NewsBriefs BUTTONS

Study of Reckless Drivers' Drug Use


September-October 1994

A study in the August 25 issue of the New England Journal of Medicine finds that more than half of all reckless drivers who were not apparently driving under the influence of alcohol tested positive for cocaine and/or marijuana use (Daniel Brookoff, Charles S. Cook, Charles Williams, and Calvin Mann, "Testing Reckless Drivers for Cocaine and Marijuana Use," New England Journal of Medicine, vol. 331, no. 8 (Aug. 25, 1994): p. 518-522; "Roadside Study of Reckless Drivers Shows Marijuana and Cocaine Use Among Persons Not Impaired by Alcohol," CESAR FAX, Sept. 12, 1994). NewsBriefs believes this flawed study is likely to be misrepresented.

An unreported number of drivers were stopped for reckless driving in Memphis, TN in the summer of 1993. An unreported number of those stopped who were "primarily suspected of driving under the influence of alcohol" were processed but not tested for drug use nor included in this "study" of reckless driving. 175 drivers, the subjects of this study, were judged not under the influence of alcohol and were subjected to a "standard field sobriety test" to evaluate the driver's degree of intoxication (e.g., walking a straight line, touching finger to nose, standing on one leg). The study does not say whether the group screened out for alcohol intoxication was given the same test. Intoxication: 68% (119) of the study's sample were classified as moderately or extremely intoxicated. 32% (56) were not or mildly intoxicated. Most (150) complied with a request to submit to a urine test for marijuana and cocaine use. 14% of those not or mildly intoxicated tested positive for marijuana only; 16% of all those who tested positive for marijuana were identified as not or mildly intoxicated. 59% of those 150 tested positive for marijuana, cocaine, or both (33% for marijuana, 13% for cocaine, 12% for both). For those who were considered by police to be moderately or extremely intoxicated after a sobriety test (94 of the 150), 85% tested positive for cocaine or marijuana. The study concludes that half of the drivers stopped for reckless driving and not under the influence of alcohol were under the influence of marijuana, cocaine, or both.

[This study has many fundamental methodological flaws. First, the basic data collection technique is flawed by eliminating an unknown number of suspected alcohol intoxicated subjects. Second, 28% of the 175 in the non-alcohol group were tested for alcohol and were positive for recent alcohol use at levels up to 0.21 mg/dl, a level far above legal intoxication. The authors point out that other studies have shown that up to one-third of subjects with blood alcohol levels in the intoxicated range have no appreciable odor of alcohol. Thus some of the intoxicated "drug" cases were certainly intoxicated by alcohol and probably not by use of cocaine or marijuana. The authors fail to correlate any of the alcohol intoxication data they had with either the drug use data or the degree of intoxication. Third, the authors fail to quantify or define "under the influence" and imply that it means testing positive for the drug metabolite in urine at any level of concentration. This indicates nothing about the level of impairment the person is experiencing. Nor does it tell us what drug is responsible for the impairment. In fact, the study states that some of those that were at first classified as not under the influence of alcohol but intoxicated later were found to have blood alcohol levels establishing legal intoxication and that some who tested positive for marijuana did not appear intoxicated. Fourth, the drivers could also have been under the influence of any number of prescription or over-the-counter drugs, for which the researchers did not test. While they tested positive for marijuana or cocaine or both, many other drugs (legal or illegal) could have been responsible for their impairment. This study falls prey to the fallacy that correlation equals causation. Not knowing the total number stopped for reckless driving makes it impossible to support the authors' conclusion that the use of marijuana and cocaine is a major cause of reckless driving. -- Editors]