as tothe efficacy of this conservative option for the manage-ment of patients with knee OA.Competing interestsThe authors declare that they have no competing interests.Authors’ contributionsKLB, FJK conceived the project and KLB is leading the co-ordination of thetrial. KLB, FJK, CB, GJ, MH, JK, MN, AF and AH designed the protocol andprocured the project funding. FJK developed the pain coping skills trainingprotocol, developed and ran the 4-day workshop to train thephysiotherapists and consults with the site psychologists regarding theprotocol on an as needed basis. GJ is leading the Brisbane study site. GJ, KLBand MH designed the exercise intervention and GJ and KLB wereresponsible for training the physiotherapists in the exercise protocol andgrading x-rays for participant inclusion. YA is a PhD student responsible formanaging the Melbourne study site and for recruiting and screeningparticipants. CB and JK are the study psychologists involved in overseeingthe site psychologists. AF provided the sample size calculations, designedthe statistical analysis and provided the randomization schedule. AHdesigned the economic evaluation. TE and BM assisted with development ofstudy material and are blinded assessors. BM set up the initial studydatabase. All authors provided feedback on drafts of this paper and read andapproved the final manuscript.AcknowledgementsThis trial is being funded by Australian Health Management. None of thefunders have any role in the study other than to provide funding. ProfessorBennell is partly funded by an Australian Research Council Future Fellowship.We wish to thank the Brisbane project manager Paul Connellan, the sitepsychologists Prue Lewis (Melbourne) and Denae Bacon (Brisbane) as well asthe Melbourne project physiotherapists: Nick Economis, Marie-LouiseFrancken, Arthur Lee, Ross Fraser, Gabrielle Molan, Frankie Mullen, BarryNguyen, Adrian Quinn, Anjelo Ratnachandra, Michelle Raymundo, andChristine Roberts and Brisbane project physiotherapists: Colwen Bacon,Sandra Day, Julie D’Mellow, Jane Elliot, Peter Ford, Serena Marshall, RodMcLean, Katrina Milicich, Joanne Minto, and Sonja Varendorf.Author details1Centre for Health, Exercise & Sports Medicine, Department of Physiotherapy,University of Melbourne, Melbourne, VIC, Australia. 2Psychological Sciences,University of Melbourne, Melbourne, VIC, Australia. 3Centre of ClinicalResearch Excellence in Spinal Pain, Injury & Health, School of Health andRehabilitation Sciences, University of Queensland, Brisbane, Qld, Australia.4Department of Physical Therapy, University of British Columbia, Vancouver,BC, Canada. 5Centre of National Research on Disability and RehabilitationMedicine, School of Psychology and Medicine, University of Queensland,Brisbane, Qld, Australia. 6Department of Epidemiology and PreventativeMedicine, School of Public Health and Preventative Medicine, MonashUniversity, Melbourne, VIC, Australia. 7Centre for Health Economics, MonashUniversity, Melbourne, VIC, Australia. 8Pain Management and ResearchCentre, University of Sydney, Sydney, NSW, Australia. 9Department ofPsychiatry and Behavioral Sciences, Duke University School of Medicine,Durham, NC, USA.Received: 14 June 2012 Accepted: 5 July 2012Published: 24 July 2012References1. Bijlsma JW, Berenbaum F, Lafeber FP: Osteoarthritis: an update withrelevance for clinical practice. Lancet 2011, 377:2115–2126.2. Rosemann T, Laux G, Szecsenyi J: Osteoarthritis: quality of life,comorbidities, medication and health service utilization assessed in alarge sample of primary care patients. J Orthop Surg Res 2007, 2:12.3. Somers TJ, Keefe FJ, Godiwala N, Hoyler GH: Psychosocial factors and thepain experience of osteoarthritis patients: new findings and newdirections. Curr Opin Rheumatol 2009, 21:501–506.4. Access Economics: Painful Realities: The economic impact of arthritis inAustralia. Canberra: Access Economics Pty, Ltd; 2007:1–90.5. Ettinger WH, Davis MA, Neuhaus JM, Mallon KP: Long-term physicalfunctioning in persons with knee osteoarthritis from NHANES. I: Effects ofcomorbid medical conditions. J Clin Epidemiol 1994, 47:809–815.6. Dieppe PA, Lohmander LS: Pathogenesis and management of pain inosteoarthritis. Lancet 2005, 365:965–973.7. Davis MA, Ettinger WH, Neuhaus JM, Mallon KP: Knee osteoarthritis andphysical functioning: evidence from the NHANES I EpidemiologicFollowup Study. J Rheumatol 1991, 18:591–598.8. Sayre EC, Li LC, Kopec JA, Esdaile JM, Bar S, Cibere J: The effect of diseasesite (knee, hip, hand, foot, lower back or neck) on employmentreduction due to osteoarthritis. PLoS One 2010, 5:e10470.9. Creamer P, Hochberg MC: The relationship between psychosocialvariables and pain reporting in osteoarthritis of the knee. Arthritis CareRes 1998, 11:60–65.10. Sharma L, Cahue S, Song J, Hayes K, Pai YC, Dunlop D: Physical functioningover three years in knee osteoarthritis: role of psychosocial, localmechanical, and neuromuscular factors. Arthritis Rheum 2003,48:3359–3370.11. Dieppe PA: Relationship between symptoms and structural change inosteoarthritis: what are the important targets for therapy? J Rheumatol2005, 32:1147–1149.12. Eitzen I, Holm I, Risberg MA: Preoperative quadriceps strength is asignificant predictor of knee function two years after anterior cruciateligament reconstruction. Br J Sports Med 2009, 43:371–376.13. Somers TJ, Keefe FJ, Pells JJ, Dixon KE, Waters SJ, Riordan PA, Blumenthal JA,McKee DC, LaCaille L, Tucker JM, Schmitt D, Caldwell DS, Kraus VB, Sims EL,Shelby RA, Rice JR: Pain catastrophizing and pain-related fear inosteoarthritis patients: relationships to pain and disability. J PainSymptom Manage 2009, 37:863–872.14. van Baar ME, Dekker J, Lemmens J, Oostendorp RA, Bijlsma J: Pain anddisability in patients with osteoarthritis of hip or knee: the relationshipwith articular, kinesiological and psychological characteristics.J Rheumatol 1998, 25:125–133.15. Smith BW, Zautra AJ: The effects of anxiety and depression on weeklypain in women with arthritis. Pain 2008, 138:354–361.16. Rosemann T, Laux G, Szecenyi J, Wensing M, Grol R: Pain and osteoarthritisin primary care: factors associated with pain perception in a sample of1,021 patients. Pain Med 2008, 9:903–910.17. Sale JE, Gignac M, Hawker G: The relationship between diseasesymptoms, life events, coping and treatment, and depression amongolder adults with osteoarthritis. J Rheumatol 2008, 35:335–342.18. Roddy E, Zhang W, Doherty M, Arden N, Barlow J, Birrell F, Carr A,Chakravarty K, Dickson J, Hay E, Hosie G, Hurley M, Jordan K, McCarthy C,McMurdo M, Mockett S, O'Reilly S, Peat G, Pendleton A, Richards S:Evidence-based recommendations for the role of exercise in theBennell et al. BMC Musculoskeletal Disorders 2012, 13:129 Page 15 of 17http://www.biomedcentral.com/1471-2474/13/129management of osteoarthritis of the hip or knee- the MOVE consensus.Rheumatology 2005, 44:67–73.19. Zhang W, Moskowitz R, Nuki G, Abramson S, Altman R, Arden N, Bierma-Zeinstra S, Brandt K, Croft P, Doherty M, Dougados M, Hochberg M, HunterD, Kwoh K, Lohmander L, Tugwell P: OARSI recommendations for themanagement of hip and knee osteoarthritis, Part II: OARSI evidence-based, expert consensus guidelines. Osteoarthritis and Cartilage 2008,16:137–162.20. National Institute of Health and Clinical Excellence (NICE) Guidelines:London: The care and management of osteoarthritis in adults 2008 NationalInstitute of Health and Clinical Excellence: 1–25.21. Fransen M, McConnell S: Land-based exercise for osteoarthritis of theknee: a metaanalysis of randomized controlled trials. J Rheumatol 2009,36(6):1109–1117.22. Lange AK, Vanwanseele B, Fiatarone Singh MA: Strength training fortreatment of osteoarthritis of the knee: a systematic review. ArthritisRheum 2008, 59:1488–1494.23. Woodard CM, Berry MJ: Enhancing adherence to prescribed exercise:structured behavioral interventions in clinical exercise programs.J Cardiopulm Rehabil 2001, 21:201–209.24. Keefe FJ, Blumenthal J, Baucom D, Affleck G, Waugh R, Caldwell DS, BeaupreP, Kashikar-Zuck S, Wright K, Egert J, Lefebvre J: Effects of spouse-assistedcoping skills training and exercise training in patients with osteoarthriticknee pain: a randomized controlled study. Pain 2004,110:539–549.25. Schilke JM, Johnson GO, Housh TJ, O'Dell JR: Effects of muscle-strengthtraining on the functional status of patients with osteoarthritis of theknee joint. Nurs Res 1996, 45:68–72.26. Thomas KS, Muir KR, Doherty M, Jones AC, O'Reilly SC, Bassey EJ: Homebased exercise programme for knee pain and knee osteoarthritis:randomised controlled trial. BMJ 2002, 325:752.27. O'Reilly SC, Muir KR, Doherty M: Effectiveness of home exercise on painand disability from osteoarthritis of the knee: a randomised controlledtrial. Ann Rheum Dis 1999, 58:15–19.28. Focht BC: Effectiveness of exercise interventions in reducing painsymptoms among older adults with knee osteoarthritis: a review.J Aging Phys Act 2006, 14:212–235.29. Smeets RJ, Vlaeyen JW, Kester AD, Knottnerus JA: Reduction of paincatastrophizing mediates the outcome of both physical and cognitive-behavioral treatment in chronic low back pain. J Pain 2006, 7:261–271.30. van Koulil S, van Lankveld W, Kraaimaat FW, van Helmond T, Vedder A, vanHoorn H, et al: Tailored cognitive-behavioral therapy and exercisetraining for high-risk patients with fibromyalgia. Arth Care Research 2010,62:1377–1385.31. Scully D, Kremer J, Meade MM, Graham R, Dudgeon K: Physical exerciseand psychological well being: a critical review. Br J Sports Med 1998,32:111–120.32. Korstjens I, May AM, van Weert E, Mesters I, Tan F, Ros WJ, Hoekstra-Weebers JE, van der Schans CP, van den Borne B: Quality of life after self-management cancer rehabilitation: a randomized controlled trialcomparing physical and cognitive-behavioral training versus physicaltraining. Psychosom Med 2008, 70:422–429.33. May AM, Van Weert E, Korstjens I, Hoekstra-Weebers JE, Van Der Schans CP,Zonderland ML, Mesters I, Van Den Borne B, Ros WD: Improved physicalfitness of cancer survivors: a randomised controlled trial comparingphysical training with physical and cognitive-behavioural training. ActaOncol 2008, 47:825–834.34. Keefe FJ, Caldwell DS, Baucom D, Salley A, Robinson E, Timmons K, BeaupreP, Weisberg J, Helms M: Spouse-assisted coping skills training in themanagement of knee pain in osteoarthritis: long-term followup results.Arthritis Care Res 1999, 12:101–111.35. Keefe FJ, Caldwell DS, Williams D, Gil K, Mitchell D, Robertson C: Paincoping skills in the management of osteoarthritis knee pain:A comparative study. Behavior Therapy 1990, 21:49–62.36. Turk DC: Biopsychosocial perspective on chronic pain. New York: GuilfordPress; 1996.37. Hurley MV, Walsh NE, Mitchell HL, Pimm TJ, Patel A, Williamson E, Jones RH,Dieppe PA, Reeves BC: Clinical effectiveness of a rehabilitation programintegrating exercise, self-management, and active coping strategies forchronic knee pain: a cluster randomized trial. Arthritis Rheum 2007,57:1211–1219.38. Pengel LH, Refshauge KM, Maher CG, Nicholas MK, Herbert RD, McNair P:Physiotherapist-directed exercise, advice, or both for subacute low backpain: a randomized trial. Ann Intern Med 2007, 146:787–796.39. Boutron I, Moher D, Altman DG, Schulz KF, Ravaud P: Extending theCONSORT statement to randomized trials of nonpharmacologictreatment: explanation and elaboration. Ann Intern Med 2008,148:295–309.40. Altman R, Asch E, Bloch D, Bole G, Borenstein D, Brandt K, Christy W, CookeTD, Greenwald R, Hochberg M, et al: Development of criteria for theclassification and reporting of osteoarthritis. Classification ofosteoarthritis of the knee. Diagnostic and Therapeutic CriteriaCommittee of the American Rheumatism Association. Arthritis Rheum1986, 29:1039–1049.41. Kellgren JH, Lawrence JS: Radiological assessment of osteo-arthrosis.Ann Rheum Dis 1957, 16:494–502.42. Fransen M, McConnell S, Bell M: Exercise for osteoarthritis of the hip orknee. Cochrane Database Syst Rev 2003, :CD004286.43. Day ML, McGuigan MR, Brice G, Foster C: Monitoring work intensitiesduring resistance training using a session RPE scale. J Strength Cond Res2004, 18:358–358.44. Bellamy N: Osteoarthritis clinical trials: candidate variables and clinimetricproperties. J Rheumatol 1997, 24:768–778.45. McConnell S, Kolopack P, Davis AM: The Western Ontario and McMasterUniversities Osteoarthritis Index (WOMAC): a review of its utility andmeasurement properties. Arthritis Rheum 2001, 45:453–461.46. ten Klooster PM, Drossaers-Bakker KW, Taal E, van de Laar MA: Patient-perceived satisfactory improvement (PPSI): interpreting meaningfulchange in pain from the patient's perspective. Pain 2006,121:151–157.47. Pua YH, Wrigley TV, Cowan SM, Bennell KL: Intrarater test-retest reliabilityof hip range of motion and hip muscle strength measurements inpersons with hip osteoarthritis. Arch Phys Med Rehabil 2008, 89:1146–1154.48. Newcomer KL, Krug HE, Mahowald ML: Validity and reliability of thetimed-stands test for patients with rheumatoid arthritis and otherchronic diseases. J Rheumatol 1993, 20:21–27.49. Dunlop DD, Semanik P, Song J, Sharma L, Nevitt M, Jackson R, Mysiw J,Chang RW: Moving to maintain function in knee osteoarthritis: evidencefrom the osteoarthritis initiative. Arch Phys Med Rehabil 2010, 91:714–721.50. Hill KD, Bernhardt J, McGann AM, Maltese D, Berkovits D: A new test ofdynamic standing balance for stroke patients: reliability, validity andcomparison with healthy elderly. Physiotherapy Canada 1996, 48:257–262.51. Podsiadlo D, Richardson S: The timed "Up & Go": a test of basic functionalmobility for frail elderly persons. J Am Geriatr Soc 1991, 39:142–148.52. Martin KA, Rejeski WJ, Miller ME, James MK, Ettinger WH Jr, Messier SP:Validation of the PASE in older adults with knee pain and physicaldisability. Med Sci Sports Exerc 1999, 31:627–633.53. Tudor-Locke C, Bassett DR Jr: How many steps/day are enough?Preliminary pedometer indices for public health. Sports Med 2004, 34:1–8.54. Richardson J, Peacock S, Iezzi A, Day NA, Hawthorne G: The Assessment ofQuality of Life (AQoL) II Instrument Melbourne.: Monash University; 2007.55. Meenan RF, Mason JH, Anderson JJ, Guccione AA, Kazis LE: AIMS2. Thecontent and properties of a revised and expanded Arthritis ImpactMeasurement Scales Health Status Questionnaire. Arthritis Rheum 1992,35:1–10.56. Lorig K, Chastain RL, Ung E, Shoor S, Holman HR: Development andevaluation of a scale to measure perceived self-efficacy in people witharthritis. Arthritis Rheum 1989, 32:37–44.57. Porter LS, Keefe FJ, Wellington C, De Williams A: Pain communication inthe context of osteoarthritis: patient and partner self-efficacy for paincommunication and holding back from discussion of pain and arthritis-related concerns. Clin J Pain 2008, 24:662–668.58. Tonkin L: The pain self-efficacy questionnaire - Commentary. AustralianJournal of Physiotherapy 2008, 54:77–77.59. Nicholas MK: The pain self-efficacy questionnaire: Taking pain intoaccount. Eur J Pain 2007, 11:153–163.60. Osman A, Barrios FX, Gutierrez PM, Kopper BA, Merrifield T, Grittmann L: ThePain Catastrophizing Scale: further psychometric evaluation with adultsamples. J Behav Med 2000, 23(4):351–365.61. Rosenstiel AK, Keefe FJ: The Use of Coping Strategies in Chronic Low-Back-Pain Patients - Relationship to Patient Characteristics and CurrentAdjustment. Pain 1983, 17:33–44.Bennell et al. BMC Musculoskeletal Disorders 2012, 13:129 Page 16 of 17http://www.biomedcentral.com/1471-2474/13/12962. Harland NJ, Georgieff K: Development of the coping strategiesquestionnaire 24, a clinically utilitarian version of the coping strategiesquestionnaire. Rehabilitation Psychology 2003, 48:296–300.63. Lovibond SH, Lovibond PF: Manual for the Depression Anxiety Stress Scales(2nd Ed). Sydney, Australia: Psychology Foundation; 1995.64. Henry JD, Crawford JR: The short-form version of the Depression AnxietyStress Scales (DASS-21): construct validity and normative data in a largenon-clinical sample. Br J Clin Psychol 2005, 44:227–239.65. Snell WE: Development of the emotional self-disclosure scale. Sex Roles1988, 18:59–67.66. Pistrang N, Barker C: The partner relationship in psychological responseto breast cancer. Soc Sci Med 1995, 40:789–797.67. Figueiredo MI, Fries E, Ingram KM: The role of disclosure patterns andunsupportive social interactions in the well-being of breast cancerpatients. Psychooncology 2004, 13:96–105.68. Porter LS, Keefe FJ, Hurwitz H, Faber M: Disclosure between patients withgastrointestinal cancer and their spouses. Psychooncology 2005,14:1030–1042.69. Porter LS, Keefe FJ, Baucom DH, Hurwitz H, Moser B, Patterson E, Kim HJ:Partner-assisted emotional disclosure for patients with gastrointestinalcancer: results from a randomized controlled trial. Cancer 2009,115:4326–4338.70. Kroenke K, Spitzer RL, Williams JB: The PHQ-9: validity of a briefdepression severity measure. J Gen Intern Med 2001, 16:606–613.71. Sallis JF, Hovell MF, Hofstetter CR, Faucher P, Elder JP, Blanchard J,Caspersen CJ, Powell KE, Christenson GM: A multivariate study ofdeterminants of vigorous exercise in a community sample. Prev Med1989, 18:20–34.72. Borkovec T, Nau S: Credibility of analogue therapy rationales. Journal ofBehavioral Therapy and Experimental Psychiatry 1972,3:257–260.73. Angst F, Aeschlimann A, Stucki G: Smallest detectable and minimalclinically important differences of rehabilitation intervention with theirimplications for required sample sizes using WOMAC and SF-36 qualityof life measurement instruments in patients with osteoarthritis of thelower extremities. Arthritis Rheum 2001, 45:384–391.74. Lim BW, Hinman RS, Wrigley TV, Sharma L, Bennell KL: Does kneemalalignment mediate the effects of quadriceps strengthening on kneeadduction moment, pain, and function in medial knee osteoarthritis?A randomized controlled trial. Arthritis & Rheumatism-Arthritis Care & Research2008, 59:943–951.75. Pocock SJ, Assmann SE, Enos LE, Kasten LE: Subgroup analysis, covariateadjustment and baseline comparisons in clinical trial reporting: currentpractice and problems. Stat Med 2002, 21:2917–2930.76. Fenwick E, Marshall DA, Levy AR, Nichol G: Using and interpreting cost-effectiveness acceptability curves: an example using data from a trial ofmanagement strategies for atrial fibrillation. BMC Health Serv Res2006, 6:52.77. Hoch JS, Briggs AH, Willan AR: Something old, something new, somethingborrowed, something blue: a framework for the marriage of healtheconometrics and cost-effectiveness analysis. Health Econ 2002,11:415–430.78. van Baar ME, Assendelft WJ, Dekker J, Oostendorp RA, Bijlsma JW:Effectiveness of exercise therapy in patients with osteoarthritis of thehip or knee: a systematic review of randomized clinical trials. ArthritisRheum 1999, 42:1361–1369.79. Dixon KE, Keefe FJ, Scipio CD, Perri LM, Abernethy AP: Psychologicalinterventions for arthritis pain management in adults: a meta-analysis.Health Psychol 2007, 26:241–250.80. Steultjens MP, Dekker J, Bijlsma JW: Coping, pain, and disability inosteoarthritis: a longitudinal study. J Rheumatol 2001,28:1068–1072.81. Keefe FJ, Caldwell DS: Cognitive behavioral control of arthritis pain.Med Clin North Am 1997, 81:277–290.82. Keefe FJ, Caldwell DS, Queen KT, Gil KM, Martinez S, Crisson JE, Ogden W,Nunley J: Pain coping strategies in osteoarthritis patients. J Consult ClinPsychol 1987, 55:208–212.83. Keefe FJ, Caldwell DS, Queen K, Gil KM, Martinez S, Crisson JE, et al:Osteoarthritis knee pain: a behavioural analysis. Pain 1987, 28:309–321.84. Penninx BW, Rejeski WJ, Pandya J, Miller ME, Di Bari M, Applegate WB, PahorM: Exercise and depressive symptoms: a comparison of aerobic andresistance exercise effects on emotional and physical function in olderpersons with high and low depressive symptomatology. J Gerontol BPsychol Sci Soc Sci 2002, 57:P124–P132.85. Harrison AL: The influence of pathology, pain, balance, and self-efficacyon function in women with osteoarthritis of the knee. Phys Ther 2004,84:822–831.doi:10.1186/1471-2474-13-129Cite this article as: Bennell et al.: A physiotherapist-delivered integratedexercise and pain coping skills training intervention for individuals withknee osteoarthritis: a randomised controlled trial protocol. BMCMusculoskeletal Disorders 2012 13:129.Submit your next manuscript to BioMed Centraland take full advantage of: • Convenient online submission• Thorough peer review• No space constraints or color figure charges• Immediate publication on acceptance• Inclusion in PubMed, CAS, Scopus and Google Scholar• Research which is freely available for redistributionSubmit your manuscript at www.biomedcentral.com/submitBennell et al. BMC Musculoskeletal Disorders 2012, 13:129 Page 17 of 17http://www.biomedcentral.com/1471-2474/13/129


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items