Initiation, continuation or discontinuation of cannabis use in the general population.

A nationally representative sample (n = 1997) of young people (17-20 years of age) was examined twice with an interval of two years using postal questionnaires supplemented by telephone interviews of non-responders. Sixty-eight per cent of the original sample participated in both surveys. Factors found to increase the probability of having ever used cannabis were sex, place of residence, divorced parents, earlier problems with education and employment, and mental and somatic health problems. However, these factors could not predict the persistence/cessation of cannabis use which was significantly related to the establishment of an adult social role with a partner/spouse and/or having children, and negatively correlated to long-term unemployment. Among men, high aspirations as to future occupational activity predicted cessation of cannabis use. The hypothesis that alcohol consumption would be increased amongst those individuals who stopped using cannabis was not supported. The results replicate some of the main findings of Kandel and coworkers (1984, 1986, 1989) supporting the role incompatibility theory of Thornton (1975) as an explanation of the cessation of cannabis use.