Cannabis use in adolescence : studies of trends, correlates and adverse mental health consequences

Background: Cannabis is currently the most commonly used illicit psychoactive substance. Cannabis use is typically initiated during adolescence and is more common among young people than among adults. Cannabis use is associated with negative health and social consequences and early initiation increases the risk of negative outcomes. Cannabis may be particularly harmful to individuals whose brain is not yet fully developed. Overall aims: The overall aim was to improve our understanding of various aspects of the epidemiology of cannabis use among adolescents. The four studies within the thesis highlight the following main research questions regarding adolescents: (I) Are trends in cannabis use related to trends in drinking? (II) Is socioeconomic status (SES) related to cannabis use? (III) Do pupils in introductory programs (IPs) have an increased risk of cannabis use compared with pupils in national programs? (IV) Is there an association between cannabis use and mental distress? Data and methods: Data for three of the four studies came from a nationally representative, self-reported, and anonymous school survey conducted annually by the Swedish Council for Information on Alcohol and Other Drugs. The survey is cross-sectional and the sample consists of Swedish 9th grade pupils (15–16 years of age) and pupils in year 2 in upper secondary school (17–18 years of age). To assess whether trends in cannabis use were related to trends in drinking, autoregressive integrated moving average (ARIMA) time-series analysis was used (Study I). To assess the association between self-reported cannabis use and socioeconomic status (SES) and academic orientation, respectively, multi-level mixed-effects Poisson regression with robust standard errors was used. Because the pupils were clustered in school classes, the assumption of independence between observations was not met. Hence, data were analyzed with a random intercept for the school/school class level (Studies II and III). In the fourth study, data stemmed from the Young in ...