Secondary Logo

Journal Logo


Impact of Cannabis Use on Road Traffic Collisions and Safety at Work

Systematic Review and Meta-analysis

Els, Charl MBChB, MMed Psych, FCPsych[SA], Dip. ABAM1; Jackson, Tanya D. PhD, RPsych2; Tsuyuki, Ross T. BSc(Pharm), PharmD, MSc, FCSHP, FACC, FCAHS3; Aidoo, Henry MBChB, MSc2; Wyatt, Graeme BA (Hons)2; Sowah, Daniel PhD2; Chao, Danny CCFP2; Hoffman, Harold MD, FRCPC2; Kunyk, Diane BScN, MN, PhD4; Milen, Mathew MSW, SAP2; Stewart-Patterson, Chris MD, CCBOM, FACOEM5; Dick, Bruce D. PhD, RPsych6; Farnan, Paul MB, BCh, FCFPC, Dip. ABAM7; Straube, Sebastian BM BCh, MA (Oxon), DPhil2

Author Information
Canadian Journal of Addiction: March 2019 - Volume 10 - Issue 1 - p 8-15
doi: 10.1097/CXA.0000000000000046
  • Free



With the legalization of cannabis for recreational purposes in Canada, potential challenges to workplaces, unions, employers, employees, and occupational physicians emerge regarding how to adequately mitigate the risks of impairment associated with cannabis consumption by workers employed in decision-critical1 or safety-sensitive settings. We do not aim to replace decision-making by trained practitioners with this review and meta-analysis, but rather hope that our work may serve as an aid in this task.

In 2000, Canada became the first country to offer legal access to cannabis for medical purposes.2 In 2015, the Government of Canada issued a commitment to legalizing cannabis for recreational purposes.3 Bill C-45, the Cannabis Act, was passed and became effective on October 17, 2018, for the purposes of legalizing, regulating, and restricting access to cannabis, and also to amend the Controlled Drugs and Substances Act to allow for the legalization of cannabis.4 Canada is now only the second country in the world, and the first of the G7 countries, to legalize cannabis for recreational use on a national scale.

Employers occasionally encounter workers in safety-sensitive settings with an authorization for cannabis for medical purposes, while others may not disclose this. For context, Health Canada projects the number of registered medical users to be 308,755 by 2024,5 up from 477 in 2002. Reasonable evidence currently supports therapeutic efficacy of cannabis for a limited number of indications, for example, spasticity (primarily in multiple sclerosis), chemotherapy-induced nausea and vomiting, and neuropathic pain, where benefit is likely small.6 In light of the common occurrence of adverse effects associated with the use of cannabis, benefits have to be considerable to justify its use.

Cannabis, which remains illicit in most countries, is the most commonly used such substance globally, with a 1-year prevalence rate of 3.8%.7 In 2015 the Canadian Tobacco, Alcohol and Drugs Survey8 reported a lifetime prevalence of cannabis use of 44.5% and a past-year prevalence of 12% (or approximately 3.6 million Canadians), making it the most commonly used illicit drug in Canada, before its legalization. The Canadian Cannabis Survey, however, suggested higher rates of consumption, with 18% of those aged 25 years and older reporting use in the past 12 months.9

That survey found that, for persons aged 25 or older who had used cannabis in the past 12 months, the frequency of use to achieve euphoria before or at work was as follows: 80.4% reported no such cannabis use, 7.0% reported daily or weekly use, 3.4% reported monthly use, and 9.2% reported use which occurred less than once a month.

The cannabis plant delivers numerous cannabinoids, with unique pharmacokinetics; the best studied of which are tetrahydrocannabinol (THC), the primary psychoactive compound and considered the most occupationally relevant cannabinoid, and cannabidiol (CBD). THC content can vary widely, as cannabis strains are cultivated specifically for potency. Recent methods, such as carbon dioxide supercritical fluid extraction and short path distillation, have produced THC concentrations of above 90%.10 CBD is also present in high concentrations in some cannabis strains, but has been traditionally assumed to be nonpsychoactive.

Cannabis use is associated with a range of adverse effects,11 including in the short term: impaired memory, impaired motor coordination, impaired driving, injury, altered judgment, increased risk of sexually transmitted infections, paranoia, and, in high doses, psychosis.12 In the long term, cannabis interferes with brain development, may result in poor educational outcomes, cognitive impairment, diminished life satisfaction and achievement, chronic bronchitis (if smoked), and increased risk for psychotic disorder.11

THC is implicated in achieving euphoria and is considered addictive, and aside from cannabis intoxication (which includes intoxication delirium), cannabis use is associated with the potential for development of cannabis use disorder (cannabis dependence or addiction), as well as cannabis withdrawal. The Diagnostic and Statistical Manual of Mental Disorders13 further recognizes cannabis-induced psychotic disorder, cannabis-induced anxiety disorder, and cannabis-induced sleep disorder.

Various tests have been proposed to detect impairment in safety-sensitive settings, but at the time of this publication, there is insufficient evidence in support of any testing other than blood levels as providing adequate sensitivity, specificity, or predictive value to allow for adoption as a screening method. It has been suggested that plasma levels of 5 ng/mL of THC (the parent compound), along with the presence of medical signs indicative of consumption, could be used as 1 indicator of acute impairment; plasma levels of THC above 5 to 10 ng/mL have been shown to be indicative of severe impairment.14 Reasonable suspicion, or probable cause testing, remains the default for detection of workplace impairment.

The duration of impairment following cannabis consumption is not well established. The College of Family Physicians of Canada suggests that patients consuming dried cannabis should be advised not to drive for at least 4 hours after inhalation, 6 hours after oral ingestion, and 8 hours after any use if the patient experiences euphoria.15 Health Canada, however, suggests that the ability to drive or perform activities requiring alertness may be impaired for up to 24 hours,5 well after other effects may have subsided.

Ramaekers et al16 caution that studies in the past have typically relied on low-potency cannabis (4% THC), and may be a conservative estimate of detrimental effects. Methodological limitations and small study size notwithstanding, an aircraft simulator study17 reported carry-over effects of THC lasting 24 hours. Although the majority of pilots in this study demonstrated some degree of impairment at 24 hours after consumption, only 1 subject reported awareness of the substance's effects.

The absence of a legislative standard to define cannabis-related impairment in the workplace, combined with the lack of a dependable, practical, and validated instrument to measure such impairment, raises questions about occupational risk and safety.

To address the question of safety, the risk from motor vehicle crash data was extrapolated to the impact of cannabis in safety-sensitive workplaces. Such extrapolation is commonly used in occupational medicine; for example, studies of driving collision risk from alcohol intoxication were used to inform alcohol impairment-related risk assessments in safety-sensitive settings.14 This is the same methodology now followed by both the Railway Association of Canada18 and the American College of Occupational and Environmental Medicine in their guideline on the use of opioids for workers in safety-sensitive settings.19 Generalizing the performance impairments confirmed in cannabis research—including impaired motor skills, concentration, and coordination on divided attention tasks—to safety-sensitive occupational settings in general is therefore also believed reasonable here.


We considered publications from peer-reviewed academic journals and grey literature sources (such as government and agency reports or working papers and conference proceedings, if sufficient detail was provided). Publications needed to report on participants who had cannabis use confirmed by testing of blood, saliva, or urine. The presence of cannabis needed to be reported exclusively (i.e., the number of participants who tested positive for cannabis and other substances could be distinguished from the number who tested positive for cannabis alone). One of two outcomes of interest, either road traffic collisions or a safety-related work outcome, needed to be presented. The presence of a control group who had not used cannabis and who were representative of the general population was also required. Both cohort and case–control research designs were acceptable. Reflecting the linguistic limitations of the research team, the included publications needed to be in English, French, or German.

A search strategy for MEDLINE (Table 1) was developed by a health sciences librarian. This search strategy utilizes the specialized Medical Subject Headings and combines groups of cannabis-related, work-related, and collision/accident-related terms. The date of the last search in MEDLINE was May 24, 2017. Subsequently, the database search was supplemented by screening the reference lists of the included studies and review articles in the field; authors also identified further potentially relevant papers known to them, published at any time. Study selection and data extraction were performed in duplicate (CE, MM, TJ, HA, DS, DC, BD, DK, PF, GW), with the senior author (SS) arbitrating differences.

Table 1
Table 1:
MEDLINE Search Strategy

The following data were extracted from the studies:

  1. citation details;
  2. when the study data collection took place;
  3. what country the study took place in;
  4. characteristics of the study participants;
  5. testing process and specimen types;
  6. numbers of participants who tested positive for cannabis;
  7. numbers of participants who had road traffic collisions or workplace safety incidents.

We planned to use fixed effect or random effects meta-analysis, depending on between-study heterogeneity (I-squared statistic). Odds ratios (OR) and 95% confidence intervals (CIs) were calculated using the generic inverse variance method in RevMan.20 ORs were deemed statistically significant if the 95% CI did not include the number 1. We conducted a meta-analysis of all eligible studies as our primary analysis, and also conducted a sensitivity analysis of the peer-reviewed publications only.


A total of 869 hits were generated from our search including 5 duplicates, leaving 864 unique search results. Of these, 426 were excluded based on the title and abstract. The remaining 438 publications were assessed as full texts. Eleven case–control studies were ultimately selected for inclusion; all had road traffic collisions as the outcome of interest. There were no cohort studies, and no studies investigating other safety outcomes that met our eligibility criteria. The study selection process is shown in Figure 1.

Figure 1
Figure 1:
Study selection.

The characteristics of the included studies are detailed in Table 2. Our main analysis is illustrated in Figure 2. We found that the presence of cannabis (as indicated by the parent compound THC or its carboxy metabolite, 11-nor-9-carboxy-Δ9-THC, or THC-COOH) resulted in an increased risk estimate of road traffic collisions, with an OR of 2.49 (95% CI 1.68–3.71, P < 0.00001, n = 49,870). When including only the peer-reviewed studies, the OR was 2.84 (95% CI 1.71–4.71, P < 0.0001, n = 38,947). Both these meta-analyses were performed with the random effects model.

Table 2
Table 2:
Characteristics of Included Studies
Figure 2
Figure 2:
Effect of cannabis on road traffic collisions (“Log” here refers to the natural logarithm; CI - confidence interval; df - degrees of freedom; IV - inverse variance; Random - random effects model; SE - standard error).


The results of our systematic review and meta-analysis demonstrate a robust risk increase, exceeding a doubling of the risk, with cannabis consumption for the outcome of road traffic collisions. These findings warrant urgent attention from the occupational health and safety community.

Extrapolation of data is commonly used for occupational risk assessment, as mentioned above. The skills (cognitive and other) required for safe, effective, and predictable functioning in safety-sensitive settings typically overlap with those needed to safely operate a motor vehicle, so that crash risk data have been utilized here as a proxy for determining levels of impairment in other safety-sensitive work settings. However, we acknowledge the limitation that the known functional impacts of cannabis use on driving may not map precisely to other safety-sensitive tasks.

Our results are consistent with 2 earlier meta-analyses, by Asbridge et al21 and Li et al,22 suggesting that cannabis consumption is associated with an increased motor vehicle crash risk. Li et al22 calculated an OR of 2.66 (95% CI 2.07–3.41) for the risk of being involved in motor vehicle crashes; Asbridge et al21 found that cannabis consumption is associated with an increased risk for motor vehicle crashes with an OR of 1.92 (95% CI 1.35–2.73). The magnitude of these risk estimates was challenged by Rogeberg and Elvik,23 whose calculations were in turn contested by Gjerde and Morland.24 Another systematic review by Hostiuc et al25 concluded that the effect size for driving under the influence of cannabis on unfavourable traffic events was not statistically significant, based on an OR with a confidence interval that did not include 1, but with a prediction interval that did include 1.

Our systematic review and meta-analysis serve as a basis to suggest cannabis is not recommended for workers who perform safety-sensitive tasks. This includes operation of motor vehicles and other machinery and heavy equipment, working with corrosive chemicals or high-pressure valves, high-voltage electric lines, and any other task where impaired performance could result in risk of injury or damage to the employee, others, or the environment.21

Cannabinoid products in Canada include warnings of potential risks in their monographs as per Health Canada's regulations; for Sativex (THC plus CBD; manufactured by GW Pharma Ltd., Histon, Cambridge, UK), it reads “patients should be warned not to drive or engage in activities requiring unimpaired judgement and coordination.”26(p.8) The Cesamet (nabilone; manufactured by Valeant Canada limitée/Limited, Montreal, Quebec, Canada) warning similarly says that patients “should not be permitted to drive or engage in dangerous tasks until the effects of nabilone are no longer present.”27(p.4) The monograph for CBD,28 the active ingredient of Epidiolex (manufactured by Greenwich Biosciences, Inc., Carlsbad, California, USA; licensed in the United States but not available in Canada), alerts physicians that patients may have impaired physical or mental abilities when taking Epidiolex, and “must be cautioned about performing tasks that require mental alertness (e.g., operating machinery or driving).”28(p.2)

There is no dependable evidence to suggest that legalized recreational cannabis, or cannabis for medical purposes, is less impairing than pharmaceutical cannabinoids. The occupational physician, in the absence of direct empirical evidence, should not be expected to be able to declare that cannabis products are safe for use. This reasoning would also apply to CBD-only or CBD-predominant preparations, where there is a lack of evidence to conclusively prove that CBD is not associated with impairing effects.

A limitation of our systematic review and meta-analysis that should be considered is that the evidence base discussed here consists of 11 studies about motor vehicle crashes, with no studies about nondriving workplace safety incidents. There may be cannabis-related impairments that are not apparent in traffic collision research, or, conversely, driving-related performance issues that do not apply to a specific safety-sensitive job.

As driving safety refers to a normobaric environment, work in hypo- or hyperbaric environments (e.g., diving, aviation) will need to be assessed separately.

Another limitation of our analysis is that some cannabis-consuming workers may not be measurably impaired, or the circulating THC level may not reach the threshold for impairment. However, there is presently no reliable evidence to confirm a safe level of THC. We therefore suggest that the use of cannabis is not recommended for persons working in safety-sensitive work settings.

Considerable uncertainty exists regarding the duration of impairment after cannabis use, which depends on the cannabinoids consumed and their doses. Estimating impairment in practice is further complicated by interindividual variability. However, it can be argued that practical guidance is needed even if there are only limited data from which to conclude the duration of impairment in individual workers. The evidence that we do have suggests that the duration of impairment can last much longer than 4 hours. Until there is definitive evidence, it would therefore appear prudent to restrict any safety-sensitive duties for at least 24 hours, or for longer, if there is concern about risk of ongoing impairment. Needless to say, each worker is unique, and individual assessment is needed for physicians to appropriately determine fitness to work.

Position Statement of the Occupational and Environmental Medical Association of Canada

Taking the evidence presented here into consideration, we formulated a Position Statement on cannabis use and safety-sensitive work, which has been endorsed by the Occupational and Environmental Medical Association of Canada (OEMAC). Further information can be found on the OEMAC website at The Canadian Society of Addiction Medicine has now endorsed this OEMAC position. Further information can be found on the Canadian Society of Addiction Medicine website:


We are grateful to Ms L. Dennett (University of Alberta Libraries) for the design and execution of our literature search.

Funding for the project “Marijuana in the Workplace: Position Paper” was provided by an Occupational Health and Safety Innovation and Engagement Grant from the Government of Alberta to SS.


1. Fan X, Els C, Corbet KJ, et al. “Decision-critical” work: a conceptual framework. J Occup Med Toxicol 2016; 11:22.
2. Parker v. ONCA, (2000), 49 OR (3rd) 481.
3. Government of Canada. Speech from the Throne to Open the 1st Session 42nd Parliament of Canada. 2015; Ottawa, ON: Government of Canada, Available from: Accessed December 24, 2018.
4. Bill C-45, Cannabis Act, 1st Sess, 42nd Parl, 2018, c. 16. (as passed by the House of Commons 21 June 2018).
5. Government of Canada, Health Canada. Marihuana for Medical Purposes Regulations. 2016; Ottawa, ON: Government of Canada, Health Canada, Available from: Accessed December 24, 2018.
6. Allan GM, Ramji J, Perry D, et al. Simplified guideline for prescribing medical cannabinoids in primary care. Can Fam Physician 2018; 64:111–120.
7. United Nations Office on Drugs and Crime (UNODC). World Drug Report 2014. 2014; Vienna, AT: United Nations Office on Drugs and Crime (UNODC), 93. Available from: Accessed December 24, 2018.
8. Government of Canada, Statistics Canada on behalf of Health Canada. Canadian Tobacco, Alcohol and Drugs Survey (CTADS): 2015 Summary (Corrections Posted March 2017). 2017; Ottawa, ON: Government of Canada, Statistics Canada on behalf of Health Canada, Available from: Accessed December 24, 2018.
9. Government of Canada, Health Canada. Canadian Cannabis Survey 2017—Summary. 2017; Ottawa, ON: Government of Canada, Health Canada, Available from: Accessed December 24, 2018.
10. Rovetto LJ, Aieta NV. Supercritical carbon dioxide extraction of cannabinoids from Cannabis sativa L. J Supercrit Fluids 2017; 129:16–27.
11. Volkow ND, Baler RD, Compton WM, et al. Adverse health effects of marijuana use. N Engl J Med 2014; 370:2219–2227.
12. Di Forti M, Sallis H, Allegri F, et al. Daily use, especially of high-potency cannabis, drives the earlier onset of psychosis in cannabis users. Schizophr Bull 2014; 40:1509–1517.
13. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 5th ed.Arlington: American Psychiatric Pub; 2013.
14. Phillips JA, Holland MG, Baldwin DD, et al. Marijuana in the workplace: guidance for occupational health professionals and employers: joint guidance statement of the American Association of Occupational Health Nurses and the American College of Occupational and Environmental Medicine. J Occup Environ Med 2015; 57:459–475.
15. College of Family Physicians of Canada. Authorizing Dried Cannabis for Chronic Pain or Anxiety: Preliminary Guidance. 2014; Mississauga, ON: College of Family Physicians of Canada, Available from: Accessed December 24, 2018.
16. Ramaekers JG, Kauert G, van Ruitenbeek P, et al. High-potency marijuana impairs executive function and inhibitory motor control. Neuropsychopharmacology 2006; 31:296–303.
17. Leirer VO, Yesavage JA, Morrow DG. Marijuana carry-over effects on aircraft pilot performance. Aviat Space Environ Med 1991; 62:221–227.
18. Canadian Railway Medical Rules Handbook. Railway Association of Canada. 2016; Ottawa, ON: Railway Association of Canada, 208. Available from: Accessed December 24, 2018.
19. Hegmann KT, Weiss MS, Bowden K, et al. ACOEM practice guidelines: opioids for treatment of acute, subacute, chronic, and postoperative pain. J Occup Environ Med 2014; 56:143–159.
20. The Nordic Cochrane Centre, The Cochrane Collaboration. Review Manager (RevMan) [Computer Program.]. Version 5.3. Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration; 2014.
21. Asbridge M, Hayden JA, Cartwright JL. Acute cannabis consumption and motor vehicle collision risk: systematic review of observational studies and meta-analysis. BMJ 2012; 344:e536.
22. Li M, Brady JE, DiMaggio CJ, et al. Marijuana use and motor vehicle crashes. Epidemiol Rev 2011; 34:65–72.
23. Rogeberg O, Elvik R. The effects of cannabis intoxication on motor vehicle collision revisited and revised. Addiction 2016; 111:1348–1359.
24. Gjerde H, Mørland J. Risk for involvement in road traffic crash during acute cannabis intoxication. Addiction 2016; 111:1492–1495.
25. Hostiuc S, Moldoveanu A, Negoi I, et al. The association of unfavorable traffic events and cannabis usage: a meta-analysis. Front Pharmacol 2018; 9:1–14.
26. Bayer Canada Inc.. Product Monograph: Sativex. 2015; Mississauga, ON: Bayer Canada Inc., Available from: [Cited December 21, 2018].
27. Valeant Canada limitée/Limited. Product Monograph: Cesamet. 2009; Montreal, QC: Valeant Canada limitée/Limited, Available from: [Cited December 21, 2018].
28. Greenwich Biosciences, Inc. Cannabidiol (Epidiolex) [monograph]. Lexicomp® [Database on the Internet] 2018; Carlsbad, CA: Greenwich Biosciences, Inc, [Cited December 24, 2018].
29. Assum T. The Prevalence and Relative Risk of Drink and Drug Driving in Norway. 2005; Oslo, Norway: Institute of Transport Economics, 48.
    30. Bogstrand ST, Gjerde H. Which drugs are associated with highest risk for being arrested for driving under the influence? A case–control study. Forensic Sci Int 2014; 240:21–28.
    31. Brault M, Dussault C, Bouchard J, Lemire AM. The contribution of alcohol and other drugs among fatally injured drivers in Quebec: final results [Conference Proceedings]. Jean-Lesage (QC): Proceedings of the 17th International Conference on Alcohol, Drugs and Traffic Safety; 2004.
      32. Chihuri S, Li G, Chen Q. Interaction of marijuana and alcohol on fatal motor vehicle crash risk: a case–control study. Inj Epidemiol 2017; 4:8.
        33. Del Balzo G, Gottardo R, Mengozzi S, et al. “Positive” urine testing for Cannabis is associated with increased risk of traffic crashes. J Pharm Biomed Anal 2018; 151:71–74.
        34. Gjerde H, Christophersen AS, Normann PT, et al. Associations between substance use and fatal road traffic accidents among car and van drivers in Norway: a case–control study. Transp Res Part F 2013; 17:134–144.
          35. Kuypers KPC, Legrand S, Ramaekers JG, et al. A case–control study estimating accident risk for alcohol, medicines and illegal drugs. PLoS One 2012; 7:e43496.
          36. Marquet P, Delpla PA, Kerguelen S, et al. Prevalence of drugs of abuse in urine of drivers involved in road accidents in France: a collaborative study. J Forensic Sci 1998; 43:806–811.
          37. Mathijssen R, Houwing S. The Prevalence and Relative Risk of Drink and Drug Driving in the Netherlands: A Case–Control Study in the Tilburg Police District. 2005; Leidschendam, The Netherlands: SWOV, 57.
            38. Mura P, Kintz P, Ludes B, et al. Comparison of the prevalence of alcohol, cannabis and other drugs between 900 injured drivers and 900 control subjects: results of a French collaborative study. Forensic Sci Int 2003; 133:79–85.
            39. Pelissier AL, Leonetti G, Kerguelen S, et al. Urinary screening of drugs of abuse among drivers involved in road accidents. Ann Biol Clin 1996; 54:365–371.

            cannabis; motor vehicle collisions; safety-sensitive work; cannabis; travail critique pour la sécurité; collisions de véhicules à moteur

            © 2019 by Lippincott Williams & Wilkins, Inc.