Send to

Choose Destination
Int J Drug Policy. 2014 Nov;25(6):1103-12. doi: 10.1016/j.drugpo.2014.07.002. Epub 2014 Jul 12.

Is there a cannabis epidemic model? Evidence from France, Germany and USA.

Author information

Institut national d'études démographiques (INED), Paris, France; INSERM, U669, Paris, France; University Paris-Sud and University Paris Descartes, UMR-S0669, Paris, France. Electronic address:
IFT Institut für Therapieforschung, Munich, Germany.
University of Colorado Boulder, United States.
Institut national d'études démographiques (INED), Paris, France.
IFT Institut für Therapieforschung, Munich, Germany; Centre for Social Research on Alcohol and Drugs, SoRAD, Stockholm University, Stockholm, Sweden.



Cannabis is the most popular illicit drug in the world, but the process of its diffusion through the population has rarely been studied. The unfolding of the tobacco epidemic was accompanied by a shift in the educational gradient of users across generations. As a consequence, cannabis may show the same pattern of widening social inequalities. We test the diffusion hypotheses that a positive value in older cohorts - the more educated experimenting more - shifts to a negative one in younger cohorts - the more educated experimenting less, first for males and then females.


Three nationwide subsamples (18-64 years old) of representative surveys conducted in France (n=21,818), Germany (n=7887) and USA (n=37,115) in 2009-2010 recorded age at cannabis experimentation (i.e., first use), educational level, gender, and age. Cumulative prevalence of experimentation was plotted for three retrospective cohorts (50-64, 35-49, 18-34 years old at data collection) and multivariate time-discrete logistic regression was computed by gender and generation to model age at experimentation adjusted on age at data collection and educational level. This latter was measured according to four categories derived from the International Standard Classification of Education (ISCED) and a relative (rather than absolute) index of education.


The findings demonstrate a consistent pattern of evolution of the prevalence, gender ratio and educational gradient across generations and countries that support the hypothesis of an "epidemic" of cannabis experimentation that mimics the epidemic of tobacco.


We provide evidence for a cannabis epidemic model similar to the tobacco epidemic model. In the absence of clues regarding the future of cannabis use, our findings demonstrate that the gender gap is decreasing and, based on the epidemic model, suggest that we may expect widening social inequalities in cannabis experimentation if cannabis use decreases in the future.


Cannabis epidemic; Cannabis experimentation; Cross-national comparison; Educational inequalities; Gender inequalities

[Indexed for MEDLINE]
Free full text

Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center