Open Access Protocol Effectiveness and cost-effectiveness of a novel, group self-management course for adults with chronic musculoskeletal pain: study protocol for a multicentre, randomised controlled trial (COPERS) Dawn Carnes,1 Stephanie JC Taylor,1 Kate Homer,1 Sandra Eldridge,2 Stephen Bremner,1 Tamar Pincus,2 Anisur Rahman,3 Martin Underwood4 Tocite:CarnesD, ABSTRACT ARTICLE SUMMARY TaylorSJC,HomerK,etal. Introduction:Chronic musculoskeletal pain is a Effectivenessandcost- common condition that often responds poorly to Article focus effectivenessofanovel, groupself-management treatment. Self-management courses have been ▪ We outline all procedures for a pragmatic trial courseforadultswith advocated as a non-drug pain management investigating the effectiveness of a self-manage- chronicmusculoskeletalpain: technique, although evidence for their effectiveness ment intervention for chronic musculoskeletal studyprotocolfora is equivocal. We designed and piloted a pain. multicentre,randomised self-management course based on evidence for controlledtrial(COPERS). effectiveness for specific course components and Key messages BMJOpen2013;3:e002492. characteristics. ▪ Complex interventions need to be based on doi:10.1136/bmjopen-2012- Methods/analysis: COPERS (coping with persistent evidenceandhavegoodtheoreticalunderpinning. 002492 pain, effectiveness research into self-management) is ▪ The value of pilot studies cannot be a pragmatic randomised controlled trial testing the underestimated. ▸ Prepublicationhistoryfor effectiveness and cost-effectiveness of an intensive, ▪ Identifyingpatientswith non-electronically indexed thispaperareavailable group, cognitive behavioural-based, theoretically conditionsisdifficult. online.Toviewthesefiles informed and manualised self-management course Strengths and limitations of this study pleasevisitthejournalonline (http://dx.doi.org/10.1136/ for chronic pain patients against a control of best ▪ Thisisacomplexinterventionwithgoodtheoret- usual care: a pain education booklet and a relaxation bmjopen-2012-002492). icalunderpinnings CD. The course lasts for 15h, spread over 3days, ▪ The trial protocol was informed comprehensive Received17December2012 with a –2h follow-up session 2weeks later. We aim pilotstudy Revised17December2012 to recruit 685 participants with chronic ▪ Difficulties surround the identification patients Accepted20December2012 musculoskeletal pain from primary, intermediate and with non-electronically indexed conditions such secondary care services in two UK regions. The aschronicpain. Thisfinalarticleisavailable study is powered to showa standardised mean foruseunderthetermsof difference of 0.3 in the primary outcome, pain-related theCreativeCommons disability. Secondary outcomes include generic AttributionNon-Commercial BACKGROUND health-related quality of life, healthcare utilisation, 2.0Licence;see Chronic conditions, especially musculoskel- pain self-efficacy, coping, depression, anxiety and http://bmjopen.bmj.com social engagement. Outcomes are measured at 6 and etal conditions, impose an increasing burden 12months postrandomisation. Pain self-efficacy is on healthcare systems and society.1 Point esti- measured at 3months to assess whether change mates of the prevalence of chronic musculo- mediates clinical effect. skeletal pain (pain lasting for longer than Ethics/dissemination: Ethics approval was given the normal soft tissue healing time of by Cambridgeshire Ethics 11/EE/046. This trial will around 12weeks2) range from 46% to 76%.3 provide robust data on the effectiveness and In 2008, the annual report on public health cost-effectiveness of an evidence-based, group of the UK Chief Medical Officer identified self-management programme for chronic chronic pain as a major issue which needed Fornumberedaffiliationssee musculoskeletal pain. The published outcomes will to be addressed.4 Despite increased under- endofarticle. help to inform future policy and practice around such standing of the factors contributing to the self-management courses, both nationally and development of chronic pain,5 there has Correspondenceto internationally. DrDawnCarnes; Trial registration: ISRCTN24426731. been little improvement in how successfully [email protected] it is treated and managed.6 Current CarnesD,TaylorSJC,HomerK,etal.BMJOpen2013;3:e002492.doi:10.1136/bmjopen-2012-002492 1 COPERS protocol treatment tends to centre on drug regimens for pain related to improving posture and social activities, for control and depression, physiotherapy and pain- example, art and massage. Self-efficacy has been management programmes. However, many patients will reported as an important mediator for success in these experience complex care pathways involving multiple types of ‘treatment’ approaches.21 Since healthcare pro- referrals to secondary care and often unnecessary and fessionals and lay tutors have both been reported as repeated diagnostic tests.7 effective facilitators for self-management programmes, Several UK Department of Health reports have recom- we used both.22 23 There is mixed evidence for the mended self-care for chronic disease as a strategy for optimal duration of self-management courses and ‘dose’ improving the long-term future of health in the UK.8–10 per week. Short weekly sessions over 6weeks or more Since 2000, the UK has invested in the implementation are the usual model, but attendance can be poor.14 of lay-led (ie, peer-led) self-management training However, evidence from mental health research shows courses, the Expert Patients Programme (EPP).10 The that brief/intensive interventions can be effective and available evidence, however, suggests that these courses are often preferred.24 25 The COPERS course is deliv- maynotreducehealthcareresourceuseasexpected11–13 ered as a 3-day intensive course on consecutive days, the and that there are only modest short-term beneficial rationale being to address the high attrition often effects on other outcomes. Very few studies have exam- reported anecdotallyin other programmes.14 inedthelong-termclinicaleffects.13–15 The protocol has been designed as a result of a feasi- In response to the anomaly between continued UK gov- bility study which enabled us to modify our approach to ernment support of self-management programmes and recruitment, intervention and to test a variety of equivocal evidence of their effectiveness, the COPERS outcome measures. Recruitment was problematic in the (coping with persistent pain, effectiveness research into feasibility studyas there is no specific electronic diagnos- self-management) study (http://blizard.qmul.ac.uk/ tic code forchronic pain. We have since piloted a search research-generation/329-copers.html) was commissioned strategy based on repeat prescriptions, ‘Read’ codes and by the National Institute of Health Research (NIHR), as a frequency of visits, which identify many of our target 5-yearprogrammegranttoidentifywhichcomponentsand population. To optimise recruitment, we decided to also characteristics of self-management and pain-management use face-to-face consultations and self-referral via adver- programmes are the most effective, as well as to evaluate tisements in surgeries and clinics. We found that filling these components in a trial, in order to inform both the courses within a set time period was difficult especially development of a new pain self-management course and when, by chance, the majority of participants were ran- the next generation of self-management courses in domised to the control. We have accommodated this by general.Wecompletedtwosystematicreviewsandaqualita- using randomly varied permuted block randomisation tive study that informed the development of the COPERS and phasing the recruitment of patients to specific time intervention. One review collated the evidence base periods between schoolholidays. for components and characteristics of pain and self- Weconvenedafocusgroupconsistingoftwoexpertsin management courses and the other collated the evidence outcome measures, two general practitioners (GPs), two for predictors, mediators and moderators of outcomes in psychologists and two patients with chronic pain. This these programmes.16 17 In a qualitative study, we explored focus group selected the most appropriate measures to the views and experiences of participants and tutors from test in the pilot trial. In addition to piloting quantitative self-management courses about expectations, course datacollection,weevaluatedtheinterventionthoroughly content,recruitment,tutoringandattendance.18Thefind- duringthepilotstudyusingfeedbackfromobservers,par- ings from this work suggested that the group-delivered ticipants and facilitators. The pilot study indicated that courses led by healthcare professionals and/or lay people theCOPERSself-managementcoursewasviableandwell weremostlikelytobeeffective.Wealsofoundthatcourses received by participants, with good attendance (85% full lasting for a shorter time appeared to be as effective as attendance). We found that the qualityof facilitation was longer courses and that the setting in which courses were a key determinant for the success of a course and delivered (ie, in healthcare settings or the community) invested more time and resources in facilitator training appearedtomakelittledifferencetoanybeneficialeffects. and recruitment for the main trial. Additionally, we Evidence also suggested that participants particularly found that the group nature of the course was valued enjoyedtherelaxationelementsofcoursesbutwereaverse highly by participants; therefore, we allowed for more toexercisebeingincluded. socialisingduringtheintervention. Our systematic reviews found evidence to support psy- chological approaches; therefore, we based our course Aims of the study around concepts from cognitive behavioural therapyand The aim of this study isto establish the effectiveness and acceptance and commitment therapy.19 20 Some ele- cost-effectivenessofthenewself-management (COPERS) ments of the COPERS intervention are similar to the courseforthosewithchronicmusculoskeletalpain,when EPP, such as the inclusion of relaxation, goal-setting and compared with usual care plus a CD recording of simple action-planning and improving consultations with relaxation exercises (also received by the intervention healthcare professionals. We also included sessions armviatheself-managementcourse). 2 CarnesD,TaylorSJC,HomerK,etal.BMJOpen2013;3:e002492.doi:10.1136/bmjopen-2012-002492 COPERS protocol Trial purpose attendance in the previous 3-month period. During pilot This trial will provide the data needed by the National testing, these searches identified 5–7% of general practi- Health Service (NHS) purchasers to decide whether or tioners as possibly eligible and roughly identified our not tofund programmes ofthis nature. targetpopulationingeneralpractices. The list of potential participants identified via elec- tronic search is screened by clinicians at the practice or METHODS/DESIGN clinic to check suitability based on the eligibility criteria. Trial design Eligible patients are invited to join the study by letter A multicentre pragmatic randomised controlled trial from their GP or secondary care clinic. They are asked (RCT) was conducted in the UK with unbalanced ran- to complete an expression of interest form to send to domisation (1.33: 1) (intervention:control) to accommo- the study team or to contact the study team directly. date for the effects of clustering. The unbalanced Clinicians working in participating practices are also randomisation may also improve recruitment; if partici- able to invite patients consulting with chronic pain to pants are told that they are more likely to receive the participate by handing them an information pack. active treatment, then they may be more likely to join Finally, advertisements and invitation packs will be the study. placed in general practice and clinic waiting rooms to attract interested patients, who can contact the study Trial setting teamdirectlyfor further information. The trial is based in northeast London, and in the Coventry and Warwickshire region. Participants are Recruitmentand informed consent recruited from primary and secondary care settings, People expressing an interest in the study, who are eli- such as general practices, musculoskeletal physiotherapy gible and available to participate, are sent a baseline unitsand pain clinics. The population ofthesetwo local- questionnaire, a trial consent form to complete and a ities, taken together, is broadly representative of the UK freepost envelope to return the questionnaire and trial as awhole. consent form to the study team. All participants are telephoned and asked whether they need more infor- Target population mation, and the consent form is discussed before a Inclusion criteria member of the study team counter signs the consent ▸ People aged 18 or over with chronic musculoskeletal form indicating informed consent. A summaryof partici- pain of more than 3months.2 This includes pain pant flow through the study is shown infigure 1. from osteoarthritis and fibromyalgia or chronic wide- spread pain. Intervention group Exclusion criteria The intervention is a group-based, facilitated learning ▸ Inability togiveinformed consent course supporting patientsto manage theirchronic pain ▸ Not fluent in English better. It is a complex intervention using psychological ▸ Chronic pain arising from activemalignant disease approaches which have been shown to promote behav- ▸ Pain from inflammatory arthritis, such as rheumatoid iourchange of proven benefit in low back pain.26–28 The arthritis course also includes pain education, attention control, ▸ Serious activecomorbidityor terminalillness relaxationandvisualisationtechniques, socialinteraction ▸ Serious mental health or substance abuse issues and the opportunity to try a new activity unrelated to which render the individual unable to attend the pain (table 1). Teaching and learning methods include course. facilitated learning through group discussion, sharing narratives and experiences to foster social interaction, Participant recruitment problem solving, an educational DVD, role play, practis- Participants are recruited from primary and secondary ing attention and distraction techniques and ‘good’ careusing threemethods. posture. We hypothesise that these group-delivered Our mainmethod of recruitment willbe bulk mail outs courses with plenty of breaks for social interaction will from targeted searches of the secondary care clinic and build confidence to manage chronic pain better, that is, general practice electronic records. Electronic records are improve self-efficacy, improve social integration and, searchedbyidentified keyliaisonstaff inthepracticesand where necessary, help re-activate participants to clinics.Thesepersonneluseagenericsearchstrategydevel- re-engage in social activities.29 30 The intervention con- opedandtestedbythestudyteam.Thesearchstrategyhas sists of a short course run in 1weekover 3days (typically a tiered approach focusing on demographic data (patient Monday, Wednesday and Friday) during schoolhours registered at the practice, alive and >18years old), repeat (10:00—14:45), with a 2h follow-up session 2weeks prescriptions for analgesics, tricyclic antidepressants and later. The total course comprises 24 individual ‘compo- anxiolytics, followed by clinical symptoms (eg, low-back nents’ spread over 15–16h. A detailed training manual pain, back pain, osteoarthritis, fibromyalgia) and for facilitators was developed, tested and refined in the CarnesD,TaylorSJC,HomerK,etal.BMJOpen2013;3:e002492.doi:10.1136/bmjopen-2012-002492 3 COPERS protocol Healthcare professional facilitators are recruited via press releases about the study in professional magazines. Lay facilitators are recruited via local contacts with com- munity interest companies delivering self-management type programmes (typically the Expert Patient Programme, http://www.expertpatients.co.uk). Trainingfacilitators Facilitators attend a 2-day training course that covers the course content, how to facilitate, dealing with difficult situations and what to do if an adverse event occurs. During the courses, the trainee facilitators are required to demonstrate actively that they are good listeners, empathic, flexible, able to encourage equal participation and laughter, able to manage difficult people, and able to introduce, lead and summarise sessions while putting the course content into a chronic pain context. We also assess their understanding of the course via a written assessment at the end ofthe training. Only facilitators who are assessed as competent (ie, demonstrate the skills required during role plays and provide written answers that illustrate an understanding of the key concepts) will deliver the intervention. They are to be paired with trained and experienced facilita- tors from the pilot study for their first course to ensure consistency of delivery and to promote confidence in the newly trained facilitators. Control group In our pilot study, we found that a usual care control arm was insufficient to encourage participation in the trial. Our interviews and pilot studies suggested that par- ticipants particularly enjoyed the relaxation part of the courses, while our systematic review found little evidence that the relaxation component of pain-management courses had much observable beneficial effect. In this trial, we include a ‘relaxation pack’ as the control inter- vention to make participation more appealing. This includes a simple audio CD of three breathing and Figure1 Flowdiagramofthestudy. relaxation sequences, instructions and the rationale behind them. The CD is also used and distributed on the COPERS course. After randomisation, participants pilot study. The manual provides the aims and theoret- in the control group receive the relaxation pack and are ical basisforeach component. asked to practise relaxation at least once a day every day In the absence of any evidence favouring a particular for 3weeks (the same duration as the intervention) and setting for such group self-management courses, courses as oftenasthey likethereafter. are held in local medical or community venues that are In an attempt to ensure that all the study participants accessible to participants (eg, with disabled parking and receive best usual care, they are also given a copy of a near to public transport). booklet, called the ‘The Pain Toolkit’,31 which is pub- lished by the UK Department of Health and freely dis- Recruitmentof facilitators for the intervention tributed to UK GPs to give to any patient with chronic Each course is led by two facilitators, a healthcare pain. professional (a professionally registered physiotherapist, osteopath, chiropractor, occupational health practitioner or psychologist) and a lay person with experience of Hypothesis both living with chronic pain and of tutoring or facilitat- We hypothesised that the benefits of the intervention ingsmall groups. will be improved function, greater self-confidence, 4 CarnesD,TaylorSJC,HomerK,etal.BMJOpen2013;3:e002492.doi:10.1136/bmjopen-2012-002492 COPERS protocol Table1 Outlineofthecourseandtheoreticalmodels Day Sessions Moduleaims Theory 1 1:Introduction Understandingpainandacceptance AcceptanceandCommitment 2:PaineducationDVD Therapy1945 3:Acceptance:theuninvited guest Lunchbreak Tasteractivity(eg,art) Distractfrompainperceptionwithphysicalactivity Attentionanddistraction 4:Pain,whenisitbearable Painisnotjustphysiological;itisapsychological, Biopsychosocialmodelof andwhenisitnot? emotionalandsocialphenomenon medicine46 5:Thepaincycle Recognisingthepaincycleandsignpostingways Fearavoidanceand out catastrophising47 6:Movementandposture Reducemuscletensiontoeasepainandbecome Physicaltherapyprinciples 7:Breathingandrelaxation awareofphysicalweaknessandstrengthenareas andAlexandertechnique, (CDtrack1) biofeedback 2 8:Reflectionsfromdayone Improvesocialbonding,groupcohesionand Socialcognitivetheory,3048 communitysocialsupport sociallearningtheory29 9:Identifyingproblems,goal Recognisingerrorsinthinkinginordertopromote Cognitivetherapy,2049 settingandactionplanning aconstructive/rationalviewofasituation theoriesofreasonedaction 10:Barrierstochange— andplannedbehaviour,50–52 unhelpfulthinking rationalemotivetherapy53 Lunchbreak Tasteractivity(eg,,art) Distractfrompainperceptionwithphysicalactivity Attentionanddistraction 11:Barrierstochange— Recognisingerrorsinthinkinginordertopromote Cognitivetherapy2049and reframingnegativesto aconstructive/rationalviewofasituation theoriesofreasonedaction positives andplannedbehaviour50–52 12:Attentioncontroland Distractfrompainperceptionusingvisualisation Attentioncontrol54 distraction 13:Identifyingthingsthat Reminderstoapplytechniquesascoping Embeddinglearning makepainmoremanageable strategies 14:Movementandbalance Reducemuscletensiontoeasepain,become Physicaltherapyprinciples 15:Breathing,relaxationand awareofphysicalweaknessandstrengthenareas. andattentionmanagement54 visualisation Distractionfrompainperceptionusingvisualisation (CDtrack2) 3 16:Reflectionsfromday2 Improvesocialbonding,groupcohesionand Socialcognitivetheory3048 communitysocialsupport 17:Communicationwith Promoteconstructivehealthcareconsultationsand Theoriesofreasonedaction healthprofessionals effectivecommunication andplannedbehaviour50–52 18:Listeningskills andSocialcognitive theory3048 19:Anger,irritabilityand Recognisingerrorsinthinkinginordertopromote Cognitivetherapy2049 frustration aconstructive/rationalviewofasituation theoriesofreasonedaction andplannedbehaviour50–52 Lunchbreak Tasteractivity(eg,art) Distractfrompainperceptionwithphysicalactivity Attentionmanagement 20:Movementandstretch Reducemuscletensiontoeasepain,become Physicaltherapyprinciples 21:Breathing,relaxationand awareofphysicalweaknessandstrengthenareas. andattentionmanagement54 mindfulness Distractionfrompainperceptionusingmindfulness (CDtrack3) 22:Summingup Reminderstoapplytechniquesascoping Embeddinglearning strategies Follow-up 23:Reflectionsandnarratives Improvesocialbonding,groupcohesionand Socialcognitivetheory3048 communitysocialsupport 24:Managingsetbacks Reminderstoapplytechniquesascoping Embeddinglearning strategies better coping skills, increased social engagement and Outcomes more appropriate healthcare resource use (table 2). For the main trial, we selected a group of measures Ouroutcome measures reflect these domains. based on both the published literature regarding their CarnesD,TaylorSJC,HomerK,etal.BMJOpen2013;3:e002492.doi:10.1136/bmjopen-2012-002492 5 COPERS protocol Table2 Outcomemeasuresandotherdatacollection Follow-up Domain Measures (months)* Painduration Numericalscale 0 Painintensity ChronicPainGrade(painintensitysubscale)33 0,6,12 Paindisability ChronicPainGrade(paindisabilitysubscale)33 0,6,12 QualityofLife EQ-5D(healthutility)55 0,6,12 Self-efficacy PainSelf-EfficacyQuestionnaire56 0,3,6,12 Mood HospitalAnxietyandDepressionScale57 0,6,12 Coping ChronicPainAcceptanceQuestionnaire58 0,6,12 Socialactivity HEIQ(Socialintegrationsubscale)59 0,6,12 GeneralHealth Censusglobalhealthquestion60 0,6,12 Demographics Age,NHSnumber,sex,ethnicity,educationalbackground,employmentstatus, 0 languagefluency,livingarrangements(aloneorwithothers) Economic Healthcareresourceuse:primarycareconsultations,secondarycareconsultations, 0to12 analysis hospitaladmissions,surgeries,imaging,testsandprescriptionsfromgeneralpractice electronicrecords Comorbidities FromgeneralpracticeelectronicrecordsaccordingtotheCumulativeIllnessRating 12 Scale7 *Postrandomisation. validity and reliability and the extent of their use in almost doubled when incentives were not conditional on other studies and on the focus group and pilot study. response (1.71;1.29 to2.26).35 Secondaryoutcomes Primaryoutcome Table 2 shows our secondaryoutcomes. Our primary outcome is pain-related disability (we are not aiming to change pain severity in the short term, as current understanding of the mechanism of chronic Criteriafor withdrawal pain suggests that it is unlikely to improve rapidly).32 To All participants are free to withdraw from the study at measure function, we chose a well-validated tool, the any timewithout having to giveanyexplanation. Chronic Pain Grade (CPG).33 The CPG is made up of two constructs—pain intensityand pain-related disability. Each construct has been validated separately. They are Sample size scored independently and can be combined to produce The sample size calculation was based on detecting a the Chronic Pain Grade.34 Our primaryoutcome is pain- standardised mean difference of 0.3 in pain-related dis- relateddisability.Thismeasurehasthreequestionsabout ability between the intervention and control groups, pain-related function, and the responder rates their cir- with a power of 80% at the 5% significance level. This cumstance on a scale from 0 to 10. The pain-related dis- effect size is commensurate with the largest change seen abilityscoreisthemeanofthethreequestions. in a recent systematic review of expert patient pro- Since there is a paucity of evidence on the long-term grammes,14 and also with the sort of change effected by effectiveness of these types of interventions,14 we will interventions for other chronic pain syndromes, such as measure outcomes by a patient-completed postal ques- low back pain, on any continuous outcome measure.36 tionnaire at baseline (before randomisation) and at 6 A simple sample size calculation indicates that we would and 12months after randomisation. There will be one require data on 350 subjects. We inflated the sample size postal reminder for each follow-up datacollection point, because of the possibility of a ‘clustering’ effect in the after which non-responders will be contacted by phone group intervention arm and chose the ratio between to obtain the primary outcome and quality of life intervention and control participants to increase statis- measures. To encourage completion and return of tical efficiency.37 Using an intracluster correlation coeffi- questionnaires, wewill send participants a £5 ‘high-street cient of 0.1 and assuming, on average, 9 individuals shop’ voucher redeemable in multiple stores on a non- providing data from each group resulted in 480 indivi- conditional basis along with their 6-month and duals being needed with 275 in the intervention group 12-month questionnaires. This expression of appreci- and 205 in the control group (1.33:1 intervention: ation has been shown to improve questionnaire return control). Allowing, conservatively, for a 30% loss to rates. A Cochrane review has found that ‘the odds of follow-up (from an average of 13 individuals recruited response’ were more than doubled when a monetary per group), we sought to randomise 685 participants incentive was used (OR 2.02; 95% CI 1.79 to 2.27) and (391 intervention participants and 294 controls). 6 CarnesD,TaylorSJC,HomerK,etal.BMJOpen2013;3:e002492.doi:10.1136/bmjopen-2012-002492 COPERS protocol Previous research and electronic record searches from Blindingand protection from bias the pilot study indicated that around 5% of adults on Allbaselinedataarecollectedbyself-completedquestion- GP registers consult with chronic musculoskeletal pain. nairepriortorandomisation.Afterallocation,however,it Of these, based on our pilot study, we estimate that 10% is impossible to blind researchers and patients to alloca- may be interested in participating in the trial and tion due to the nature of the intervention and the around half of thesewill berecruited into the trial. unequalrandomisation.Incompliancewithourresearch We estimated that around 80% of the population are ethics committee requirements, GPs are informed of adults over 18years of age (www.statistics.gov.uk 2010). their patient’s enrolment into the trial but not of their This means that to recruit our 685 participants, we allocation. Although participants are free to divulge this needed a population base of around 342000 (342000 informationtotheirprimarycareteam,wefeelthisinfor- registered patients of whom approximately 80% are mation in itself should have little impact on their care. adults (274000). Five per cent of these may have Outcome data are by patient completed postal question- chronic pain (13700), and of these, we estimate that naire. For non-responders, primary outcome data and around 10% may express interest in the study (1370). the Euroqol instrument are collected over the phone by Half of these may be recruited and enrolled (685). researchpersonnelblindtotreatmentallocationwhouse Using an average, total GP list size of 7000, this equates asetscriptaskingparticipantsnottodivulgetheiralloca- to around 49 practices. We estimate that we will recruit tionpriortoaskingthemfortheirdata. between 12 and 14 patients per practice. This may be an overestimate of the number of prac- Data management tices needed as it does not account for participants All data are managed in line with PCTU standard oper- recruited from pain or musculoskeletal physiotherapy ating procedures and subject to review by audit and the clinics or from face-to-face invitations and advertise- Data Monitoring and Ethics Committee (DMEC). All ments within general practices. After signing up to the electronic participant data are stored in encrypted and/ study,generalpracticesaregivenachoiceofrecruitment or password-protected files in a secure environment. A phases corresponding to blocks of prebooked course database has been designed to manage the data input to dates. Recruitment will be split roughly equally between ensure consistencyof practice and coding, with a built-in eachof our studysites. audit trail enabling us to track all entries and changes. Weareaimingto book participants into courses within Regular audit and double-checking of all primary out- 12weeks of randomisation. The minimum course size is comes will be conducted to ensure accuracy of data nine, due to the cost implications of running more input, and a further random 10% of all data entry will courses. If a course is undersubscribed (<9) at the be double-checked. outset, it is either cancelled and those registered offered alternative dates for other courses, or those from other Adverse event reporting courses will be transferred to the more imminent These are reported via the facilitators. Minor adverse courses wherepossible. incidents (eg, a participant being tearful and distressed during a session) are logged and fed back to the study team by the end of the course. Serious adverse events related to the study (eg, extreme distress or expressing Randomisation suicidal thoughts) are reported immediately to the study The randomisation plan was developed by the Pragmatic principal investigator in accordance with good clinical Clinical Trials Unit (PCTU) at Queen Mary, University practice guidelines and, where necessary, reported to of London (http://blizard.qmul.ac.uk/pragmatic- the DMEC and the studysponsor. clinical-trials-unit.html). Randomisation was stratified by centre (the Midlands or London). To ensure that alloca- Statisticalanalysis tion provides sufficient participants for each course, but We will use Stata,37 R38 and MLwiN,39 as appropriate, to which cannot be predicted by researchers, we used ran- analyse the data. Descriptive statistics will be used to domly varied permuted blocks of size 7 and 14 with an summarise the characteristics of participants in each allocation ratio of 1.33:1 (intervention:control). Those arm of the trial. Outcomes at follow-up will be analysed returning completed questionnaires were telephoned as dependent variables in mixed effects regression by the study team. People who were able and willing to models. The model will include as covariates: baseline participate and gave valid informed consent were rando- outcome measure, allocation (intervention or control), mised over the phone. During the phone call, research- age, gender, study centre (London or the Midlands) as ers entered the participant’s trial identification number fixed effects, and the intervention group as a random and recruiting centre into a central randomisation effect. Where participants withdraw, we will compare the service, ensuring allocation concealment and partici- characteristics of those withdrawing against those who pants received immediate notification of theirallocation. remain inthe study. Participants are then informed and either immediately Our primary analysis is an available case analysis booked on acourse or sent arelaxation pack. following the intention-to-treat principles (http:// CarnesD,TaylorSJC,HomerK,etal.BMJOpen2013;3:e002492.doi:10.1136/bmjopen-2012-002492 7 COPERS protocol www.consort-statement.org/consort-statement/further- included in this information. A secondary analysis will explanations/box6_intention-to-treat-analysis/). We will include broader societal costs such as the patients’ use mixed effects models appropriate for each outcome out-of-pocket treatments for their pain such as comple- (linear, logistic, Poisson) with intervention arm, age and mentary therapies, mobility devices, private investigations gender and baseline level of outcome as fixed effects andprivatehospitaladmissions. and group as a random effect. The random effect will The data will be collected from three main sources: only be present in the intervention arm. We will primary care utilisation from GP electronic records at examine patterns of‘missingness’and conductasecond- 12months’ follow-up, SUS data from the local strategic ary analysis using multiple imputation and assuming health bodies at 15months (since there is a 3-month variables are missing at random. We will also conduct a ‘lag’ in the availabilityof SUS data) and patient-reported complier-average causal effect analysis to estimate the cost and utility data from the 6-month and 12-month effect of the intervention on compliers.40 We define questionnaires. Data on the costs of training sessions will ‘compliers’ as those who attend more than half of the be collected from trial records. Primary and intermedi- course (ie, those present for at least 12 of the 24 course ate care unit costs will be derived from UK published components). We are also interested in the dose inter- sources.41 The unit costs for medications will be vention participants receive, and define ‘full’ exposure obtained from the Prescription Cost Analysis database to the intervention as being present at 17– 24 individual for 2010.42 The unit costs for secondary care will be course sessions, moderate exposure as being present at 9 basedon health resource groups.43 –16 components, and non-exposure as being present at If appropriate, we will use multiple imputation to eight or fewer components. We will report effect esti- guard against any bias that may result from missing mates with CIs. For selected outcomes, we will estimate EQ-5D or cost data. The imputation model will impute the proportions who improve over the threshold of 30% missing data based on age, gender, study centre from baseline, a figure commonly used in relation to (London or the Midlands), EQ-5D and cost. The imput- back pain, and from this, we will estimate the numbers ation model for the intervention arm will also account needed to be treated and the relative risks for improve- for clustering by group in the intervention arm (by ment.36 If the analysis shows an effect of the interven- including the intervention group as arandom effect). tion on our primary outcome (pain-related disability), Our primary economic analysis is a cost-utility analysis we will conduct an exploratory analysis to explore over 12months, examining the cost per QALY gained whether any of the following are moderators of the for all participants whowere assessed for EQ-5D prior to effect: baseline duration of chronic pain, severity of randomisation and had SUS data extracts. Descriptive pain-related disability and pain self-efficacy. We will also statistics will be used tosummarise the costs and QALYs. conduct a mediator analysis to determine whether the We will use a mixed effects regression model to adjust level of exposure to the intervention or a positive estimates for costs and QALYs for the following covari- changeinself-efficacyfrom baseline to12weeks postran- ates: baseline outcome measure, allocation (intervention domisation mediatesthe effects at 6 or 12months. or control), age, gender, study centre (London or the Midlands) as fixed effects, and the intervention group as Health economic analysis a random effect. The Incremental Cost Effectiveness At 12months postrandomisation, we will conduct a Ratio (ICER) will be calculated using the QALYs for cost-utility analysis using EQ-5D data to measure Quality each patient as a utility measurement along with the Adjusted Life years (QALYs) from an NHS perspective. coststothe NHS incurred byeach patient, where: Thecostswillcoverprimarycareservices,prescribedmedi- cation, investigations, intermediate care referrals and sec- ondary care (using secondary user services (SUS) data). Primary care services include consultations with GPs and ðCostofintervention(cid:2)CostofcontrolÞ ICER ¼ practice nurses, both scheduled and unscheduled. Utilityofintervention(cid:2)Utilityofcontrol Intermediate care referrals data will be collected on physiotherapy,mentalhealthservices(includingImproved Access to Psychological Therapies referrals), podiatry, community nursing, community rehabilitation and other The spread of the incremental cost-effectiveness ratio unscheduled primary or community care (out-of-hours across the four quadrants of cost effectiveness will be services and walk-in centres). SUS will include accident plotted, and we will use bootstrapping to estimate the and emergency visits, outpatient appointments and CIs of our estimates. We also plan to assess the cost inpatient episodes. The SUS data will have International utilityof the intervention using willingnessto pay thresh- Classification of Disease 10 and Office of Population olds ranging between £0 and £4000044 and to run two Censuses and Surveys codes which will provide informa- sensitivityanalyses: tion about disease and billing data. Information derived 1. Excluding high-cost individuals (top 5%). fromtheOfficeforNationalStatisticsdeathreportswillbe 2. Including costs from the societal perspective (out-of- the filter for attrition by death, as the cause of death is pockettreatments for pain). 8 CarnesD,TaylorSJC,HomerK,etal.BMJOpen2013;3:e002492.doi:10.1136/bmjopen-2012-002492 COPERS protocol Quality control andfidelityof intervention delivery 3. ParsonsS,BreenA,FosterNE,etal.Prevalenceandcomparative The first day of each course is observed, and where we troublesomenessbyageofmusculoskeletalpainindifferentbody locations.FamPract2007;24:308–16. have first time facilitators, or facilitators who lack confi- 4. DonaldsonL.150yearsoftheAnnualReportoftheChiefMedical dence, these courses are observed in their entirety to Officer:onthestateofpublichealth2008.DepartmentofHealth. StationeryOffice;2008.http://www.dh.gov.uk/en/ check the fidelity of the intervention. In addition, every Publicationsandstatistics/Publications/AnnualReports/DH_096206 course is audiorecorded. A random selection of audio- (accessed17Jan2013). taped sessions will be chosen for evaluation of fidelity. 5. WScoieonlfcCeJ2,0S0a0lt;e2r88M:1W7.6N5e–u9r.onalplasticity:increasingthegaininpain. The evaluators will use a checklist to record adherence 6. CroftP.Islifebecomingmoreofapain?BMJ2000;320:1552–3. to structure and content and facilitator competence. 7. HudonC,FortinM,VanasseA.CumulativeIllnessRatingScalewas areliableandvalidindexinafamilypracticecontext.JClin Feedback to facilitators will be provided, where neces- Epidemiol2005;58:603–8. sary,so theycan modifyand improvetheir performance. 8. DarziA.Highqualitycareforall:NHSNextStageReviewfinal report.DepartmentofHealth,StationeryOffice;2008.http://www.dh. gov.uk/en/Publicationsandstatistics/Publications/Publications PolicyAndGuidance/DH_085825(accessed17Jan2013). CONCLUSION 9. WanlessD.Securingourfuturehealth:takingalong-termview—the Thisdefinitivetrialwillproviderobustdataontheeffect- WanlessReport.DepartmentofHealth,StationeryOffice;2002. http://webarchive.nationalarchives.gov.uk/+/http://www.hm-treasury. iveness and cost-effectiveness of an optimised package of gov.uk/consult_wanless_final.htm(accessed17Jan2013). care designed to improve self-management of chronic 10. DonaldsonL.Expertpatientsusherinaneweraofopportunityfor theNHS.BMJ2003;326:1279–80. pain. This will help to serve to inform future policy and 11. DepartmentofHealth.Theexpertpatient:anewapproachtochronic practice for running self-management courses both diseasemanagementforthe21stcentury.DepartmentofHealth, nationallyand internationally. StationeryOffice;2001.http://www.dh.gov.uk/en/Publicationsand statistics/Publications/PublicationsPolicyandGuidance/DH_4006801 (accessed17Jan2013). Trial status 12. Expertpatientprogramme.http://www.expertpatients.co.uk/ (accessed17Jan2013). We anticipate that our results will be complete and sub- 13. GriffithsC,FosterG,RamsayJ,etal. Howeffectiveareexpert mitted for peer review publication inJanuary 2014. patient(layled)educationprogrammesforchronicdisease?BMJ 2007;334:1254–6. 14. FosterG,TaylorSJC,EldridgeSE,etal.Self-management Authoraffiliations educationprogrammesbylayleadersforpeoplewithchronic 1CentreforPrimaryCareandPublicHealth,BlizardInstituteBartsandThe conditions.CochraneDatabase SystRev2007;(4):CD005108. 15. KennedyA,ReevesD,BowerP,etal.Theeffectivenessandcost LondonSchoolofMedicineandDentistry,QueenMaryUniversityofLondon, effectivenessofanationallay-ledselfcaresupportprogrammefor London,UK patientswithlong-termconditions:apragmaticrandomised 2DepartmentofPsychology,RoyalHollowayUniversityofLondon,London, controlledtrial.JEpidemiolCommunityHealth2007;61:254–61. UK 16. MilesCL,PincusT,CarnesD,etal.Canweidentifyhowprogrammes 3DepartmentofRheumatology,UniversityCollegeHospital,London,UK aimedatpromotingself-managementinmusculoskeletalpainwork 4ClinicalTrialsUnit,WarwickMedicalSchool,WarwickUniversity,Coventry, andwhobenefits?Asystematicreviewofsub-groupanalysiswithin RCTs.EurJPain2011;15:775.e1–775.e11. UK 17. CarnesD,HomerKE,MilesCL,etal.Effectivedeliverystylesand Contributors MU,DCandSTconceivedtheoriginalideaforthestudy.ST, contentforself-managementinterventionsforchronicmusculoskeletal pain:asystematicliteraturereview.ClinJPain2012;28:344–54. MU,DC,KH,TPandARwereintegraltothedesignoftheinterventionand 18. CarnesD,HomerKE,UnderwoodM,etal.Self-management thetrial,andSE,SBandMUwerespecificallyresponsibleforthestatistical coursesforthosewithchronicmusculoskeletalpain:whotorefer componentsofthetrialandthestatisticalanalysisprotocol.DCwrotethefirst andwhy?EurJPainSupplements2011;5:70–1. draftoftheprotocol.KH,ST,MU,KH,AR,TPandSEcontributedtoeach 19. HayesSC,StrosahlKD,WilsonKG.Acceptanceandcommitment successivedraftofthemanuscript,andallauthorsapprovedthefinalversion. therapy:anexperientialapproachtobehaviorchange.NewYork, MUandSTarejointPIsforthestudy;DCisthetrialmanagerandKHthe USA:GuildfordPress,2004. 20. BeckJS.Cognitivetherapy.NewYork,USA:TheGuildfordPress, seniorresearcher;SBandSEarethetrialstatisticians.ARandTPareonthe 1995. trialmanagementboardandprovidedsubstantialinputintotheintervention 21. OsborneRH,ElsworthGR,WhitfieldK.TheHealthEducation design. ImpactQuestionnaire(heiQ):anoutcomesandevaluationmeasure forpatienteducationandself-managementinterventionsforpeople Funding Thispaperpresentsindependentresearchcommissionedbythe withchronicconditions.PatientEducCouns2007;66:192–201. NationalInstituteforHealthResearch(NIHR)underitsProgrammeGrantsfor 22. FransenM,McConnellS.Exerciseforosteoarthritisoftheknee. AppliedResearchscheme(RP-PG-0707-10189).Theviewsexpressedinthis CochraneDatabaseSystRev2008(4):CD004376. publicationarethoseoftheauthorsandnotnecessarilythoseoftheNHS,the 23. HaydenJA,vanTulderMW,TomlinsonG.Systematicreview: NIHRortheDepartmentofHealth.Thisprojectbenefitedfromfacilitiesfunded strategiesforusingexercisetherapytoimproveoutcomesinchronic lowbackpain.AnnInternMed2005;142:776–85. throughtheBirminghamScienceCityTranslationalMedicineClinicalResearch 24. WattMC,StewartSH,LefaivreMJ,etal.Abriefcognitive-behavioral andInfrastructureTrialsPlatform,withsupportfromAdvantageWestMidlands. approachtoreducinganxietysensitivitydecreasespain-related anxiety.CognBehavTher2006;35:248–56. Competinginterests None. 25. WisemanCV,SundaySR,KlapperF,etal.Short-termgroupCBT Ethicsapproval CambridgeshireEthicsCommitteeprovidedafavourable versuspsycho-educationonaninpatienteatingdisorderunit.Eat DisordJTreatPrev2002;10:313–20. ethicalreviewforthisstudyRef:11/EE/046.TheISCTRNis24426731 26. LambSE,LallR,HansenZ,etal.Amulticentredrandomised Provenanceandpeerreview Notcommissioned;internallypeerreviewed. controlledtrialofaprimarycare-basedcognitivebehavioural programmeforlowbackpain.TheBackSkillsTraining(BeST)trial. HealthTechnolAssess2010;14:1–253,iii–iv. 27. HenschkeN,OsteloRW,vanTulderMW,etal.Behavioural REFERENCES treatmentforchroniclow-backpain.CochraneDatabaseSystRev 1. ManiadakisN,GrayA.TheeconomicburdenofbackpainintheUK. 2010(7):CD002014. Pain2000;84:95–103. 28. HoffmanBM,PapasRK,ChatkoffDK,etal.Meta-analysisof 2. IASP.Paintermsacurrentlistwithdefinitionsandnotesonusage. psychologicalinterventionsforchroniclowbackpain.HealthPsychol Pain1986;24(Suppl1):S215–21. 2007;26:1–10. CarnesD,TaylorSJC,HomerK,etal.BMJOpen2013;3:e002492.doi:10.1136/bmjopen-2012-002492 9 COPERS protocol 29. MillarNE,DollardJ.Sociallearningandimitation.NewHaven, 45. HayesSC,SmithSX.Getoutofyourmindandintoyourlife:the Connecticut,USA:YaleUniversityPress,1941. newacceptanceandcommitmenttherapy.Oakland,California,USA: 30. BanduraA.Self-efficacy:towardaunifyingtheoryofbehavioral NewHarbingerPublications,2005. change.PsycholRev1977;84:191–215. 46. TurkDC,MonarchES.Psychologicalapproachestopainmanagement: 31. MooreP,ColeF.ThePainToolkit.2009.http://www.paintoolkit.org/ apractitioner’shandbook.2ndedn.NewYork,USA:TheGuilford assets/downloads/Pain-Toolkit-Booklet-Nov-2012.pdf(accessed17 Press,2002. Jan2013). 47. VlaeyenJW,LintonSJ.Fear-avoidanceanditsconsequencesin 32. GwilymSE,PollardTC,CarrAJ.Understandingpainin chronicmusculoskeletalpain:astateoftheart.Pain osteoarthritis.JBoneJointSurgBr2008;90:280–7. 2000;85:317–32. 33. VonKorffM,OrmelJ,KeefeFJ,etal.Gradingtheseverityof 48. BanduraA.SocialFoundationsofthoughtandaction:asocial chronicpain.Pain1992;50:133–49. cognitivetheorypearsoneducation.1986. 34. SmithBH,PennyKI,PurvesAM,etal.TheChronicPainGrade 49. WillsonR,BranchR.Cognitivebehaviouraltherapyfordummies. questionnaire:validationandreliabilityinpostalresearch.Pain Chichester,England,UK:JohnWiley&Sons,2006. 1997;71:141–7. 50. AjzenI,FishbeinM.Belief,attitude,intentionandbehaviour:an 35. EdwardsP,RobertsI,ClarkeM,etal.Increasingresponseratesto introductiontotheoryandresearch(Addison-Wesleyseriesin postalquestionnaires:systematicreview.BMJ2002;324:1183. socialpsychology).Harlow,Essex,UK:LongmanHigherEducation, 36. OsteloRW,DeyoRA,StratfordP,etal.Interpretingchangescoresfor 1976. painandfunctionalstatusinlowbackpain:towardsinternational 51. AjzenI,FishbeinM.UnderstandingAttitudesandPredictingSocial consensusregardingminimalimportantchange.Spine2008;33:90–4. BehaviorPearsonEducation.Harlow,Essex,UK,1980. 37. StataCorp.StataStatisticalSoftware.Release10.1.In.College 52. KuhlJ,BeckmanJ.eds.Actioncontrol:fromcognitiontobehavior Station,TX:StataCorpLP,2007. (springerseriesinsocialpsychology).Berlin,Germany: 38. RDevelopmentCoreTeam.R:ALanguageandEnvironmentfor Springer-VerlagBerlinandHeidelbergGmbH&Co.,1985. StatisticalComputing.In.:RFoundationforStatisticalComputing, 53. EllisA,DrydenW.Thepracticeofrationalemotivebehaviortherapy. Vienna,Austria,2011. 2ndedn.NewYork,USA:SpringerPublishing,2007. 39. RasbashJ,CharltonC,BrowneWJ,etal.MLwiNVersion2.02.In: 54. MorleyS,ShapiroDA,BiggsJ.Developingatreatmentmanualfor CentreforMultilevelModelling,UniversityofBristol,2005. attentionmanagementinchronicpain.CognBehavTher 40. DunnG,MaracyM,DowrickC,etal.Estimatingpsychologicaltreatment 2004;33:1–11. effectsfromarandomisedcontrolledtrialwithbothnon-complianceand 55. EuroQolGroup.TheEuroQolinstrument:EQ-5D.1990.http://www. losstofollow-up.BrJPsychiatry2003;183:323–31. euroqol.org/eq-5d(accessed17Jan2013). 41. CurtisL.UnitCostsofHealthandSocialCare2010.PersonalSocial 56. NicholasMK.Thepainself-efficacyquestionnaire:takingpaininto ServicesResearchUnit(PSSRU),UniversityofKent,2010.https:// account.EurJPain2007;11:153–63. www.education.gov.uk/publications/eOrderingDownload/ 57. ZigmondAS,SnaithRP.Thehospitalanxietyanddepressionscale. PSSRU-1368–230X.pdf(accessed17Jan2013). ActaPsychiatrScand1983;67:361–70. 42. PrescriptionCostAnalysis,England—2010.TheHealthandSocial 58. McCrackenLM,VowlesKE,EcclestonC.Acceptanceofchronic CareInformationCentre,2011.http://www.ic.nhs.uk/pubs/ pain:componentanalysisandarevisedassessmentmethod.Pain prescostanalysis2010(accessed17Jan2013). 2004;107:159–66. 43. DepartmentofHealth:2010–11referencecostspublication.2011. 59. UniversityofMelbourne.HealthEducationImpactQuestionnaire. http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/ http://www.crd.unimelb.edu.au/heiq/(accessed17Jan2013). PublicationsPolicyAndGuidance/DH_131140(accessed17Jan2013). 60. OfficeforNationalStatistics.Census2011.2011. 44. TowseA.ShouldNICE’sthresholdrangeforcostperQALYbe http://www.ons.gov.uk/census/2011-census(accessed raised?Yes.BMJ2009;338:b181. 17Jan2013). 10 CarnesD,TaylorSJC,HomerK,etal.BMJOpen2013;3:e002492.doi:10.1136/bmjopen-2012-002492