Cost-utility of exercise therapy added to general practitioners’ care versus general practitioners’ care alone in patients with hip osteoarthritis

Abstract

Purpose: To determine the cost-effectiveness over a period of 12 months of exercise therapy added to general practitioners’ care compared to general practitioners’ care alone in patients with a new episode of hip OA in general practice. Methods: A cost-utility study was performed in conjunction with a multi-center randomized controlled trial with a parallel group design. Patients participated if they were 45 years or older, comply with the clinical American College of Rheumatology criteria for hip OA, and visited their general practitioner (GP) for a new episode of complaints due to hip osteoarthritis. Patients were excluded if they: 1) were already treated with exercise therapy in the present episode of hip OA, 2) had a hip pain score of <2 on the 11-point numeric rating scale (0 to 10), 3) had a high level of physical function, a score of <2 on the walking ability and the physical function sections of the Algofunctional index, 4) had undergone hip surgery or on the waiting list, 5) had severe disabling co-morbidity and 6) had insufficient comprehension of the Dutch language and/or were mentally incapable of participation. The patients were allocated at random in two treatment groups: one group received exercise therapy supervised by a physiotherapist (up to 15 sessions in the first 3 months and 3 follow-up sessions in month 5, 7 and 9) added to GP care and the control group received GP care only. The cost-utility study was primarily conducted from a societal perspective, but the healthcare perspective was also applied. Data on direct medical costs, productivity costs and quality of life was collected at baseline and at 6, 13, 26, 39 and 52 weeks follow-up. Annual costs were determined by adding up the costs per period. The costs for the time between the measurement periods (week 6–7) were established through linear interpolation. All costs were based on Euro 2011 cost data. The quality of life score per patient during the 52 weeks follow up was estimated by combining the EQ-5D scores at all measurement moments. Differences between the intervention and control group were assessed by means of the independent sample T test (for variables showing a normal distribution), the Mann Whitney U test (for variables not normally distributed) or Pearson Chi-square test (for variable fractions). Using nonparametric bootstrapping (drawing 2,500 observations at random), the degree of uncertainty for costs and health effects and the cost-utility ratio was examined on the so-called CE-plane. In addition, an acceptability curve was generated to indicate the probability that the intervention has lower incremental costs per quality adjusted life year (QALY) gained than various thresholds for the maximum willingness to pay for an extra QALY. Results: The study took place in the period 2009–2012 and finally 203 patients were included. The annual direct medical costs per patient were significantly lower for the exercise group (€ 1,233) compared to the control group (€ 1,331) despite additional physiotherapy visits. The average annual societal costs per patient were lower in the exercise group (€ 2,634 versus € 3,241; P = 0,002). Productivity costs were higher than direct medical costs. Patients in the exercise group experienced a slightly, but not significantly, higher quality of life (0.776 versus 0.770). We found a societal average cost effectiveness (CE)-ratio of – € 107,505 per quality adjusted life year (exercise cost effective). When only direct medical costs were included, the average cost per quality adjusted life year amounted to € −17,441, but the uncertainty around both CE-ratios was substantial. Conclusions: Over a period of 52 weeks, with a CE-ratio of – € 107,505 per QALY from the societal perspective and a CE-ratio of € −17,441 per QALY from the healthcare perspective, our study revealed a considerable probability that exercise therapy added to GP care is cost saving or cost effective as compared to GP care alone.