Abstract
In spite of having been formulated nearly two decades back, there is as yet no consensus on the validity of the clinically popular self-medication hypothesis (SMH) of substance use disorders in patients with dual diagnosis. SMH broadly proposes that patients use substances in a non-random fashion so that the psychopharmacologic characteristics of particular substances are used to alleviate a variety of psychiatric symptoms and emotional distress. In order to test the SMH empirically, it was broken down to five sub-hypotheses, which were tested in a group of dual-diagnosis schizophrenia (DDS) patients vis-à-vis a group of only-schizophrenia (S) patients (n = 22 each). The DDS group scored lower than the S group regarding general and some specific psychopathology. The DDS patients ascribed reasons for substance use more often for hedonistic pursuit but also for reduction in symptoms and distress. There was a trend for alcohol to be used more for self-medication purposes compared to opioids and cannabis. The perceived effects of these three substances were significantly different on several symptom/distress dimensions. Finally, there was some degree of “match” between symptom-oriented reasons for use of substances and the effect that was perceived. All of this evidence provides a consistent but modest support for the SMH for “some patients, some substances, and some symptoms.” The implications are discussed. (Am J Addict 2004;13:139–150)