Examining the link between collision involvement and cocaine use (2012)

Background: Cocaine is one of the more commonly found illicit drugs in injured drivers. In this work, we examine the association between self-reported past year cocaine use and past year collision involvement in a large representative sample of adult drivers in Ontario.
Methods: Data are based on the CAMH Monitor, an ongoing cross-sectional telephone survey of Ontario adults aged 18 and older. Five years of data (2002, 2003, 2004, 2006, 2008) were merged for this study (N = 8107) due to survey item availability. Logistic regression analysis was performed to identify the risk of self-reported collision involvement within the past 12 months associated with past year use of cocaine, while controlling for sociodemographic, driving exposure and drinking-driving (as a function of drinking status) factors. Due to listwise deletion, the logistic regression model was based on a reduced sample (N = 7284).
Results: The prevalence of self-reported collision involvement within the past year was 18.9% among those who used cocaine in the past year compared to 7.4% of non-users. Logistic regression analysis, controlling for the potential confounding effects of age, gender, income, driving exposure and drinking-driving measures, found the odds of collision involvement in the preceding year among cocaine users was over twice that of non-users (OR = 2.11, 95% CI = 1.06-4.18).
Conclusions: This study suggests that cocaine users are significantly more likely to report collision involvement in the past year. Additional work to confirm these observations, and to assess possible causal pathways, is needed.