To the Editor.—
In the useful article on the cost-effectiveness of clinical trials,1 the reader is left with the impression that clinical trials are a good investment, even though only a few, selected trials were studied. The author is careful to point out that this was not a random selection of trials, but that's the problem, it should have been.Presumably, there are clinical trials that do not result in the development of a useful health care intervention (and also do not prevent the use of ineffective medical care activities). These studies also have costs, possibly quite large. How are these costs to be handled? Should they be considered indirect costs of the trials that produce useful results? A cost-effectiveness study of clinical trials would be more convincing if it were done on a randomly selected group of trials?