Designing naturalistic prospective studies of economic and effectiveness outcomes associated with novel antipsychotic therapies.

The cornerstone of recent pharmacoeconomic work in schizophrenia is the hypothesis that the improved efficacy of novel antipsychotic medications will lead to a reduction in medical services utilization, thereby reducing direct medical costs associated with treatment. Creating the most valid design to prospectively examine the effectiveness and costs of competing pharmacotherapies requires a dialectic of opposing research paradigms. The final protocol must represent a series of decisions that strike a careful balance between being scientifically sound (internal validity) and generalizable to the real world of clinical treatment (external validity). The results must be useful to decision-makers in determining to what extent reductions in healthcare expenditures can offset higher drug acquisition costs within their type of treatment environment. This article is a review of several methodological challenges in the design of medical effectiveness trials, including whether to blind the study, definition of the patient population, degree of physician discretion in treatment, and how to collect and analyze data for patients who discontinue their originally assigned medication. The article also provides a discussion of how clinical practices can inform decisions made to meet these challenges. The issues are illustrated through a prospective study designed to evaluate the cost-effectiveness of the newer antipsychotics in general and olanzapine in particular. Cost-effectiveness studies of novel antipsychotic medications, particularly those with naturalistic designs, will increase in importance as the use of these second-generation agents continues to expand.