Whilst significant advances have been made in persuading clinical researchers of the value of conducting economic evaluation alongside clinical trials, a number of problems remain. The most fundamental is the fact that economic principles are almost entirely ignored in the traditional approach to trial design. For example, in the selection of an optimal sample size no consideration is given to the marginal costs or benefits of sample information. In the traditional approach this can lead to either unbounded or arbitrary sample sizes. This paper presents a decision-analytic approach to trial design which takes explicit account of the costs of sampling, the benefits of sample information and the decision rules of cost-effectiveness analysis. It also provides a consistent framework for setting priorities in research funding and establishes a set of screens (or hurdles) to evaluate the potential cost-effectiveness of research proposals. The framework permits research priority setting based explicitly on the budget constraint faced by clinical practitioners and on the information available prior to prospective research. It demonstrates the link between the value of clinical research and the budgetary restrictions on service provision, and it provides practical tools to establish the optimal allocation of resources between areas of clinical research or between service provision and research.