Choosing Sample Sizes to Maximize Expected Health Benefits Subject to a Constraint on Total Trial Costs

Abstract

The authors present a method for choosing sample sizes for randomized controlled trials that maximizes expected health benefits (measured in expected discounted life years gained) subject to the decision maker's budget constraint. In comparison with similar approaches, the method introduces richer and more realistic models for the following quantities: costs and benefits during and after the trial, rates of adopting interventions after a positive rec ommendation, the distribution of data collected in the trial, and the decision to make a positive recommendation based on the results of the trial. Although the methodology is applicable to any type of trial, the emphasis in the paper is on prevention trials. Calculations involve Monte Carlo methods. An example is provided.