As a result of long delays for physiotherapy for musculoskeletal problems, several areas in the UK have introduced PhysioDirect services in which patients telephone a physiotherapist for initial assessment and treatment advice. However, there is no robust evidence about the effectiveness, cost-effectiveness or acceptability to patients of PhysioDirect. To investigate whether or not PhysioDirect is equally as clinically effective as and more cost-effective than usual care for patients with musculoskeletal (MSK) problems in primary care. Pragmatic randomised controlled trial to assess equivalence, incorporating economic evaluation and nested qualitative research. Patients were randomised in 2 : 1 ratio to PhysioDirect or usual care using a remote automated allocation system at the level of the individual, stratifying by physiotherapy site and minimising by sex, age group and site of MSK problem. For the economic analysis, cost consequences included NHS and patient costs, and the cost of lost production. Cost-effectiveness analysis was carried out from the perspective of the NHS. Interviews were conducted with patients, physiotherapists and their managers. Four community physiotherapy services in England. Adults referred by general practitioners or self-referred for physiotherapy for a MSK problem. Patients allocated to PhysioDirect were invited to telephone a senior physiotherapist for initial assessment and advice using a computerised template, followed by face-to-face care when necessary. Patients allocated to usual care were put on to a waiting list for face-to-face care. Primary outcome was the Short Form questionnaire-36 items, version 2 (SF-36v2) Physical Component Score (PCS) at 6 months after randomisation. Secondary outcomes included other measures of health outcome [Measure Yourself Medical Outcomes Profile, European Quality of Life-5 Dimensions (EuroQol health utility measure, EQ-5D), global improvement, response to treatment], wait for treatment, time lost from work and usual activities, patient satisfaction. Data were collected by postal questionnaires at baseline, 6 weeks and 6 months, and from routine records by researchers blind to allocation. A total of 1506 patients were allocated to PhysioDirect and 743 to usual care. Patients allocated to PhysioDirect had a shorter wait for treatment than those allocated to usual care [median 7 days vs 34 days; arm-time ratio 0.32, 95% confidence interval (CI) 0.29 to 0.35] and had fewer non-attended face-to-face appointments [incidence rate ratio 0.55 (95% CI 0.41 to 0.73)]. The primary outcome at 6 months' follow-up was equivalent between PhysioDirect and usual care [mean PCS 43.50 vs 44.18, adjusted difference in means -0.01 (95% CI -0.80 to 0.79)]. The secondary measures of health outcome all demonstrated equivalence at 6 months, with slightly greater improvement in the PhysioDirect arm at 6 weeks' follow-up. Patients were equally satisfied with access to care but slightly less satisfied overall with PhysioDirect compared with usual care. NHS costs (physiotherapy plus other relevant NHS costs) per patient were similar in the two arms [PhysioDirect £ 198.98 vs usual care £ 179.68, difference in means £ 19.30 (95% CI -£ 37.60 to £ 76.19)], while QALYs gained were also similar [difference in means 0.007 (95% CI -0.003 to 0.016)]. Incremental cost per QALY gained was £ 2889. The probability that PhysioDirect was cost-effective at a £ 20,000 willingness-to-pay threshold was 88%. These conclusions about cost-effectiveness were robust to sensitivity analyses. There was no evidence of difference between trial arms in cost to patients or value of lost production. No adverse events were detected. Providing physiotherapy via PhysioDirect is equally clinically effective compared with usual waiting list-based care, provides faster access to treatment, appears to be safe, and is broadly acceptable to patients. PhysioDirect is probably cost-effective compared with usual care.