ETHICS OF CLINICAL-TRIALS - DO THEY HELP THE PATIENT

Ethical justification of a randomized clinical trial requires at the o utset that the investigators be able to state an honest null hypothesi s regarding the therapies to be compared. For example, in a randomized clinical trial with only two arms (A and B), it is necessary to state that there is no scientifically validated reason to predict that ther apy A will prove to be either superior or inferior to therapy B. Moreo ver, there must be no therapy C known to be superior to A and B unless there is a cogent reason to reject therapy C; the population of resea rch subjects will consist either of those in whom therapy C has been t ried without success or individuals who are aware of therapy C and hav e rejected it for various reasons. Thus, theoretically, the patients e nrolled in a randomized clinical trial are assured that they will rece ive one of the two best known therapies for their condition. At first glance, one might conclude that randomized clinical trials do, in gene ral, help the patient. There are some features of at least some clinic al trials, however, that may be detrimental to the interests of some p atients. It commonly is said, for example, that patients in randomized clinical trials are deprived of the ''good of personal care.''1 The p hysician-investigator may be constrained by rigid protocol requirement s from being responsive to the particular needs and desires of the pat ient. Additionally, the process of obtaining informed consent to rando mized clinical trials may be preemptive: patients who might have chose n palliative care rather than chemotherapy may not be afforded adequat e opportunity to make such a choice. Any randomized clinical trial sho uld maximize the likelihood of helping the patient.