Objective: To estimate the 3-year incremental cost effectiveness and cost utility of introducing leflunomide into sequential therapy, consisting of the most frequently used disease-modifying antirheumatic drugs (DMARDs), for patients with rheumatoid arthritis in specialised, i.e. rheumatological, care in Germany.
Design and setting: The analysis was conducted from the societal perspective in Germany using an existing 3-year simulation model, which was adapted to the German healthcare system after secondary analysis of relevant publications and data. DMARD sequences including leflunomide were compared with those excluding leflunomide. Costs comprised direct costs incurred by treatment and indirect costs incurred by loss of productivity (sick leave and premature retirement) of rheumatoid arthritis patients. Effectiveness parameters were given by response years gained (RYGs) according to the American College of Rheumatology (ACR) criteria for 20%, 50% and 70% improvement (ACR20/50/70RYGs) and by QALYs gained (QALYGs). Costs, effects and QALYs were discounted by 5% per annum. In the base-case analysis, average values of costs, response years and QALYs were applied. Costs were in 1998-2001 values (euro 1 approximately equal to $US 0.91, average of the period from the year 2000 through 2001).
Main outcome measures and results: After 3 years, adding leflunomide was less costly and more effective than the strategy excluding leflunomide when total (direct and indirect) costs were considered. There were savings of euro 271,777 and 8.1, 4.3, 5.1 and 4.9 ACR20RYGs, ACR50RYGs, ACR70RYGs and QALYGs per 100 patients, respectively, obtained through adding leflunomide. Focusing on direct costs, adding leflunomide was more costly and more effective compared with excluding leflunomide, with an incremental cost effectiveness of euro 5004 per ACR20RYG, euro 9535 per ACR50RYG, euro 7996 per ACR70RYG, and an incremental cost utility of euro8301 per QALYG, after 3 years. The robustness of the results was shown in comprehensive sensitivity analyses. In the analysis of extremes, different combinations of the limits of cost, effectiveness and utility parameters were investigated. Adding leflunomide to sequential DMARD therapy remained dominant in 79% of the possible cases, i.e. was less costly and more effective than the strategy excluding leflunomide. Focusing on direct costs, adding leflunomide became dominant in 29% and remained more costly and more effective in 50% of possible cases.
Conclusions: Our analysis suggests, with its underlying data and assumptions, that having leflunomide as an additional option in a DMARD treatment sequence extends the time patients benefit from DMARD therapy at reasonable additional direct costs. Adding leflunomide may even be cost saving when total (direct and indirect) costs are considered. As data on DMARD effectiveness were extracted from the results of clinical trials, real-world data from observational studies would be needed to corroborate the findings of the present analysis.