The burden of ankylosing spondylitis and the cost-effectiveness of treatment with infliximab (Remicade).

OBJECTIVES: In the past, treatment options for ankylosing spondylitis (AS) have been limited, and the introduction of new treatments such as infliximab will have a noticeable impact on health-care budgets. The objective of this study was therefore to assess the current burden of the disease and estimate the cost-effectiveness of infliximab treatments.

METHODS: A cross-sectional retrospective observational study of resource consumption and utility related to disease severity was performed in patients who had participated in a population survey between 1992 and 1994 at the University of Bath and patients regularly followed at the Royal National Hospital for Rheumatic Diseases in Bath for up to 9 years. Mean costs and utility were estimated using a regression model including age, gender, disease duration, disease activity and functional status, and disease development was expressed as annual progression of functional disability. Cost-effectiveness of infliximab was modelled using a 3-month placebo-controlled clinical trial with open 1-yr extension in 70 patients, over a total time frame of 2 yr. In the model, costs and utility controlled for disease severity and age from the observational study were assigned to individual patients. The effect of long-term treatment was evaluated in a hypothetical model over 30 yr.

RESULTS: Fifty-seven per cent of patients answered the questionnaires. The mean age was 57 (s.d. 11.2) yr, 74% were male and mean disease duration was 30.2 (11.7) yr. Mean total costs were estimated at pound 6765 (s.d. pound 166). Indirect costs represented 57.9% and non-medical costs such as investments and informal care accounted for 16.5% of total costs. Mean utility was 0.67 (0.21). In the main model, mean costs for untreated patients are estimated at pound 25,128. For the infliximab group, mean costs (excluding treatment) are estimated at pound 17,240, a reduction of 31%. Thus, part of the treatment cost was offset by savings in other resources ( pound 7888), leaving an incremental cost of pound 6214. Treatment increased the number of quality-adjusted live years (QALYs) by 0.175 QALYs, leading to a cost per QALY gained of pound 35,400 for the first year of treatment. When treatment is assumed to continue for the full 2 yr, the cost per QALY is pound 32,800. When infliximab infusions are given every 8 weeks instead of every 6 weeks, the cost per QALY is reduced to pound 17,300. In the long-term model, the cost per QALY is estimated at pound 9600.

CONCLUSIONS: Non-medical costs and production losses dominate costs in AS, and economic evaluation must therefore adopt a societal perspective. The cost of treatment with infliximab is partly offset by reductions in the cost of the disease and patients' quality of life is increased, leading to a cost per QALY gained in the vicinity of pound 30,000 to pound 40,000 in the short term, but potentially below pound 10,000 in the long term.