Analytic choices in economic models of treatments for rheumatoid arthritis: What makes a difference?

Objectives. To compare the analytic judgments, data, and assumptions of different models used in the economic evaluation of infliximab, one of a new class of drugs for rheumatoid arthritis (RA). Methods. A detailed assessment was made of 4 models, 1 submitted (in a reimbursement dossier) by the manufacturer, 1 produced by an independent academic group, and 2 recently published in the literature. Factors considered included the key data inputs, assumptions about the sequencing of treatments for RA, the methods used to calculate health utilities, and the estimation of cost offsets. Results. Two of the 4 models, although embodying different methodological approaches, gave fairly similar results (approximately 25,000- pound 35,000 pound cost per additional quality-adjusted life year [QALY] gained). The other 2 models, both by an independent academic group, gave much higher estimates, ranging from 50,000 pound to 60,000 pound to more than 100,000 pound per additional QALY. The differences appeared to depend mainly on differences in model structure, the assumptions about the positioning of infliximab in the treatment sequence, and the relationship between Health Assessment Questionnaire (HAQ) states and QALYs. Conclusions. Economic models of treatments for RA incorporate different key data inputs and analytic judgments. However, convergence was observed in some of the estimates produced by the models, particularly when adjustments were made for some of the differences in input parameters. Nevertheless, differences in the choice of model structure and in key assumptions also had a major impact on results. Therefore, more discussion is needed to reach a consensus on some of these methodological issues.