The cost effectiveness of orlistat in a 1-year weight-management programme for treating overweight and obese patients in Sweden : a treatment responder approach.

OBJECTIVE: To calculate the cost effectiveness (from the Swedish healthcare perspective) of orlistat plus diet for an obese and overweight population in a 1-year weight-management responder programme versus a 1-year weight-management programme based on diet only. As a reference, orlistat plus diet and diet only were also compared with a no-diet alternative.
METHOD: Costs and effectiveness were calculated in a decision-tree model by means of Monte Carlo simulation. Efficacy was derived from a pooled analysis of the orlistat clinical trial programme. Acquisition costs for orlistat (euro, 2003 prices), healthcare costs for visits to doctors and dieticians related to weight management, and costs related to the difference in diabetes mellitus incidence between treatment arms were included in the analysis. The health benefit of temporary weight loss was measured in the number of quality-adjusted life-years (QALYs) gained.
RESULTS: The number of responding (those with >5% weight loss) patients at month 3 was almost twice as high with orlistat compared with diet only: 48.9% versus 26.3%. Responding orlistat patients had a weight loss of 15.5% at month 12 compared with 7.9% for all patients on diet only. The incremental cost-effectiveness ratio (ICER) per QALY gained versus diet only was estimated to be 13,125 euro for the average patient starting on orlistat. When orlistat was compared with no diet, the cost effectiveness was improved. However, comparing diet only with no diet gave a slightly higher ICER, indicating that orlistat had an extended dominance over the diet-only alternative.
CONCLUSION: Our estimates indicated that orlistat in a 12-month dietary responder programme increased the number of QALYs and reduced the cumulative incidence of diabetes compared with diet only. Patients starting on orlistat in addition to a dietary programme achieved an ICER that was similar to many other well accepted healthcare treatment programmes. In order to improve the precision of our calculations, we need to confirm the key assumptions regarding temporary weight loss and utility gains, and the relationship between temporary weight loss and diabetes, as well as other co-morbidities, and to have better knowledge of the long-term impact of weight-management programmes in clinical practice, such as changes in weight-controlling behaviours and sustainability of weight loss.