One of the more frequently voiced empirical generalizations to come out of psychopharmaeological research states the existence of an inverse relationship between the therapeutic effectiveness of a given drug and the length of time that has elapsed since its introduction. This generalization epitomizes in its very cynicism the major difficulty which confronts the investigator concerned with evaluating drug action, viz., that the effects which arise solely from the action of the particular drug under investigation are ubiquitously confounded with various non-drug effects.
Indications of the heterogeneity of these potentially operative nondrug factors have accumulated with each passing year. There now exists substantial evidence, for example, that the reported occurrence of beneficial change subsequent to the ingestion of a given psychopharmacological agent can be a function not simply of the drug, but also to a greater or lesser extent of the size and composition of the patient samples 1, of patients' attitudes and expectations 2, of various components of what has come to be called the "treatment milieu "a , of the measurement process per se --including such things as the selection of i We include under this heading the necessity that the patient cohort be homogeneous in regards the "target" symptom or symptoms to which the medication is aimed, that patients be randomly allocated to treatment and control groups, and that they be of sufficient number to insure that random allocation will accomplish its task.
2 Under patient's attitudes and expectations we refer substantively to whatsoever goes to make up what WOLF (1950, p. 106), in explaining the effects of placebos on his patients, referred to as "the conviction of the patient that this or that would result." Several excellent summaries of the placebo literature have recently appeared (BARB]S~ 1959; FRANK 1961; KURLA~D ]957; I~OSENTttAL and FRANK 1956; T~OUTO~ T 1957).
3 We here refer to such diverse matters as inter-patient effects (CLA~K 1956;NOWLIS 1958), the usually unforeseen and thus usually uncontrolled effect on patients of increased staff attention to them which typically accompanies inpatient drug studies (RAStIKIS and SMA~R ]957), and the general evaluative orientation of the non-medical staff of the treatment facility to psychopharmacological treatment. l~sychopharmacologia, Bd. 5 24