Use of randomized controlled trials in organizational decision making: A cost-minimization approach

Objective: To determine whether conducting a randomized controlled trial (R CT) of an optimal size could result in a net reduction in regret costs to a n organization in a policy decision regarding drug coverage. Study Design: Probabilistic decision model and cost-minimization analysis. Patients and Methods: We used a probabilistic decision model and cost-minim ization analysis to estimate overall 5-year regret costs (ie, the financial loss associated with making an incorrect decision) to a health plan of per forming an RCT for a new asthma medication and making the coverage decision using the results from the RCT. Our analyses were based on a hypothetical cohort of 10,000 patients with moderate to severe asthma enrolled in the he alth plan. Results: An RCT with 150 patients per group would decrease the expected 5-y ear regret costs by $568,000. Under the baseline assumptions, the sample si ze resulting in the greatest reduction in losses was about 400 per group, w ith a loss reduction of $697,000. The cost-minimizing sample size and amoun t of loss reductions varied substantially with assumptions regarding baseli ne rates of hospitalization for asthma, cost of hospitalization for asthma, and drug utilization rates and costs. Moderate-sized RCTs, however, nearly always resulted in loss reductions. Conclusions: Randomized controlled trials can be a useful organizational de cision-making tool to reduce financial regret.