BackgroundIncreasing demand for hospital services and staff shortages has led NHS organisations to review workforce configurations. One solution has been to employ physician associates (PAs). PAs are trained over 2 years at postgraduate level to work to a supervising doctor. Little is currently known about the roles and impact of PAs working in hospitals in England.Objectives(1) To investigate the factors influencing the adoption and deployment of PAs within medical and surgical teams in secondary care and (2) to explore the contribution of PAs, including their impact on patient experiences, organisation of services, working practices, professional relationships and service costs, in acute hospital care.MethodsThis was a mixed-methods, multiphase study. It comprised a systematic review, a policy review, national surveys of medical directors and PAs, case studies within six hospitals utilising PAs in England and a pragmatic retrospective record review of patients in emergency departments (EDs) attended by PAs and Foundation Year 2 (FY2) doctors.ResultsThe surveys found that a small but growing number of hospitals employed PAs. From the case study element, it was found that medical and surgical teams mainly used PAs to provide continuity to the inpatient wards. Their continuous presence contributed to smoothing patient flow, accessibility for patients and nurses in communicating with doctors and releasing doctors’ (of all grades) time for more complex patients and for attending to patients in clinic and theatre settings. PAs undertook significant amounts of ward-based clinical administration related to patients’ care. The lack of authority to prescribe or order ionising radiation restricted the extent to which PAs assisted with the doctors’ workloads, although the extent of limitation varied between teams. A few consultants in high-dependency specialties considered that junior doctors fitted their team better. PAs were reported to be safe, as was also identified from the review of ED patient records. A comparison of a random sample of patient records in the ED found no difference in the rate of unplanned return for the same problem between those seen by PAs and those seen by FY2 doctors (odds ratio 1.33, 95% confidence interval 0.69 to 2.57;p = 0.40). In the ED, PAs were also valued for the continuity they brought and, as elsewhere, their input in inducting doctors in training into local clinical and hospital processes. Patients were positive about the care PAs provided, although they were not able to identify what or who a PA was; they simply saw them as part of the medical or surgical team looking after them. Although the inclusion of PAs was thought to reduce the need for more expensive locum junior doctors, the use of PAs was primarily discussed in terms of their contribution to patient safety and patient experience in contrast to utilising temporary staff.LimitationsPAs work within medical and surgical teams, such that their specific impact cannot be distinguished from that of the whole team.ConclusionsPAs can provide a flexible advanced clinical practitioner addition to the secondary care workforce without drawing from existing professions. However, their utility in the hospital setting is unlikely to be fully realised without the appropriate level of regulation and attendant authority to prescribe medicines and order ionising radiation within their scope of practice.Future researchComparative investigation is required of patient experience, outcomes and service costs in single, secondary care specialties with and without PAs and in comparison with other types of advanced clinical practitioners.Study registrationThe systematic review component of this study is registered as PROSPERO CRD42016032895.FundingThe National Institute for Health Research Health Services and Delivery Research programme.