How Large Should a Clinical Trial Be?
Journal of the Royal Statistical Society: Series D (The Statistician), 49 (2):
177-187 (July 2000)2979<m:linebreak></m:linebreak>Farmacoeconomia.DOI: 10.1111/1467-9884.00228
Abstract
The determination of the optimal sample size of a clinical trial is considered when the number of subsequent users of a new treatment is a function of both the statistical significance of the difference and of the magnitude of the apparent difference between the performance of the new treatment and that of the treatment in current use. An objective function is proposed consisting of the total benefit from the resulting change in the number of patients using the new treatment minus the cost of the trial. From this the optimal sample size may be calculated. The model has features which allow for the following contingencies: a cost difference between the two treatments; a pay-off function defined either from the public health or from a drug company standpoint; the performance of the existing treatment is either known or unknown; varying degrees of severity of the condition to be treated; a set-up cost for conducting the trial.