Drug and Driving: A Compendium of Research Studies

Annotated Sources


1. Bates, M.N. and Blakely, T.A. (1999)

Role of cannabis in motor vehicle crashes. Epidemiology Review 21: 222-232


Review article of the role of cannabis in motor vehicle collisions

Type of study, population(s) and proportion tested

Review of 5 existing studies examining the risks associated with cannabis in motor vehicle collisions
Re-analysis of data was presented for most studies reviewed
Culpability analysis

Drugs examined
  • Cannabinoids
  • Some studies also examined cannabis in combination with alcohol
Method of testing and medium used

Various methods were used in different studies

Other dependent variables
Findings (including statistical methods)
  • 3 studies show reduced risk of culpability for fatal crash involvement -- weighted OR of .59, p=.05

  • 2 injury studies -- one shows increased risk, one shows decreased risk

  • Risk of alcohol/cannabis combination higher than alcohol alone but not significantly so (small number of cases)

  • Author identifies several limitations of culpability studies

  • Drivers who have used cannabis modify their driving behaviour to compensate for perceived impairment by driving slower, taking fewer risks, and leaving more headway

  • Not possible to exclude possibility that cannabis alone or in combination leads to increased risk of crash in less serious crashes

2. Gemmell, C.; Moran, R.; Crowley, J.; and Courtney, R. (1999).

Literature Review on the Relation between Drug Use, Impaired Driving and Traffic Accidents. CT.97.EP.14. Lisbon, Portugal: European Monitoring Centre for Drugs and Drug Addiction


This report summarizes the general findings of drugged driving, with respect to a variety of drugs, and discusses some of the problems associated with epidemiological and experimental studies in this area

Type of study, population(s) and proportion tested

Literature review

Drugs examined (threshold values for detection)
  • Alcohol
  • Methadone
  • Cannabis
  • Benzodiazepines
  • Antihistamines
  • Antidepressants
  • Amphetamines
  • Ecstasy and other synthetic drugs
Findings (including statistical methods)
Epidemiological studies:
Unclear whether accidents occur as a direct result of the drug in question
Often no comparison group
Drugs often mixed with alcohol
Presence of drugs does not necessarily imply impairment
Simulation techniques:
Introduces artificiality and ethical issues, with respect to giving high doses of drugs to subjects
No drug found as frequently as alcohol
Not found often and studies do not suggest sufficient impairment
One of the most prevalent drugs, though often found mixed with alcohol, which is more likely to result in an increased risk of accident
Most frequently detected licit drug which can double the risk of accident, especially in older (over 65) drivers and when mixed with alcohol
Seldom suggested as causal in accidents
Effects of severe depression may have greater impact on Driving
Insufficient evidence, largely due to lack of controlled studies, but, in high does, it is likely to increase driving impairment
Ecstasy and other synthetic drugs:
More research is required, but they are likely to be a danger to driving

3. Kelly, E., Darke, S. and Ross, J. (2004). A review of drug use and driving: Epidemiology, impairment, risk factors and risk perceptions.

Drug and Alcohol Review 23: 319-344


Review of drug use and driving literature

Type of study, population(s) and proportion tested


Drugs examined (threshold values for detection)
  • Cannabis
Method of testing and medium used
Other dependent variables
Findings (including statistical methods)
  • Drug driving prevalence approximately 4% in past 12 months
  • Up to 25% of crash-involved drivers positive for drugs
  • Cannabis is the most common, followed by benzodiazepines, cocaine, amphetamines, opioids -- polydrug use common, as is overlap with alcohol
  • Cannabis and benzodiazepines increase risk
  • Unclear as to whether there exists an association between drug use problems and drug driving
  • Less concern about impaired driving among drug drivers and drink drivers -- perception is that drugs do not significantly impair driving performance
  • Risk perceptions vary according to drug type -- cannabis seen as producing less impairment than others, including alcohol
  • Research and interventions should focus on high-risk groups.

4. Ramaekers, J. G., Berghaus, G., van Laar M., and Drummer, O. H. (2004). Dose Related risk of motor vehicle crashes after cannabis use.

Drug and Alcohol Dependence 73: 109-119


Studies of cannabis prevalence and relative risk to driving are reviewed. Both epidemiological and experimental studies are discussed.

Type of study, population(s) and proportion tested

Literature review

Drugs examined (threshold values for detection)
  • Cannabis
  • Alcohol
Findings (including statistical methods)
Epidemiological studies:
  • Prevalence of THC between 4 and 14% in injured or fatally injured drivers
  • In the absence of control groups, role of THC in collisions cannot be determined
Culpability studies:
  • Cannabis in combination with alcohol have shown significant increases in culpability rates
  • Looking specifically at THC presence, crash culpability increases with increases doses of THC (OR of 2.7 compared to drug-free drivers)
Case-control studies:
  • OR for accident or injury, for those with THC present or having used cannabis within so many hours of driving, ranged from 2.3-2.9 When in combination with alcohol, OR ranged from 4.6-80.5
Experimental studies:
  • In interactive simulators (dose up to 200 kg), THC has been found to increase lateral position variability, headway variability, and speed variability and cause subjects to ignore navigational information, hit roadway obstacles, and react more slowly to secondary demands
  • In actual driving situations, (THC dose of 8.4 mg) concentration, care while driving, and judgement were significantly affected
  • When cannabis combined with alcohol, more severe performance impairments were found

5. Thomas, R.E. (1998). Benzodiazepam use and motor vehicle accidents -- systematic review of reported association.

Canadian Family Physician 44: 799-808


Review of relationship between benzodiazepine use and motor vehicle collisions

Type of study, population(s) and proportion tested

Review of case-control, police or emergency studies and experimental driving studies of the effects of benzodiazepines

Drugs examined (threshold values for detection)


Method of testing and medium used
Other dependent variables

Examines impaired driving, crashes, mortality, medical cases, hospitalization

Findings (including statistical methods)
  • OR for mortality and emergency medical treatment range 1.45 to 2.4
  • In police and emergency ward studies, benzodiazepines a factor in 1% to 56% of crashes (usually 5 to 10%)
  • In cases where BAC was below legal limit, benzodiazepines were found in 43 to 65% of cases
  • In controlled one controlled study, 5% of drivers and 2% of controls in crashes had benzodiazepines present
  • Benzodiazepines double risk of motor vehicle collision
  • For driver >65 years of age, risk is higher when taking longer acting and larger quantities of benzodiazepine

6. Ward, N.J. and Dye, L. (1999). Cannabis and Driving: A Literature Review and Commentary.

Road Safety Research Report, No. 12. London: Department for Transport, Local Government and the Regions


Literature review on cannabis and driving

Type of study, population(s) and proportion tested

Literature review and commentary

Drugs examined (threshold values for detection)


Method of testing and medium used
Other dependent variables
Findings (including statistical methods)
  • Cannabis was considered to cause clear impairment on laboratory tasks such as tacking and attention
  • Impairment was not as pronounced on real driving or simulator tasks
  • Compensatory effort could be invoked to offset the impairing effects 4-12% of fatalities test positive for cannabis -- most cases confounded by alcohol
  • Baseline data for non-crash cases absent
  • Authors suggest that the equivalent of legal alcohol limit (80 mg/100 ml) for cannabis is 11 ng/ml THC
  • Study notes that the research has been piecemeal with numerous problems and gaps. No firm conclusions could be drawn.
Date modified: