Current issues of ACP Journal Club are published in Annals of Internal Medicine


Anxiolytic benzodiazepines and zopiclone increased road traffic crashes

ACP J Club. 1999 Mar-April;130:48. doi:10.7326/ACPJC-1999-130-2-048

Source Citation

Barbone F, McMahon AD, Davey PG, et al. Association of road-traffic accidents with benzodiazepine use. Lancet. 1998 Oct 24;352:1331-6.



Is psychoactive drug use associated with an increased risk for road traffic crashes?


A case-crossover study.


Tayside, Scotland, United Kingdom.


1731 persons (55% men) who were ≥ 18 years of age, had a first road traffic crash between 1 August 1992 and 30 June 1995 (identified by Tayside police records), and had received a psychoactive drug during the study period.

Assessment of risk factors

Use of psychoactive drugs, including tricyclic antidepressants, selective serotonin-reuptake inhibitors [SSRIs], and benzodiazepines (zopiclone was classified as a benzodiazepine), and participant and crash data.

Main outcome measure

A first road traffic crash.

Main results

Results are shown in the Table. On the days of the first road traffic crashes, 235 participants were receiving benzodiazepines, 189 were receiving tricyclic antidepressants, 84 were receiving SSRIs, and 47 were receiving other psychoactive drugs. The risk for a first crash increased with use of benzodiazepines but not with use of tricyclic antidepressants, SSRIs, or other psychoactive drugs. Users of benzodiazepines had an increased risk for a crash if they were ≤ 44 years of age; they were also more likely to be at fault. The risk for a crash was greater if the alcohol breath test was positive. No difference in risk was shown between men and women. Analysis of benzodiazepine subgroups showed that the risk for crashes was greater with benzodiazepines used as anxiolytics, zopiclone, and long-acting benzodiazepines.


Benzodiazepines as a group (including zopiclone) increased the risk for a first road traffic crash, whereas tricyclic antidepressants, selective serotonin-reuptake inhibitors, and other psychoactive drugs did not. Benzodiazepines used as anxiolytics, zopiclone, and long-acting benzodiazepines were associated with an even greater risk for crashes.

Source of funding: SmithKline Beecham.

For correspondence: Dr. T.M. MacDonald, Medicines Monitoring Unit, Department of Clinical Pharmacology and Therapeutics, University of Dundee, Ninewells Hospital and Medical School, Dundee DD1 9SY, Scotland, UK. FAX 44-1382-667120.

Table. Risks for a first road traffic crash while receiving benzodiazepines

Drug Participants Odds ratio (95% CI)
Benzodiazepine All 1.62 (1.24 to 2.12)
< 30 y of age 2.66 (1.35 to 5.25)
30 to 44 y of age 2.18 (1.30 to 3.64)
Driver at fault 1.88 (1.36 to 2.60)
Positive breath test 8.15 (2.06 to 32.34)
Anxiolytics All 2.18 (1.52 to 3.13)
Zopiclone All 4.00 (1.31 to 12.2)
Long-acting benzodiazepines All 2.03 (1.41 to 2.93)


Barbone and colleagues observed a significant increase in risk for motor vehicle crashes with use of benzodiazepines and zopiclone in this case-crossover study. A dose-response relation was seen. Analysis according to drug elimination half-life showed that anxiolytics with long half-lives were specifically associated with an increased occurrence of road traffic crashes.

Surprisingly, risk associated with benzodiazepine use was highest among drivers < 30 years of age and decreased with advancing age. Among persons ≥ 65 years of age, no increase in risk was found. These results conflict with those of 2 epidemiologic studies that focused on the geriatric population (1, 2). In view of the widespread concerns about the harmful effects of long half-life benzodiazepines on the elderly, the results reported by Barbone and colleagues seem incongruous. There are several possible explanations for these findings, including insufficient statistical power to detect an increased risk in the elderly, reduced driving by elderly persons while receiving benzodiazepines, and the common use of long-term benzodiazepine therapy among elderly persons. The last possibility is particularly important because the case-crossover study design is appropriate only for transient exposures. As the authors point out, a person taking a drug throughout the study period would have the same exposure at the time of the crash as in the previous control period, which underestimates the risks in persons who receive long-term therapy.

Despite the somewhat confusing findings, this study reinforces public health concerns about the risk for motor vehicle crashes among persons who use benzodiazepines as anxiolytics and those who receive zopiclone. All such persons should be specifically cautioned about this risk, and prescribers should consider advising such patients not to drive.

Jerry H. Gurwitz, MD
Fallon Healthcare System and University of Massachusetts Medical SchoolWorcester, Massachusetts, USA


1. Hemmelgarn B, Suissa S, Huang A, Boivin JF, Pinard G. Benzodiazepine use and the risk of motor vehicle crash in the elderly. JAMA. 1997;278:27-31.

2. Ray WA, Fought RL, Decker MD. Psychoactive drugs and the risk of injurious motor vehicle crashes in elderly drivers. Am J Epidemiol. 1992;136:873-83.