Stratified Primary Care for Musculoskeletal Pain

The programme has 4 linked Workpackages. Workpackages 1 and 2 run in parallel and inform Workpackages 3 and 4. Workpackage 1 aims to confirm the validity and optimise the predictive performance of the STarT-Musc screening tool and determine the cut-points for each patient risk strata. We plan an observational, prospective cohort of 1800 adults with one or more of the five most common musculoskeletal pain presentations at 8 general practices in the West Midlands North Primary Care Research Network (WMN PCRN). Recruitment uses methods previously shown to be successful, in which an electronic tag is stamped on the medical records of eligible patients based on Read codes entered at the time of consultation, to identify patients for invitation. This ensures that recruitment is not dependent on GPs remembering about the study. Participants will complete baseline, 2 and 6 month questionnaires and are asked to give consent for medical record review. Attrition bias will be minimised by reminders and collection of a minimal data set. Poor outcome is lack of improvement on the SF-36v2 Physical Component Scale (PCS) between baseline and 6 month follow-up. Analyses will determine the discriminant and construct validity of the screening tool, establish risk strata cut-points and describe patients classified at low, medium and high risk of poor outcome. Predictive validity will be optimised using decision analysis, assessing the benefit of additional or alternative variables in the screening tool. Workpackage 2 focuses on specifying the intervention of stratified care. Study 2.1 will provide a detailed understanding of the perceived added value, and anticipated pitfalls and solutions to using a stratified primary care model that is based on prognosis with musculoskeletal pain patients. In 6 to 8 focus groups with GPs, other health professionals and patients, we will follow good qualitative research practice to analyse the data, including independent analysis of transcripts. Study 2.2 is an evidence synthesis of recent guidelines and systematic reviews of treatments that are suitable and available to NHS primary care. We will also examine whether information is available according to patient risk strata. In study 2.3, we will develop consensus about the appropriate matched treatment options for patients in each risk strata. We plan 3 consensus workshops with patients, clinicians, managers and commissioners using nominal group technique. In study 2.4, we will agree the content of the clinician support packages for stratified care delivery and the optimal and most effective ways of delivering the support packages. Workpackage 3 will test the feasibility of delivering stratified care in 2 general practices that participated in Workpackage 1, and is a before/after study. Stratified care is both use of the STarT-Musc tool and matched treatment options. We will train and support the practice teams to deliver stratified care to eligible patients consulting over a 3 month period. We will recruit 250-300 adults to complete a study questionnaire in order to test patient recruitment and data collection. Up to 30 audio-taped GP-patient consultations will be examined to explore the feasibility of our stratified care model in real-time. Two theoretical models will inform the programme as whole, but this study in particular, a model of behaviour change and a model of broader systems change. The quantitative and qualitative data will be analysed iteratively and fed back to the practice teams in workshops. Feasibility will be established if by the end of the 3 month period, the STarT-Musc tool is completed for more than two-thirds of patients coded with musculoskeletal problems and over 50% of patients receive their risk appropriate matched treatment options. Workpackage 4 is a pragmatic, cluster randomised trial with 20 GP practices (10 intervention, 10 usual care control) and 3100 patients, to test the clinical and cost-effectiveness of stratified p