BMJ 1998; 317 doi: https://doi.org/10.1136/bmj.317.7167.1248a (Published 31 October 1998) Cite this as: BMJ 1998;317:1248

(Reprinted from BMJ 1948:ii:791-2)

The clinical trial of a remedy is as old as medicine itself. The idea of controlled investigation and statistical analysis is recent enough to call new. With the older hit-or-miss method the patient was treated, and subsequent observation showed with what result. If the patient-or patients-died, or at least failed to get well, the results that flowed from the remedy were unequivocal. It is when the patients, or anyway some of them, recover that we are faced with the post hoc propter hoc dilemma. Is the result due to the remedy or to the vis medicatrix naturae? Until quite recent years that difficulty was principally met-when it was met at all-by comparing patients submitted to the new treatment with the patients the clinician had observed in past years. The problems of such a comparison were many. It was often not certain that the patients selected for the trial were comparable with those previously seen; those earlier patients were often not equally well documented, so that records were lacking and the comparisons necessarily crude; the numbers involved were frequently far too few to merit the confidence placed in them; and mortality from the same disease at different periods of time is affected by factors such as breeding out of highly susceptible stock, economic and social conditions, and variations in the infecting organism. With a new discovery that produces dramatic results there would rarely be any …

View Full Text

Sign in

Log in through your institution

Free trial

Register for a free trial to thebmj.com to receive unlimited access to all content on thebmj.com for 14 days.
Sign up for a free trial