Abstract
Legalization of nonmedical cannabis occurred in Canada in October 2018, necessitating a stronger understanding of potential consequences such as driving under the influence of cannabis (DUIC). Evidence associating DUIC with increased risk of motor vehicle collisions stems from natural experiments upon cannabis legalization, 1 simulation research, 2 and linkage studies to actual crash data. 3,4 To augment this knowledge base, relevant items have been included in recent population-based surveys in Canada.
We analyzed cross-sectional population-representative data collected by the Canadian Community Health Surveys (CCHS) 5 from 2015 to 2018 and the Canadian Tobacco, Alcohol and Drugs Survey (CTADS, 2017). Two populations examined were (1) individuals aged 15 and older and (2) individuals aged 15 and older who reported use of cannabis in the past 12 months. This age range was chosen to accommodate the minimum age of obtaining a learner’s license in Alberta (age 14) and Ontario (age 16). The relevant CCHS content was collected only in 2 provinces in the survey cycles. In 2015 and 2016, the CCHS driving and drug use modules were both administered in Alberta only, and in 2017 and 2018, these modules were administered in Ontario only. CTADS reported relevant data only in 2017 but covered the national population. The CCHS survey questionnaire item was “Have you ever tried marijuana or hashish?” followed by “Was this in the past 12 months?” DUIC was assessed by a single item: “In the past 12 months, have you driven a motor vehicle under the influence of an illicit substance such as marijuana or hashish?” The CTADS item was “In the past 12 months, have you driven a motor vehicle within 2 hours of using marijuana?” We examined prevalence in the overall population and with stratification for age and sex among those reporting use in the past 12 months. Appropriate weighting and bootstrap procedures were used to estimate 95% confidence intervals (CIs).
The demographics of the study population were consistent in 2015 to 2018; individuals aged 15 to 24 comprised 14% of the population in both Alberta and Ontario, and females comprised 49% of the population in Alberta and 51% of the population in Ontario. In Alberta, the proportion of individuals who reported cannabis use in the past 12 months was 25% in 2015 and 29% in 2016, while in Ontario, the proportion was 33% in 2017 and 39% in 2018.
The prevalence of DUIC for the provincial (CCHS) and national (CTADS) populations and age group–specific estimates for individuals who reported use of cannabis in the past 12 months are shown in Table 1. Higher prevalence of DUIC was found in Alberta in 2015 to 2016 compared to Ontario in the years 2017 to 2018. For all 4 years, the highest rate of DUIC was that for males aged 15 to 24, where the frequency ranged between 20% and 40% of cannabis users. The lowest frequency was in women aged 25+, where the prevalence ranged from 7% to 17% of those reporting cannabis use. In aggregate, these frequencies represent approximately 2% to 3% of the population in each year.
Estimated Overall and Age Group–specific Population Prevalence (%) and 95% CI for Driving Under the Influence of Cannabis, from the CCHSa and CTADSb Surveys.
a Canadian Community Health Survey: “In the past 12 months, have you driven a motor vehicle under the influence of an illicit substance such as marijuana or hashish?”
b Canadian Tobacco, Alcohol and Drugs Survey: “In the past 12 months, have you driven a motor vehicle within 2 hours of using marijuana?”
We assessed the impact of age category and sex on DUIC in the past year. Males were more likely to report DUIC compared to females when adjusted for age category in Alberta in 2015 (odds ratio [OR] 2.13; 95% CI, 1.12 to 4.06) and in Ontario in 2017 (OR 2.53; 95% CI, 1.61 to 4.00). In Ontario, individuals aged 15 to 24 were more likely to drive under the influence of cannabis than individuals aged 25 and older when adjusted for sex in 2017 (OR 1.64; 95% CI, 1.00 to 2.68) and 2018 (OR 1.91; 95% CI, 1.16 to 3.15). Despite the different populations and the differing wording of the DUIC items, the overall prevalence and the age- and sex-specific patterns were similar in CCHS and CTADS.
We characterized time trends of DUIC using Canadian population survey data, finding a consistent pattern over 2015 to 2018. The CCHS estimates augment those of CTADS, a random-digit dialing-based survey, and older data from the Canadian Alcohol and Drug Use Monitoring Survey (CADUMS). CADUMS reported DUIC prevalence of 2.5% in Canada and 2.7% in Ontario in 2012, 6 similar to the more recent estimates reported here. A national estimate of 1.8% arose from CTADS in 2017. In our analysis, males aged 15 to 24 were consistently found to have the highest prevalence of DUIC. As a group, young males are more likely to engage in risk behaviors including drug use compared to other groups. Higher rates of DUIC were noted in Alberta compared to Ontario although the estimates were insufficiently precise to fully confirm this. Future research should continue to examine possible provincial differences in DUIC. Future studies should continue to monitor DUIC in order to assess changes postlegalization.
Our work complements previous estimates of DUIC using robust population-representative Canadian data. Public policy actions toward prevention of DUIC appear necessary and will have the greatest impact if they are effective in the 15 to 24 age range and in males.
Footnotes
Authors’ Note
Data were accessed through the Statistics Canada Prairie Regional Research Data Centre. While the investigators cannot release the data, the data can be accessed through processes established by the Canadian Research Data Centre Network. The views expressed in this paper do not necessarily represent the CRDCN’s or that of its partners.
Acknowledgments
The authors would like to thank Dina Lavorato and Jeanne Williams for their assistance with the data analysis. The analysis was conducted at the Prairie Regional Data Centre, which is part of the Canadian Research Data Centre Network (CRDCN). The services and activities provided by the CRDCN are made possible by the financial or in-kind support of the SSHRC, CIHR, CFI, Statistics Canada, and participating universities whose support is gratefully acknowledged.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Cuthbertson and Fischer Chair in Pediatric Mental Health, held by Dr. Patten.
