Driving, Alcohol, and Drugs

DRIVING, ALCOHOL, AND DRUGS

Injuries, especially from motor vehicle collisions, are the leading cause of death for individuals under age 44. The presence of alcohol is the factor most frequently associated with fatalities in vehicles, drownings, falls, and fire (U.S. Department of Health and Human Services, 1987). In the first report to Congress on traffic safety and alcohol (U.S. Department of Transportation, 1968), it was concluded that more than 50 percent of fatal traffic collisions and 33 percent of serious injury traffic collisions were alcohol-related.

Although the association between alcohol consumption and traffic accidents had been recognized by the beginning of the twentieth century, the magnitude of the problem did not capture public attention until the 1970s. Public tolerance of Driving Under the Influence of alcohol decreased sharply—a shift in attitude that, combined with increased legal countermeasures, resulted in a significant decline in alcohol-related fatalities from a high of 57 percent in 1982 to 38 percent in 1998.

Voas et al. (1998) compared the relative frequency of driving under the influence of alcohol in three U.S. nationwide surveys, done in 1973, 1986, and 1996 on weekend nights. Drivers were stopped at random and asked to provide breath samples for alcohol testing. The blood alcohol concentration (BAC) levels were compared for the three surveys as a function of time, day, gender, age, ethnicity, geographical region, etc. Across nearly all population subgroups, the presence of alcohol in nighttime weekend drivers dropped from 36 percent in 1973 to 26 percent in 1986 to 17 percent in 1996. However, although the percent of decline for drivers with BACs below 0.10 percent was 54 percent from 1973 to 1996, there was only a 45 percent decline in drivers with over 0.10 percent BAC. Despite this significant drop in the number of alcohol impaired drivers in the last two and a half decades, alcohol still remains the single largest factor in traffic fatalities and serious injuries.

Epidemiological studies have compared the BAC levels in collision-involved drivers with those of randomly selected drivers passing the collision site at similar times. These studies have demonstrated that the probability of a crash increases with any departure from zero BAC and increases exponentially with increasing BAC levels. By the time BAC levels exceed 0.2 grams alcohol per deciliter of blood (200 mg/100 mL.), the probability of a collision increases more than 100 times (i.e., 10.000%).

Most areas of human behavior are impaired eventually by increasing alcohol levels. However, the examination of alcohol-related collision data from governmental investigations and police collision reports suggests that information-processing errors are common in the majority of alcohol-related traffic collisions. Information-processing deficits include impairment of attention, visual search, and perception. The second largest category of human factor errors involves judgment, such as speed selection. Failure to control a car because of decreasing motor skills remains a distant third crash category, despite the popular assumption that links driving impairment with the appearance of intoxication and motor incapacitation.

The results observed in epidemiological survey studies are supported by numerous experimental studies in which driver behavior was examined under controlled conditions. Such laboratory studies either examine one or two behaviors relevant to driving at a time or in more complex studies of driving-related behavior use driving simulators and a closed-course driving situation that preserve the safety of the driver.

Moskowitz and Robinson (1988) reviewed 177 experimental studies of alcohol's effects on driving-related behaviors that met criteria of scientific merit. The behavior found to be most affected by alcohol was divided attention, with impairment even seen at alcohol levels below 0.02 percent (20mg/100mL). Divided-attention tasks involve simultaneously monitoring and responding to more than one source of information, which is characteristic of many complex man-machine interactions such as driving and flying. While operating a vehicle, drivers under the influence of alcohol frequently fail to detect significant potential threats in the environment.

Similarly, studies have indicated that information processing and perception are affected at low BAC levels. Tracking, which is analogous to car control functions such as maintaining heading and lane position, has been shown to be impaired at low BACs when performed simultaneously with other functions in divided-attention situations, but less impaired when the tracking task is performed by itself. Similarly, complex reaction-time tasks involving several competing stimuli and responses are impaired at low BACs, whereas simple reaction time requiring little information processing was more resistant to the effects of alcohol.

Studies of psychomotor skills performance in driving simulators and closed-course driving studies have shown considerable variation in the BAC levels at which impairment appears. These variations are likely explained by the differences in information-processing requirements among these varied tasks. The review concluded that no minimum threshold for alcohol's impairment of complex human-machine tasks exists. Any reliable measure of alcohol in the human system produces some impairment.

Other areas that have been suggested as leading to alcohol-related accidents, such as poor judgment and violent and aggressive behavior, have been infrequently examined by researchers—mainly because of the difficulty of developing laboratory techniques to measure them.

The low BAC levels at which laboratory studies have indicated significant impairment and epidemiological studies have shown increased crash frequency are below the levels at which the majority of the population would exhibit symptoms of intoxication such as slurred speech and unsteady gait. Thus, the absence of signs of intoxication is not evidence that a driver is capable of operating a motor vehicle or other machinery safely.

Moskowitz and Fiorentino (2000) updated Moskowitz and Robinson's 1988 report with a review of an additional 112 studies published from 1981 to 1997. Although the main conclusions of the 1988 report remained confirmed, the most recent publications more frequently report impairments at very low alcohol levels, reflecting improvements in the sensitivity and reliability of scientific investigation. Moreover, new behavioral areas are being explored, such as the tendency to fall asleep at the wheel, which increases significantly at low BAC levels.

OTHER DRUGS

The major involvement of alcohol in traffic accidents and other injuries is well documented. What conclusions can we draw about the role of drugs other than alcohol in traffic safety? Although laboratory studies on the effects of many drugs and alcohol are similar in demonstrating the impairment of performance skills, there are difficulties in executing epidemiological studies on the effects of drugs in driving. For example, few non-crash-involved drivers volunteer to provide blood samples so their drug levels might be compared with those in blood samples obtained from collision victims.

Although several studies have been completed in hospitals with drug levels in trauma patients involved in driving collisions with blood samples from volunteers who were in the hospital for other reasons, serious questions arise regarding the representativeness of the control group.

Another problem in relating drug use to vehicle crashes has been difficulty of evaluating the behavioral significance of drug blood levels. Unlike alcohol, where levels in venous blood samples or breath samples are essentially equivalent to those from blood in the brain, the site of drug action, for nearly all other drugs have a complex relationship between blood plasma level and the degree of resulting behavioral impairment. Many drugs remain present in the plasma for weeks beyond any period in which behavioral effects may be observed. In other cases, drug levels in plasma drop extremely rapidly and become difficult to detect while behavioral impairment remains. Thus, most epidemiological studies of drugs and driving report the presence of the drug rather than the level of concentration.

One technique to circumvent control-group problems has been to assign responsibility or non-responsibility to crash-involved drivers and then correlate the presence of drugs with the frequency of crash responsibility. Within the constraints of these epidemiological studies, researchers have often concluded that tranquilizers, antihistamines, and antidepressants are overrepresented in crash-involved drivers.

Terhune and colleagues (1992) examined the presence of drugs in blood specimens from 1,882 fatally injured drivers. Drugs, both illicit and prescription, were found in 18 percent of the fatalities. Marijuana was found in 6.7 percent, Cocaine in 5.3 percent, tranquilizers in 2.9 percent, and Amphetamines in 1.9 percent of these fatally injured drivers.

When crash responsibility was assigned and correlated with drug use, the small number of individuals in each separate drug classification made statistical significance difficult to obtain despite the fact that several drug categories were associated with increased crash responsibility. Crash-responsibility rates did increase significantly as the number of drugs in the driver increased. Many drug users used several drugs simultaneously and these drivers had the highest collision rates. Alcohol was found in 52 percent of the fatalities, with more than 90 percent of the drivers with BACs over 0.08 percent considered responsible for the crash.

The most frequently used illicit drug in the United States of America for the last half-century is marijuana and epidemiological studies have demonstrated that is the most frequent drug consumed by drivers. Bates and Blakely (1999) have reviewed the epidemiological literature for marijuana's role in motor vehicle crashes. They concluded that there is no evidence marijuana alone increased either fatal or serious injury crashes. However, the evidence is inconclusive whether the presence of marijuana in combination with alcohol increases fatalities or serious injuries over that produced by alcohol alone. Nor was it possible to determine whether marijuana increases the rate of less serious vehicle crashes.

In contrast to the lack of scientific information available from epidemiological sources about the role of drugs in causing collisions, numerous experimental research has been performed to evaluate the effects of drugs on skills performance. Regulatory agencies in many countries have frequently required an evaluation of the side effects of prescription drugs on skills performance. Also, numerous governments have supported studies of the effects of illicit and abused drugs on skills performance in the laboratory.

Thus, the evaluation of the effects of drugs on driving and other human-machine interactions has depended primarily on experimental studies where changes in behavior can be observed as a function of differences in administered doses and the time after administration. However, no other drug has been evaluated in as extensive a range of behaviors as has alcohol. Nevertheless, many drugs have been studied with respect to some important variables required for driving.

The emphasis in these drug studies has tended to be on the evaluation of vision, attention, vigilance, and psychomotor skills. Driving-simulator studies have also been done on occasion. The psychomotor skill most often examined has been some form of tracking.

Reviewing this literature presents considerable difficulties since there are so many differences between classes of drugs, as well as between individual drugs within the same drug classification. For example, many minor tranquilizers, especially Benzodiazepines, have been shown to impair attention and tracking in a wide variety of studies. However, recently introduced tranquilizers, such as buspirone, exhibit little evidence of impairment.

Conclusions about impairments in a drug category are likely to change because of the pressures exerted by the drug regulatory agencies on drug companies to develop medicines that do not impair skills performance. For example, hypnotics often exhibit residual skills impairment the day following use. New drugs have been introduced whose duration of effects is shorter so there will be less residual impairment after awakening.

Another class of psychoactive drugs, the Antidepressants, especially amitriptyline, have been long known to impair performance in a variety of skills. Again, recently introduced types of antidepressants do not produce the same degree of impairment.

Although narcotic Analgesics derived from Opium (Opiates) have been shown experimentally to lead to decreased alertness, there have been reports that chronic use produces considerable tolerance to some of these side effects, which may explain why epidemiological studies have not found differences in crash rates between Narcotic users and control groups. Moreover, patients maintained on a stabilized dosage of Methadone, a synthesized narcotic, have shown little evidence of impairment in a wide variety of experimental and epidemiological studies.

Another category of drug that shows evidence of impairing skills performance in laboratory studies is the antihistamines, many of which produce impairment of performance accompanied by complaints of drowsiness and lack of alertness. Again, recent pharmacological advancements have produced antihistamine drugs, like loratadine (Claritin), which maintain antihistamine actions but have difficulty crossing the blood-brain barrier and thus produce little impairment.

Of all illicit drugs, marijuana has had the largest number of experimental studies performed to examine its effects. Many of these studies indicate that marijuana impairs coordination, tracking, perception, and vigilance, as well as performance in driving simulators and on-the-road studies.

Although there has been concern over increased driver use of Stimulants, such as amphatamines and cocaine, there is little experimental evidence demonstrating driving impairment by these drugs. On the contrary, most studies of these stimulants, as well as of Caffeine, indicate an improvement in skills performance. However, with the chronic (long-term) use of stimulants, an increased dose must be taken as tolerance develops. Thus, the dose levels examined in the laboratory may not reflect those found among drivers. Second, after the stimulation phase, a subsequent depressed phase occurs (the "crash") with increased drowsiness and lack of alertness. The stimulant drugs have not been well studied in relation to driving and should be. Further study is needed—both for acute (one-time) use and chronic use.

(See also: Dramshop Liability Laws: Drunk Driving ; Minimum Drinking Age Laws ; Mothers Against Drunk Driving ; Psychomotor Effects of Alcohol and Drugs ; Social Class of Alcohol and Drug Abuse ; Students Against Destructive Decisions ).

BIBLIOGRAPHY

Bates, M.N.& Blakely, T. A. (1999). Role of cannabis in motor vehicle crashes. Epidemiological Reviews, 21 (2), 222-232.

Moskowitz, H., & Fiorentino, D. (2000). A review of the literature on the effects of low doses of alcohol on driving-related skills. Report no. DOT HS 809 028.U.S. Department of Transportation, Washington, DC.

Moskowitz, H., & Robinson, C. D. (1988). Effects of low doses of alcohol on driving-related skills: A review of the evidence. Report No. DOT HS 807 280. U.S. Department of Transportation, Washington, DC.

Terhune, K. W., et al. (1992). The incidence and role of drugs in fatally injured drivers. Report no. DOT 808 065. U.S. Department of Transportation, Washington DC.

U.S. Department of Health and Human Services. (1987). Sixth special report to the U.S. Congress on alcohol and health. Rockville, MD: Author.

U.S. Department of Transportation. (1968). Alcohol and highway safety: A report to the Congress. Washington, DC: U.S. Government Printing Office.

Voas, R. B., et al. (1998). Drinking and driving in the United States: The 1996 national roadside survey. Accident, Analysis & Prevention, 30 (2), 267-275.

Herbert Moskowitz