Free-riding and the prisoner's dilemma: problems in funding economic analyses of phase III cancer clinical trials.

PURPOSE: Both economic and clinical data on new agents are important to policy-makers who approve pharmaceuticals for widespread use. Randomized clinical trials have been used to evaluate both clinical results and total medical costs associated with new agents. With new expensive pharmaceutical agents, early assessments of economic benefit have taken on greater importance to physicians and patients. Who should provide financial support to these integrated economic and clinical analyses in clinical trials? Here we describe issues that hinder funding of economic analyses and propose potential support mechanisms. RESULTS: The Cancer and Leukemia Group B (CALGB), a large, national cooperative group of academic and community hospitals in the United States, designed a non-small-cell lung cancer (NSCLC) treatment trial to compare two widely used supportive care regimens that varied 20-fold in cost. One important objective of this trial was to compare the cost-effectiveness of the two regimens. While funding for the clinical trial was supported by grants from the National Cancer Institute and the pharmaceutical companies involved in the trial, no specific funding agency was willing and/or able to provide financial support for the economic analyses. After 2 years of planning, the clinical trial was retracted when the funding for the economic analyses could not be secured. The prisoner's dilemma, individual reluctance to support a common social good, explains the lack of funding. CONCLUSION: Economic theory predicts difficulties in evaluating cost-effectiveness of new pharmaceuticals and reluctance to support economic analyses of clinical trials. Economic analyses will require new sources of funds that will not take scarce resources from clinical trials groups. Options for funding include a new federal agency, coordinated work by existing agencies, or academic centers for economic analysis.

Duke Scholars

Author Stephen L. George Biostatistics & Bioinformatics, Division of Biostatistics