Although concern has been voiced regarding the medical profession's relationship to nonmedicinal drug use, few data are available. In a random sample of more than 1,000 New Haven area high school students, nonmedicinal use was 1.9 to 11 times greater among those students with a history of three or more recommendations of a psychoactive drug by a physician as compared to those students who had never received such a recommendation. The relationship between medicinal psychoactive drugs and nonmedicinal drug use persisted while potentially confounding variables were controlled and was statistically significant for cigarettes, marijuana, amphetamines, barbiturates, and heroin use with odds ratios of 2.0, 1.6, 3.1, 4.0 and 6.2, respectively. The relationships of cigarette, marijuana, amphetamine, and heroin use to physician recommendation of a psychoactive drug appeared to be as strong as those of previously identified correlates of nonmedicinal drug use. The results were not attributable to artifacts of reporting. Although causal mechanisms cannot be inferred from these cross-sectional data, the results may have implications for preventive programs aimed at decreasing nonmedicinal drug use.
