Key points are not available for this paper at this time.
If you wanted to know whether a new unproved clinical test or operation or a new health program was any good, how would you proceed to find out? The best way would be to set up a controlled clinical trial, comparing the new with the old in everything from morbidity to dollar cost. Such essential trials, however, are always lengthy and very expensive. Each may focus on different aspects of the problem, and although the results of some may be available, others may still be brewing. In the meantime, as a practicing physician you want to advise your patients prudently . . .
Drummond Rennie (1981) studied this question.