Marijuana and Driving


Alcohol continues to be, and will most likely always be, the principal drug which causes deterioration of driving performance. In North Carolina for over 10 years, the bloods of more than half of the operators (57%) killed in single vehicle crashes contained more than 0.09% alcohol. The effects of drugs other than alcohol on driving are virtually unknown. Only recently have attempts been made to see if other drugs have adverse effects that could lead to accidents. The contribution of other drugs with and without alcohol should be ascertained.

There is concern that the use of marijuana may increase the risk of having an accident. In order to establish this scientifically, it will be necessary to show that operators are being adversely affected by the drug and that they are over-represented in the group of drivers having accidents compared to a non-accident group.

To obtain such information it is necessary to analyze bloods from a random sample of the non-accident driving population and also from operators involved in crashes. It is becoming increasingly difficult to obtain blood from living operators. This is not true of operators killed in crashes. If a drug is rarely found or occurs at a non-influencing concentration in the blood of those killed in crashes, the drug is a relatively insignificant factor in crashes and it is not necessary to establish the incidence of the drug in the general driving population.

Information which may be of questionable value in evaluating the effects of drugs on driving is based on anecdotal or hearsay statements, laboratory tests of physical or mental impairment, laboratory driver simulators , test course performance, and actual street-driver performance. A report published in 1977 reviews the literature up to that time. Little has been added since that time. Regardless of the results of the above methods of ascertaining impairment they do not answer the question of whether the use of a certain drug has an adverse effect on driving. This question may best be answered by determining the concentrations of drugs in the bloods of a significant number of operators involved in crashes which are not attributable to some other cause. Evidence of crashes which might be attributable to operators adversely affected by marijuana is very rare. Peripheral studies have indicated that marijuana may cause problems but not-for a very significant number of operators.

An early and widely cited report on drivers allegedly under the influence of marijuana, who might have caused fatal accidents, was based on hearsay. This is the report of Sterling-Smith et al. (). They concluded that 43 out of 267 (16%) of the drivers were under the influence of marijuana (). The results were based on interviews of friends and relatives of the deceased. Only 13 (5%) were said to have used only marijuana. The “risk taking behavior” of marijuana smokers was rated as low and in the same category with “driving without restraints” or “smoking more than 2 packages of cigarettes daily.” Since there was no evidence that there were any marijuana constituents in the blood of any operators it is doubtful that this report sheds any light on a possible safety problem that might be due to the effects of smoking marijuana.

Teale () reported on the examination of 66 bloods obtained from fatally injured drivers (). The bloods were analyzed by radio-immunoassay for cross-reacting cannabinoids, which were found in 6 specimens. Three of the blood extracts were purified by high pressure liquid chromatography to separate the tetrahydrocannabinol (THC) from the other cannabinoids. Blood of one of the victims contained 2.3 ug/L of THC and 0.34% of alcohol. The other two victims were motorcyclists who crashed into automobiles. Their bloods contained low concentrations of THC, 1.5 and 4.4 ug/L. Because cannabinoids persist in the blood for many hours, it is not possible to assess the effects these low concentrations of marijuana had on the operators. This is rather a small sample to draw any valid conclusions.

Reeve in 1979 reported on the examination of bloods from 1792 people arrested for driving under the influence (). Blood was analyzed for THC by a radioimmunoassay. Unfortunately, the procedure used was not adequately enough documented to establish its sensitivity (5 ug/L). Of the 1792 operators, 281 (16%) had positive findings of THC in their blood, 111 of the 281 had blood alcohol concentrations exceeding 0.09%. Most specimens were not analyzed for other drugs. A strong negative correlation was found between the THC positive bloods and accidents. The value of this study as an indication of the safety of driving under the influence of marijuana is questionable.

A Canadian study () sought drugs in 401 fatally injured drivers and THC was found in the bloods of 14 (3.5%) of the victims, but 7 of the 14 had more than 0.09% alcohol in their blood. The bloods of 2 of the drivers contained 5 ug/L of THC; the rest had less than 3 ug/L of THC ().

The literature does not reveal that marijuana is a factor in unsafe operation of a motor vehicle. Sparse as the reports may be, they tend to show that if there are drivers who are unsafe because of marijuana, their numbers are small and most are also influenced by alcohol.


The 100 cases chosen for this study were those where adequate and suitable specimens of blood were obtained from dead operators who were killed in single vehicle crashes. Specimens from those who lived more than 1 hour or whose blood was unsuitable because of decomposition or contamination by embalming fluid were excluded.

The blood specimens were analyzed by RIA for THC using an Iodine-125 tracer (). Only 0.1 ml of plasma or blood was required. The sensitivity was at least 0.3 ug/L for plasma and 1.1 ug/L for hemolyzed blood. The antiserum cross reacted with 11-hydroxy THC about 20%, but, fortunately, this physiologically active metabolite is present in blood in low concentrations () following marijuana smoking. It may be a significant factor when marijuana is ingested orally. The other major metabolite, 11-carboxy THC, which is physiologically inactive, is present in blood in significant quantities but fortunately cross reacted less than 0.1%. The analysis was further tested by analyzing specimens obtained from controlled smoking studies (Fig. 1). No specimen which contained less than 3 ug/L of THC was reported as positive.

Results and discussion

The safety of driving following the smoking of marijuana will have to be established. In our opinion, this has not been done. Smoking studies reveal that the concentration of THC peaks in the blood at about 50 to 150 ug/L, depending on the dose, in minutes while still smoking. The concentration drops to about half that in about 10 minutes, and to less than 10 ug/L in about an hour. The reported high lasts about two hours (). No reports were found which indicated that during the time the person was “high” their driving performance was affected.

The concentration of THC found in the victims in this study ranged from 3 to 18 ug/L with a median of 5 ug/L. At this time we are unable to assess the effects of such concentrations on driving ability ().

We do not know the significance of 18 ug/L of THC combined with 0.11% of ethanol in one of the victims. Certainly that concentration of alcohol alone has been blamed for adverse driving performance. Practically nothing is known of the effects of combinations of alcohol and marijuana or any other drugs on driving. Intuition leads us to believe that the other drugs won’t favorably improve the ability of the driver influenced by alcohol.

THC was found in the bloods of 9 of the 100 dead operators tested. In 6 of the 9 bloods, alcohol in sufficient concentration to influence operation was also found. One blood which had a low concentration of alcohol also had a low concentration of butalbital. No opiate, amphetamine, or phencyclidine (P.C.P.) was found in any victim. Thus marijuana might have added to the effects of alcohol in 6 of the 100 cases (6%). Marijuana alone might have been a factor in 3 of 100 (3%). The incidence of alcohol in concentration greater than 0.09% was 62% (62 of 100). The number of drivers who had marijuana in their blood was very small compared to the number of those who had alcohol in their blood. Marijuana was not found in the bloods of 91% of the drivers tested.

It will be difficult to establish whether or not marijuana is a significant factor in traffic safety. The effects of marijuana smoking are subtle. It produces a pleasant state of relaxation, euphoria (a sense of well-being), altered perception of distance and time, impaired memory of recent events, and impaired physical coordination. This state lasts about an hour or two. Obviously some people will overindulge with marijuana as they do with alcohol and other drugs and their performance will suffer greatly. Are there many such people, will they try to drive, and will they have accidents? If marijuana has an adverse effect on operation, how much does it affect safety, for how long, and can the effect be correlated with THC blood concentrations? Is the marijuana-influenced operator more or less aggressive, apt to speed, likely to take fewer chances, able to compensate, etc.? Is driving adversely affected by marijuana in the same way that it is by alcohol and/or other drugs?

The presence of cannabinoids on the breath or in the saliva may be an indication that someone has smoked marijuana in the past few hours and that the oral cavity was directly exposed. Unless concentrations in the breath or saliva can be correlated with effect, these specimens will only be useful to show that marijuana has been smoked at some time recently. Cannabinoids persist in the urine for several days, thus positive urine specimens only identify marijuana users. It does not appear to be possible to develop an instrument for the detection and quantitation of marijuana that is as easy to use as alcohol breath testing equipment. Are such devices really necessary before it has been established that there is a significant risk associated with marijuana and driving?

It is difficult because of legal and logistic problems to obtain blood specimens from the general driving population so that the number of non-accident drivers who smoke marijuana and the concentration of cannabinoids in their blood can be determined. If marijuana is found in a small number of drivers killed in crashes, it would mean that it is not a significant problem regardless of what is found in the driving population. If a greater percentage of driving population had marijuana in their blood than the percentage in the crash group, would that mean that it is safer to drive after smoking marijuana? Probably the most meaningful answer concerning the effect of marijuana on driving will come from analyzing the bloods of operators killed in single vehicle crashes.

A recent Department of Transportation report to Congress emphasizes the need for information to “support arguments either for or against establishing marijuana as a high priority highway safety concern”().

More research should be undertaken before legislation and counter-measures are proposed to counteract a presumed problem of driving under the influence of marijuana. Until the time that marijuana is shown to be a significant problem, erratic operators may still be removed from the highway by arresting them for reckless driving.

A. J. McBay and S. M. Owens

Selections from the book: “Problems of Drug Dependence, 1980: Proceedings of the 42nd Annual Scientific Meeting, the Committee on Problems of Drug Dependence, Inc.” Louis S. Harris, Ph.D., ed. Comprehensive assemblage of ongoing research on drug abuse, addiction, and new compounds. National Institute on Drug Abuse Research Monograph 34, February 1981.