Risk of road accident associated with the use of drugs: A systematic review and meta-analysis of evidence from epidemiological studies

Highlights

• The risk of road accident associated with use of drugs is assessed.

• A meta-analysis is performed based on 264 estimates of risk.

• Estimates of risk are developed for eleven drugs.

• Risk is found to be associated with the use of all these drugs.

• The poor quality of many studies limits causal inference.

Abstract

This paper is a corrigendum to a previously published paper where errors were detected. The errors have been corrected in this paper. The paper is otherwise identical to the previously published paper. A systematic review and meta-analysis of studies that have assessed the risk of accident associated with the use of drugs when driving is presented. The meta-analysis included 66 studies containing a total of 264 estimates of the effects on accident risk of using illicit or prescribed drugs when driving. Summary estimates of the odds ratio of accident involvement are presented for amphetamines, analgesics, anti-asthmatics, anti-depressives, anti-histamines, benzodiazepines, cannabis, cocaine, opiates, penicillin and zopiclone (a sleeping pill). For most of the drugs, small or moderate increases in accident risk associated with the use of the drugs were found. Information about whether the drugs were actually used while driving and about the doses used was often imprecise. Most studies that have evaluated the presence of a dose-response relationship between the dose of drugs taken and the effects on accident risk confirm the existence of a dose-response relationship. Use of drugs while driving tends to have a larger effect on the risk of fatal and serious injury accidents than on the risk of less serious accidents (usually property-damage-only accidents). The quality of the studies that have assessed risk varied greatly. There was a tendency for the estimated effects of drug use on accident risk to be smaller in well-controlled studies than in poorly controlled studies. Evidence of publication bias was found for some drugs. The associations found cannot be interpreted as causal relationships, principally because most studies do not control very well for potentially confounding factors.

Introduction

It is well-established that driving under the influence of alcohol increases the risk of accident involvement. This has been known at least since the famous Grand Rapids study was made in the early nineteen sixties (Borkenstein et al., 1964). Less is known about the effects of drugs (medicinal or illicit) on the risk of accident involvement. A few systematic literature reviews and meta-analyses of the effects of drugs on accident risk have been reported (Thomas, 1998, Bates and Blakely, 1999, Ramaekers et al., 2004, Baldock, 2007, Orriols et al., 2009, Rapoport et al., 2009, Smink et al., 2010, Dassanayake et al., 2011, Asbridge et al., 2012). These studies deal only with a single drug or a few drugs and not all of them include a meta-analysis providing a summary estimate of the effect of drug use on accident risk.

Thomas (1998) reviewed studies of the association between benzodiazepine use and motor vehicle accidents. He listed 23 estimates of risk (Table 3 of the paper). Twelve of these indicated an odds ratio of accident involvement for users of benzodiazepines of between 1.01 and 1.50. Three estimates indicated an odds ratio between 2.01 and 2.50. Thomas concluded that use of benzodiazepines approximately doubles the risk of motor vehicle accidents. The study did not include a meta-analysis of the estimates of risk.

Bates and Blakely (1999) reviewed studies of the role of cannabis in motor vehicle accidents. The study did not include a meta-analysis. It listed the findings of a few studies and concluded that there was no evidence that the use of cannabis alone increased the risk of being held culpable for an accident. The authors added that it cannot be ruled out that use of cannabis leads to an increased risk of accidents causing less serious injuries or property damage. Ramaekers et al. (2004) argued that the effect of cannabis on the risk of accident involvement depends on the dose taken and on how long after taking cannabis driving takes place. They pointed out that the absence of a relationship between cannabis use and risk of accident involvement in some studies is probably attributable to the fact that these studies only found inactive metabolites of cannabis in body fluids. Metabolites of cannabis can persist for a long time after it was taken, particularly in urine. The study did not include a meta-analysis.

Baldock (2007) reviewed the literature on cannabis and the risk of accident involvement. The review was a traditional narrative review and did not include a meta-analysis. Baldock argued that many studies have methodological flaws, in particular with respect to the control for potentially confounding factors.

Orriols et al. (2009) presented a systematic review of studies of the risk associated with the use of medicinal drugs. The review included 22 studies of variable methodological quality. An assessment of study quality was made and studies rated as good, average or poor. A meta-analysis was not performed. It was concluded that the use of benzodiazepines is associated with an increased risk of accident, but that there is too little evidence to conclude anything for other medicinal drugs. Poor control for confounding factors was cited as a weakness of many studies.

Rapoport et al. (2009) reported a meta-analysis of benzodiazepine use and accident risk. The meta-analysis was based on six case–control studies and three cohort studies (a short definition of study designs is given later in this paper). The summary estimate of the odds ratio of accident involvement for benzodiazepine users was 1.61 according to the case–control studies and 1.60 according to the cohort studies. The meta-analysis did not score studies formally with respect to study quality and did not test for publication bias. Smink et al. (2010) conducted a systematic literature review of studies assessing the relationship between use of benzodiazepines and accident involvement, but did not perform a meta-analysis.

Dassanayake et al. (2011) performed a systematic literature review and meta-analysis of studies of the effects on accident risk of benzodiazepines, antidepressants and opioids. A meta-analysis was only feasible for studies of benzodiazepines. The summary estimates of the odds ratio of accident involvement for benzodiazepine users were 1.59 for case–control studies, 1.81 for cohort studies and 1.41 for culpability studies. These estimates are close to those reported by Rapoport et al. (2009). The study did not score studies formally for quality and did not test for publication bias.

Asbridge et al. (2012) conducted a meta-analysis of observational studies of the effects of acute cannabis use on the risk of accident involvement. Nine studies were included. The summary estimate of the odds ratio of accident involvement associated with use of cannabis was 2.10 for fatal accidents and 1.74 for non-fatal accidents. Study quality was scored formally by means of the Newcastle-Ottawa quality assessment scale. A test for the possible presence of publication bias was not included.

The systematic reviews and meta-analyses quoted above included only a few drugs, in particular benzodiazepines and cannabis. Not all reviews included a meta-analysis. Not all meta-analyses considered study quality. No meta-analysis addressed the possibility of publication bias.

The aim of this paper is to summarize current knowledge regarding the risks associated with the use of drugs while driving. The paper seeks to improve previous reviews by: (1) including as many drugs as possible in the systematic literature review and meta-analysis; (2) assessing study quality and testing how it influences study findings; (3) testing and adjusting for the possible presence of publication bias. Alcohol is not included in this study. The focus is on drugs used in regular medical treatment or illicit drugs used recreationally.