Longitudinal patterns of problematic cannabis use in adolescence and their association with childhood psychological risk factors

Cannabis continues to be the most widely used regulated drug worldwide. Use is particularly prevalent in younger populations, who also appear to be more vulnerable to developing problems related to their cannabis use than older users. Identifying patterns of cannabis problems across adolescence and which modifiable risk factors precede them could help to develop preventative measures to reduce the incidence of cannabis use disorder.

Here we present data from the Avon Longitudinal Study of Parents and Children (ALSPAC), a prospective, birth-cohort study from the UK. Data on problematic cannabis use were assessed using the Cannabis Abuse Screening Test at ages 15, 16, 18, 20, 22 and 24. A 3-level ('no', 'low' and 'high') variable of cannabis problems was created at each time point. Longitudinal latent class analysis was conducted to determine whether different patterns of cannabis problems emerged across the 6 time points. Our analysis plan and hypothesis were preregistered on the Open Science Framework prior to analysis.

We determined a 4-class solution to be the best fit for the data, based on a consideration of model fit statistics, LMR p-value, entropy and interpretability of the classes. The classes were defined as a ‘stable-no problematic use’ class (n=5180, 86%), a ‘late-onset problematic use’ class (n=361, 6%), an ‘early-onset followed by remission’ class (n=325, 5%) and a ‘stable-high problematic use’ class (n= 183, 3%). The association between mental health and cognition assessed at ages 8-10 and these later patterns of cannabis problems will be discussed.

The results indicate distinct patterns of cannabis problems experienced by adolescent cannabis users from a birth cohort study in the UK. Identifying modifiable risk factors for severity and increased duration of cannabis problems may provide insight into the development of preventative measures.