Health Protection Branch

Safety Concerns

Safety is of major importance warranting independent consideration for a number of reasons. Although the impaired driver may be putting himself at risk, he may potentially harm others as well, so our concern here is now focused on risk of harm to others which may arise from cannabis use. Secondly, direct "harm to others" describes a moral category of behaviour that all schools of legal philosophy agree justifies the use of the criminal law power. Finally concern about safety raises a host of self-contained legal problems that deserve special attention.

The cannabis and safety issue primarily involves the operation of motor vehicles on the roads. The use of heavy machinery and flying are further illustrations of situations where cannabis use may compromise performance and thus put the safety of others at risk. Driving, however, remains the archetypal, most commonly occurring, and best-researched example of an activity which can endanger public safety. In addition, its apparent resemblance to the alcohol-and-driving situation has prompted considerable public discussion and a concomitant search for a "Breathalyzer" type of testing device, on the assumption that the problems anticipated will indeed occur.

A thorough review of the relevant research findings in this very complex field has recently been completed by R.A. Warren, a Research Associate at the Ottawa-based Traffic Injury Research Foundation. In general, his findings confirm the research and literature review presented by the Le Dain Commission in its 1972 Cannabis report.

In order to investigate possible traffic safety hazards associated with cannabis use, two major methodological strategies, experimental and epidemiological, have been adopted, each with its special strengths and weaknesses. Experimental studies are designed to investigate, under carefully controlled conditions, the potential cannabis effects on certain psychomotor skills thought to be important in safe driving. Epidemiological studies, on the other hand, search for evidence of already-existing "real world" involvement of cannabis in problematic driving, by examining users' attitudes or experiences, studying driving records, and systematically investigating accidents.

Epidemiological studies are frequently very difficult to control properly and can generally yield reliable predictive information only if the incidence of drug use in the driver population is already fairly substantial. To date, no comprehensive studies of cannabis involvement in traffic accidents have been undertaken, although a number of limited efforts have been made. There is some evidence of an association between cannabis and driving mishaps, but the data are inconsistent. Other studies have not found evidence of such a relationship.

Care must be taken in interpreting limited or inadequately controlled epidemiological studies. For example, an apparent relationship between accidents and persons who have chosen to drive after consuming an illegal drug may derive from other preexisting characteristics, such as risk-taking tendencies, rather than resulting from the use of the drug. Age, sex and region are also essential factors which must be controlled.

Studies where active drug levels in the bodies of persons involved in accidents are compared with drug levels in control subjects who have not been associated with driving mishaps have been of primary importance in clarifying the traffic hazards of alcohol. However, there are chemical and pharmacological differences inherent in the cannabis situation which limit the feasibility of such an approach.

Considerable attention has been given to the possible development of a quantitative chemical test analogous to the Breathalyzer, which could provide a reasonable estimate of the intensity of current cannabis effects. Such a device would be invaluable for basic experimental and epidemiological research, as well as for traffic law enforcement. Although qualitative methods exist that can provide some evidence of fairly recent use (e.g., employing ether finger swabs, mouthwashes, dental scrapings and urine samples), these sampling techniques are often not appropriate and, in any event, cannot provide quantitative information regarding current effect levels. In fact, they cannot generally discriminate use immediately before testing from use several days prior to taking the sample.

Available evidence, based primarily on the relatively simple case of isolated THC administration, suggest that blood sample analysis is the most practical approach to estimating the intensity of effects from levels of cannabinoids in the body. However, current methods have little applicability outside the laboratory. Even if simplified and efficient blood sample techniques were developed, significant practical and legal problems surrounding the acquisition of appropriate samples would likely preclude their use for general traffic control purposes.

In the review noted above, Warren (1978) concluded that a causal relationship between cannabis use and driving impairment has not been demonstrated and that at present no epidemiological evidence that cannabis contributes to driving collisions has been found.

Since completing this review, Warren has submitted a confidential interim report on a project investigating the level of drugs in fluid and tissue specimens from persons killed in traffic accidents in Ontario over the past year. Evidence of cannabis use was found in a significant number of the victims. It is difficult to interpret this aspect of the study, due to the small numbers involved and the preliminary nature of the report. Disclosure of more complete results is anticipated later in the year.

There are different problems involved in the evaluation of experimental studies deriving from the fact that little is known about the actual causes of traffic accidents, and small changes in one or more of the complex of driving-related skills might not be significantly related to accidents or injuries under natural conditions. It appears that such factors as driver attitudes, risk-taking traits, aggression, judgement, attention and susceptibility to distraction, and a variety of other psychological variables which are difficult to measure may be more significant in contributing to automobile accidents than are elementary psychomotor skills. Therefore, although certain basic driving parameters can be established experimentally and important issues raised, such studies can provide only a limited basis for predicting the likely "real-world" effects.

Laboratory research indicates that cannabis can produce dose-dependent short-term deficits in certain perceptual, attentional, cognitive and psychomotor abilities which are of possible significance to driving. Similarly, a few studies have revealed that experienced airplane pilots undergo deterioration in performance on flight simulators after smoking high doses of marijuana. Several experiments have demonstrated analogous detrimental effects of cannabis on certain automobile driving tasks, on test tracks and on the road. Further, it appears that alcohol and cannabis have additive deleterious effects on driving skills when used together. On the other hand, several reports suggest that cannabis reduces aggression and risk-taking in a variety of situations. Taken as a whole, these experimental findings are certainly cause for concerned attention, although, as noted above, they cannot be used to directly predict traffic hazards under natural conditions.

Surveys suggest that among regular cannabis users driving while high is not uncommon and that the combined use of alcohol and cannabis in a variety of situations is becoming increasingly frequent. In spite of the inconclusiveness of current findings, research suggests a cautious, rather than optimistic, approach to the issue of cannabis and driving.