[Planning of psychotropic drug trials. Control of interfering factors]

There are many non-specific factors affecting drug therapy outcome: Rickels classifies them as coming from: "a) the patient b) the physician c) the treatment milieu d) the nontreatment milieu". They exert most influence in the drug treatment of neurotic patients, less, but still some influence in depressed patients, and almost none in schizophrenic patients. In double-blind studies, control of the non-specific factors is achieved when their influence is the same in both groups of treatment, that means that possible non-specific factors should be rated for every patient, and after the end of the study, we should test for equal distribution in both groups. Other possibilities of control are discussed by Mr. Ferner in his paper. In pilot studies, the question is more difficult--we should be able to distinguish the influence of the drug from the influence of the non-specific factors. Two possibilities are discussed: 1. to look at differences in rated non-specific factors for the improved and the non-improved group, 2. to use the time-course of the global improvement (with a daily rating): the influence of the drug should manifest itself as a steady trend, whereas influences of non-specific factors would manifest themselves either as a trend too early in treatment or as abrupt changes in the improvement rating, possibly in connection with some external events.