Background: In order to better understand trends in cannabis use, this study estimates independent and non-confounded age, period and cohort effects on 12-month cannabis use prevalence and frequency.
Methods: Data from seven waves of the German Epidemiological Survey of Substance Abuse (ESA) conducted between 1990 and 2009 were used. The total sample included 29,836 men and 34,877 women aged 18-65 years. Cross-classified random effects models were conducted to estimate fixed effects of age and random effects of periods and birth cohorts. Analyses were stratified by gender.
Results: With regard to 12-month prevalence, substantial age variations were identified with a peak of cannabis use in young adulthood. The overall effects of periods and birth cohorts were not significant. With regard to frequency of use, there were no significant influences of either variable. The results were largely identical for men and women.
Conclusions: This study identified little variation in trend data of current cannabis use. When confounding influences of period and ageing are controlled, the widely reported increases in cannabis consumption in recent birth cohorts cannot be demonstrated.