The involvement of drugs in drivers of motor vehicles killed in Australian road traffic crashes
Introduction
Alcohol is recognised as a leading contributor to road trauma. Alcohol over-involvement in crashes has been clearly demonstrated by a number of studies which show substantial increases in crash risk when the blood alcohol concentration (BAC) exceeds 0.10 g per 100 ml (%) (Borkenstein et al., 1974, Mounce and Pendleton, 1992, Robertson and Drummer, 1994).
While many drugs detected in crash victims are liable to impair driving skills, there is still uncertainty as to whether this translates to an increased crash risk. Likely drugs include cannabis; benzodiazepines; opiate-like drugs, such as heroin, morphine and methadone; and amphetamines and other CNS stimulants.
Experimental studies using instrumented cars have provided much useful information on the role of certain drugs on performance and error rates. Many CNS active drugs, particularly cannabis, benzodiazepines, barbiturates and the sedating antihistamines reduce lane control by increasing the standard deviation of lateral position (SDLP) while the drug is having its peak activity (Brookhuis et al., 1990, Laurell and Tornros, 1986, O’Hanlon et al., 1982, O’Hanlon and Volkerts, 1986, Ramaekers et al., 2000, Robbe, 1994). Logan et al. (2000) examined the extent of driver impairment of carisoprodol, a skeletal muscle relaxant, and its major metabolite meprobamate, which has sedative properties. The authors found that at therapeutic concentrations impairment was possible with symptoms of intoxication similar to alcohol.
Unfortunately, these experimental studies may not accurately predict the effects of drugs under actual driving conditions. The association of drugs with crashes has been extensively investigated in epidemiological studies but with mixed results. Moreover, while providing evidence of an association, epidemiological methods cannot unequivocally establish that drug use causes adverse driving events.
Early studies of drugs and crash risk concluded that cannabis-users are more likely to be responsible for their crashes than drug-free drivers (Simpson, 1986, Simpson et al., 1982, Warren et al., 1981). The epidemiological design providing the strongest evidence of a causal association of drug use and crash risk is case-control responsibility analysis. Terhune et al. (1992) examined the cases of almost 2000 drivers fatally-injured in the USA to assess the contribution of drugs to crashes and found that the responsibility rate for amphetamine-positive drivers was higher than the drug-free group (Terhune et al., 1992). However, Terhune (Terhune et al., 1992) also found the responsibility rate for all THC-positive drivers was lower than that for the drug-free control group. Williams also found no evidence of an association between cannabis and the risk of injury but the numbers of drivers were small (Williams et al., 1985). In contrast, an investigation of trucker fatalities pointed to an adverse effect of Δ9-tetrahydrocannabinol (THC) over 1 ng/ml, and other psychotropic drugs, on crash risk (Crouch et al., 1993).
A study of drivers injured in South Australia found drivers who tested positive for alcohol only, benzodiazepines only and the combinations of alcohol and THC and alcohol and benzodiazepines were significantly more likely to be culpable for the crash compared with the drug-free group (Longo et al., 2000, Longo et al., 2001). A small percentage of drivers who only tested positive for THC were culpable for the crash but this was not statistically significant. It was of interest that the blood THC concentrations were particularly low, most being <2 ng/ml. A study of injured drivers who presented to an urban emergency centre in Colorado found only alcohol and alcohol in combination with other drugs increased the likelihood they were responsible for the crash (Lowenstein and Koziol-Mclain, 2001). In this study cannabinoids including THC were only measured in urine limiting the time frame of last cannabis use.
Other case-control designs compare the drug use of drivers involved in crashes with other groups. A nested case-control design of over 200,000 drivers using driver’s license files, police reports of injurious crashes, and health insurance records showed an increased risk of motor vehicle crash involvement in the elderly population using long acting benzodiazepines (Hemmelgarn et al., 1997).
A case-control study of persons involved in injurious crashes showed that use of anti-depressants and opioid analgesics by older drivers was associated with increased risk of injurious motor vehicle collisions, but not with benzodiazepines or sedating antihistamines (Leveille et al., 1994). This contrasted with another study that found the relative risk of injurious crash involvement for current users of any psychoactive drug was significantly elevated and the increase was primarily due to benzodiazepines and cyclic antidepressants (Ray et al., 1992).
Other studies of all drivers have failed to find a positive association of drugs with road crashes. A retrospective, hospital-based case-control study failed to find a significant difference between the prevalences of drugs (opiates, cannabinoids and amphetamines) in drivers injured in road crashes and those admitted to a hospital for other reasons (Marquet et al., 1998).
A study which linked prescription records with hospital admissions from road crashes showed that people who used minor tranquilizers in the past 3 months had a five-fold higher risk of a serious road accident (Skegg et al., 1979). A similar study showed the odds ratio was elevated for those persons taking benzodiazepines, particularly within a few weeks of the first prescription (Neutel, 1995). In contrast, a similar study showed no increase in accident risk with the use of benzodiazepines and sedatives (Jick et al., 1981). An association of drug use and motor crashes has also been shown by comparisons of the prevalence of drugs in crash victims with general drug use obtained from surveys. Other studies have surveyed drivers regarding their use of drugs while driving and their perception of possible impairment (Albery et al., 2000, Walsh and Mann, 1999, Wechsler et al., 1984).
Crash investigations have provided evidence of likely mechanisms of causality. Logan (1996) showed that in a population of methamphetamine positive drivers, predominantly culpable for accident causation, the use of methamphetamine most likely contributed to risk-taking behaviour or was a result of withdrawal-related fatigue and hypersomnolence (Logan and Schwilke, 1996).
It is clear that variable conclusions have emanated from various sources of any link of crash risk to use of one or more psychotropic drugs. Some studies have failed to demonstrate a convincing association of drugs with crash risk often because they lacked a suitable control population while others lacked statistical power because of their small population sizes, limited extent of toxicology testing or their assessment of culpability.
The present study was designed to avoid these problems. A validated method of responsibility analysis (Robertson and Drummer, 1994) was used to classify culpability of drivers killed in three Australian states over 10 years. The toxicological data on the incidence of alcohol and the various drugs for the 3398 subjects have been described elsewhere (Drummer et al., 2003).
Section snippets
Study population
The study population is drivers killed in motor vehicle crashes in the three Australian states of Victoria (Vic.), New South Wales (NSW) and Western Australia (WA). Only cases that were on-road motor vehicle crashes were included. Crashes that occurred off-road, or those classified as due to natural causes or suicide were excluded.
In Victoria these data were obtained from records kept at the Victorian Institute of Forensic Medicine and the State Coroners Office at Southbank. Drivers were
Characteristics of drivers
Three thousand three hundred and ninety eight drivers, were included in the study: 1611 Victorian (47.4% of study population), 1031 NSW (30.3%) and 757 Western Australian drivers (22.3%). This represented 89% (median) of the driver fatalities identified from each jurisdiction. Cases were excluded if the deceased was not a driver, if they died of natural causes or committed suicide, if the crash did not occur on a public road and if either the toxicology data or culpability data were incomplete.
Discussion
This 10-year, multi-centre longitudinal study of driver culpability has shown a number of strong positive associations between use of some psychoactive drugs and responsibility for the crash. Of particular interest is the association of THC with driver culpability. This association showed a biological gradient, similar to that observed for alcohol. The estimated association with culpability of THC in concentrations of at least 5 ng/ml was much greater than the association of all identifiable
Acknowledgements
We thank the staff of the Victorian Institute of Forensic Medicine for their assistance in these investigations. We also thank the toxicologists in each state for their support and assistance, particularly Mr. Robert Hansson (WA) and Mr. Alan Hodda (NSW), as well as coroners, registrars and clerical assistants at the respective coroners’ courts. We acknowledge the financial assistance of VicRoads for financing many of these studies, as well as AustRoads and the NSW roads and traffic authority.
References (45)
- et al.
Illicit drugs and driving: prevalence, beliefs and accident involvement among a cohort of current out-of-treatment drug users
Drug Alcohol Depend.
(2000) - et al.
Association of road-traffic accidents with benzodiazepine use
Lancet
(1998) - et al.
The prevalence of alcohol, cannabinoids, benzodiazepines and stimulants amongst injured drivers and their role in driver culpability. Part II. The relationship between drug prevalence and drug concentration, and driver culpability
Accid. Anal. Prev.
(2000) - et al.
The relationship between blood alcohol concentration and crash responsibility for fatally injured drivers
Accid. Anal. Prev.
(1992) Risk of traffic accident injury after a prescription for a benzodiazepine
Ann. Epidemiol.
(1995)- et al.
Responsibility analysis: a methodology to study the effects of drugs in driving
Accid. Anal. Prev.
(1994) - et al.
Epidemiology of road accidents involving young adults: alcohol
Drug Alcohol Depend.
(1982) - Berghaus, G., Scheer, N., Schmidt, P., 1995. Effects of cannabis on psychomotor skills and driving performance—a meta...
- Borkenstein, F.R., Crowther, R.F., Shumate, R.P., Zeil, W.B., Zylman, R., 1974. The role of the drinking driver in...
- et al.
Repeated dose effects of lormetazepam and flurazepam upon driving performance
Eur. J. Clin. Pharmacol.
(1990)
The prevalence of drugs and alcohol in fatally injured truck drivers
J. Forensic Sci.
Benzodiazepines—effects on human performance and behavior
Forensic Sci. Rev.
Benzodiazepine use and the risk of motor vehicle crash in the elderly
JAMA
Sedating drugs and automobile accidents leading to hospitalization
Am. J. Public Health
Visual search and urban driving under the influence of marijuana and alcohol
Human Psychopharmacol. Clin. Exp.
The carry-over effects of triazolam compared with nitrazepam and placebo in acute emergency driving situations and in monotonous simulated driving
Acta Pharmacol. Toxicol. (Copenh)
Opioid dependence and driving ability: a review in the context of proposed legislative change in Victoria
Drug Alcohol Rev.
Psychoactive medications and injurious motor vehicle collisions involving older drivers
Epidemiology
Methamphetamine—effects on human performance and behavior
Forensic Sci. Rev.
Methamphetamine and driving impairment
J. Forensic Sci.
Cited by (392)
Cannabis, cognitive impairment and car crash risk
2023, Cannabis Use, Neurobiology, Psychology, and TreatmentCorrelation between oral fluid and blood THC concentration: A systematic review and discussion of policy implications
2022, Accident Analysis and PreventionAn assessment of the non-fatal crash risks associated with substance use during rush and non-rush hour periods in the United States
2022, Drug and Alcohol DependenceThe incidence of psychoactive substances and alcohol among impaired drivers in Denmark in 2015–2019.
2022, Forensic Science InternationalDevelopment of a simulation-based experimental research framework for the characterization of cannabis-related driving impairment
2022, Transportation Research Interdisciplinary PerspectivesTrajectories and outcomes of adolescents that ride with an impaired driver/drive while impaired
2022, Journal of Transport and Health