Oxford Research Proposal Example: Chiropractic student to musculoskeletal health policy (Score 93)

Musculoskeletal (MSK) conditions account for approximately one in five GP consultations in England and represent the largest single driver of years lived with disability in high-income countries. Despite this burden, the pathway from first presentation to appropriate care remains poorly standardised: patients with low back pain, neck pain, or shoulder disorders may be referred directly to orthopaedic outpatient services, directed to physiotherapy, or managed entirely within primary care, depending on local commissioning decisions rather than clinical evidence. The result is wide geographic variation in secondary-care referral rates that cannot be explained by population morbidity alone. This proposal asks: to what extent do primary-care MSK triage models — specifically, first-contact physiotherapy (FCP) and extended-scope practitioner (ESP) schemes — reduce avoidable orthopaedic outpatient referrals, and does any reduction vary systematically by deprivation quintile across NHS Integrated Care Board (ICB) areas in England? The question is bounded in three ways. It focuses on a single policy instrument (structured MSK triage at primary-care level), a single measurable outcome (orthopaedic outpatient referral rate per 1,000 registered patients), and a single health system (NHS England post-2022 ICB reorganisation). This scope is tractable within a one-year MSc research project and directly relevant to current NHS priorities around elective recovery and MSK pathway redesign. Two bodies of scholarship bear on this question, and they have not been adequately joined. The first is the clinical effectiveness literature on FCP and ESP models. A series of service evaluations and small randomised trials — including work associated with the NHS England FCP pilot programme and published in journals such as Musculoskeletal Science and Practice — has established that physiotherapists operating as first-contact practitioners can safely manage a high proportion of MSK presentations without GP involvement, and that patient satisfaction is comparable to GP-led care. However, most of this evidence is single-site, measures process outcomes (such as the number of GP appointments saved) rather than downstream referral behaviour, and does not disaggregate by socioeconomic context. Whether FCP deployment across multiple ICBs translates into measurable reductions in orthopaedic outpatient demand at the system level remains empirically untested. The second body of literature is the health policy and health economics scholarship on MSK pathway variation. Work drawing on Hospital Episode Statistics and the former Clinical Commissioning Group reference costs has documented substantial unexplained variation in orthopaedic referral rates across English localities. Researchers in health geography and primary care policy have attributed part of this variation to commissioning culture, GP practice list size, and rurality, but the role of structured MSK triage as a policy lever has not been isolated as an explanatory variable in any multi-area regression analysis I have identified. The gap, then, is not a shortage of evidence about FCP effectiveness at the clinical level, nor a shortage of evidence about referral variation at the system level; it is the absence of a study that uses area-level administrative data to test whether ICBs with higher FCP or ESP penetration show lower orthopaedic referral rates, controlling for known confounders. Filling this gap would give commissioners a more defensible quantitative basis for investment decisions than currently exists. I propose a cross-sectional ecological analysis at the ICB level (n = 42 ICBs in England), supplemented by a small number of semi-structured interviews with MSK pathway leads to interpret outlier findings. The primary dataset would be NHS England's Referral to Treatment (RTT) waiting-times data, which reports orthopaedic outpatient referral volumes by ICB and can be used to construct a referral rate per 1,000 weighted registered patients. FCP and ESP deployment data would be drawn from NHS England's primary care workforce statistics and, where available, from ICB annual reports and NHS Digital general practice workforce returns. Deprivation data would come from the 2019 English Indices of Multiple Deprivation aggregated to ICB level using population-weighted means. The main analytical model would be an ordinary least squares regression of the orthopaedic referral rate on FCP/ESP penetration (measured as FCP whole-time equivalents per 100,000 registered patients), with controls for ICB-level deprivation quintile, rurality classification, GP list size, and age-sex standardised MSK prevalence estimated from GP Patient Survey data. An interaction term between FCP penetration and deprivation quintile would test the equity dimension of the research question. Sensitivity analyses would re-run the model using two-year pooled data (2022–23 and 2023–24) to assess stability. The qualitative component — approximately eight to ten interviews with ICB MSK commissioners or FCP clinical leads, sampled purposively to include high-referral and low-referral outlier ICBs — would serve an interpretive rather than confirmatory function. Interview data would be analysed thematically to identify plausible mechanisms behind statistical outliers that the ecological model cannot explain, such as local referral culture or differences in data recording practice. This design is appropriate because the policy question is inherently area-level: commissioners make decisions about FCP investment at the ICB level, not the individual patient level. An ecological design is therefore not a methodological compromise but a match between unit of analysis and unit of policy action. All primary quantitative datasets described above are publicly available under Open Government Licence and require no data-sharing agreement or ethics approval for secondary analysis of aggregate, non-identifiable statistics. The qualitative interviews with NHS commissioners would require NHS Research Ethics Committee review or, depending on classification, Health Research Authority assessment; I have budgeted time for this process in the timeline below and have identified the possibility that some ICBs may decline participation or that interview recruitment may take longer than anticipated. A contingency is to replace the interview component with a systematic review of ICB MSK strategy documents, which are publicly available, if ethics or recruitment timelines prove prohibitive. Provisional timeline (twelve months): Months 1–2, literature review and data assembly; Months 3–4, dataset construction and cleaning, ethics application submission; Months 5–7, regression analysis and sensitivity checks; Months 7–9, interview fieldwork and thematic analysis (subject to ethics approval); Months 10–11, integration of findings and draft write-up; Month 12, revision and submission. The main scope risk is data completeness: FCP workforce data at ICB level is not uniformly reported, and some ICBs may have incomplete returns. I would address this by triangulating NHS Digital workforce data with ICB annual reports and, where necessary, treating ICBs with missing FCP data as a separate category in sensitivity analysis rather than imputing values. Oxford's Nuffield Department of Population Health (NDPH) and the Blavatnik School of Government both host research programmes examining NHS commissioning, health system performance, and the social determinants of health service use — areas directly relevant to this proposal. The Health Economics Research Centre (HERC) at NDPH has particular expertise in the analysis of administrative health data and area-level variation, which aligns with the quantitative core of this project. The Blavatnik School's work on public service delivery and policy implementation is relevant to the qualitative and interpretive component. The proposed analysis requires access to publicly available NHS datasets and standard statistical software (Stata or R), both of which are available through Oxford's research computing environment. No specialist laboratory facilities or proprietary data licences are required, which keeps the project self-contained within the MSc year. I am seeking a supervisor with expertise in health system performance analysis or NHS commissioning policy. I have not approached any specific faculty member and make no claim of agreed supervision; I note that NDPH and the Blavatnik School list researchers whose published work addresses MSK service organisation, primary care workforce policy, and health inequalities, and I would welcome guidance from the admissions team on the most appropriate departmental home for this question.