Increased collision risk among drivers who report driving after using alcohol and after using cannabis.

Alcohol and cannabis are the two most commonly used psychoactive drugs in Canada: about 80% of adults consume alcohol, while nearly half report trying cannabis in their lifetime.1 It has long been known that driving under the influence of alcohol (DUIA) increases collision risk,2,3 and more recent evidence suggests that driving under the influence of cannabis (DUIC) does as well (e.g., refs. 4,5). Little research exists on the prevalence of people who report both DUIA and DUIC, or on collisions experienced by this group. Described here is the self-reported collision risk among drivers who report DUIA and DUIC in the Ontario adult population. Though these data cannot distinguish whether these behaviours occur concurrently, population data such as these can be useful in assessing the health risks facing individuals who report these behaviours. Data were based on the CAMH Monitor (CM), an annual repeated cross-sectional survey of Ontario adults. Since 2002, the CM includes an item on collision involvement in the previous 12 months; thus for the current study, data from 2002 to 2010 were merged (N=16,224). Each annual cycle of the CM is based on telephone interviews with a randomly selected sample ranging between 2,005 and 3,039 adults aged 18 years and older. The weighted sample is representative of the Ontario general adult population (response rates: 51% to 58%). Further information can be found in Ialomiteanu et al. (2012).6 Past-year self-reported collision involvement was examined in three groups (no DUIA or DUIC, either DUIA or DUIC, both DUIA and DUIC). All analyses are based on the weighted sample size, using STATA software. The association between collision involvement and independent variables was first examined using Chi-square analyses. Logistic regression analysis estimates the correlation between reporting neither DUIA nor DUIC, and reporting both DUIA and DUIC, on collision risk while controlling for demographic characteristics that significantly affected collision risk. Significant differences in the prevalence of self-reported collision involvement by driving after substance use were found (Figure 1). Drivers reporting neither DUIA nor DUIC (91.3% of the sample) had the lowest prevalence of collisions (6.7%). Those reporting either DUIA or DUIC (7.7% of the sample) reported a significantly higher prevalence of collision involvement of 9.6% (p<0.01). The highest likelihood of collision involvement (30.5%, p<0.01) was found among drivers reporting both behaviours (0.9% of the sample). Logistic regression analysis revealed that, after controlling for demographic factors, drivers reporting neither DUIA nor DUIC were significantly less likely to experience a collision than those who reported one of these behaviours (OR = 0.74, 95% CI 0.56-0.98, p<0.05) (see Table 1). People who reported both DUIA and DUIC had the highest collision risk, with over three times greater odds of collision involvement than those who reported only DUIA or DUIC (OR = 3.65, 95% CI 2.12-6.28, p<0.001). Figure 1. Percentage of respondents involved in collisions by substance(s) used while driving (CAMH Monitor, 2002-2010)