Continuity of cannabis use and violent offending over the life course

Although the association between cannabis use and violence has been reported in the literature, the precise nature of this relationship, especially whether it is causal in nature is unclear. Furthermore, previous literature has linked cannabis use with self-reports of violence, which is susceptible to recall bias and under-reporting. Hence, in the present study, we investigate the effect of exposure to cannabis at different stages of life on conviction records and self-reports of violent behaviour using data from a prospectively recruited population-based cohort of all school-aged male children from a defined geographical area in South London. Method Young males from the Cambridge Study of Delinquent Development (CSDD) (N=411) were followed up from age 8-56 to investigate the effect of cannabis use on (1) cumulative number of subsequent violent outcomes and (2) risk of violence. A multi-wave (eight assessments, T1-T8) follow-up design was employed that allowed temporal sequencing of the variables of interest and the analysis of violent outcome measures obtained from two sources, criminal records (violent conviction, VC) and self-reports (SR-V). Multivariate analyses were used to test whether exposure to cannabis, especially continued exposure, was an independent risk factor for subsequent violence, after controlling for potential confounders. Findings Univariate logistic regression revealed an increased risk for subsequent violent conviction (OR= 2.7) and self-reported violence (OR= 2.3) in those who used cannabis at least once in their life, with a stronger effect if use was reported at all three follow-up time-points (age 18, 32 and 48) as opposed to only one or two follow-up time-points, suggesting a dose-response relationship. Multivariable logistic regression revealed that compared to never users, continued exposure to cannabis as endorsed at multiple assessments at age 18, 32 and 48 was associated with a significantly higher risk of subsequent violent behaviour that continued until participants reached middle-age, as indexed by convictions (OR = 8.1, p<0.001) or self-reports (OR = 10.3, p<0.001). This effect was independent of and persisted after controlling for other putative risk factors for violence, including antisocial personality, family history, social class, ethnicity and alcohol or other drug use.