Skip navigation

Benzodiazepines and crashes

The proportion of over-65s in the population will increase. After adjusting for the number of miles driven, crash rates in the elderly population are twice as high as for middle-aged drivers, and only under-25s have higher crash rates. Over-65s often take benzodiazepines for insomnia or anxiety. Benzodiazepines affect the central nervous system in a number of ways, and the elderly may be less able to metabolise the drugs, so one consequence of all this could be reduced ability to drive safely. Two studies have now shown an association between benzodiazepine use and increased risk of motor vehicle crashes.

Elderly drivers in Quebec

The first [1] was conducted in Quebec and examined 225,000 people aged 67 to 84 years who were drivers of cars in the years 1990 to 1993. There were 5579 drivers involved in a crash in which someone was injured, and a random sample of controls, from which 55790 control person days were selected. Benzodiazepine exposure was ascertained from records of prescriptions filled, a method which had previously been validated. Cases and controls were similar.

Short half-life benzodiazepines were not associated with any increased risk of motor vehicle crashes, with an overall rate ratio of 0.96 (95% confidence interval 0.88 to 1.05). There was no increased risk in the first week of exposure, nor at any longer period (Figure 1).

Long half-life benzodiazepines, in contrast, were associated with a 28% increased risk of a crash - rate ratio 1.28 (1.12 to 1.45). The increased risk was associated with all durations of benzodiazepine use (Figure 1), but this was highest and statistically significant for the first week (45% increased risk) and use longer than two months.

In this report long half-life (> 24 hours) benzodiazepines included clonazepam, diazepam, clorazepate, chlordiazepoxide, flurazepam and nitrazepam. Those with a short half-life (≤24 hours) were alprazolam, bromazepam, lorazepam, oxazepam, temazepam and triazolam.

Benzodiazepines in Scotland

A study in Tayside looked at prescribed drugs in drivers aged 18 years and over who had a first road traffic accident attended by Tayside police between mid-1992 and mid-1995. There were 19,386 cases. Using a record linking database, cases acted as their own controls. The odds of having a crash while exposed were compared with those while not exposed.

The results showed that of the drug classes looked at (benzodiazepines, tricyclic antidepressants, SSRIs and other psychoactive drugs), only benzodiazepines were associated with increased risk of a crash. There were 235 individuals exposed to benzodiazepines who had a crash, with an odds ratio of 1.62 (95% confidence interval 1.24 to 2.12). There was a clear dose response (Figure 2), with high-dose benzodiazepines increasing risk of a crash by 170%.


These two thorough studies underline that benzodiazepines can be associated with higher risks of road accidents. Higher doses and long acting benzodiazepines will be the principal culprits. Benzodiazepine use has apparently been associated with increased fall and hip fracture rates in the elderly, especially with longer acting compounds. Reducing benzodiazepine prescribing in general practice may be one strategy, and in Bandolier 4 we highlighted a simple way to reduce the number of prescriptions in general practices with established policies to minimise benzodiazepine prescribing. By using a letter and information sheets a fifth of patients stopped taking benzodiazepines altogether and 40% reduced their prescriptions by half.


  1. B Hemmelgarn, S Suissa, A Huang, JF Boivin, G Pinard. Benzodiazepine use and the risk of motor vehicle crash in the elderly. JAMA 1997 278: 27-31.
  2. F Barbone, AD McMahon, PG Davey et al. Association of road-traffic accidents with benzodiazepine use. Lancet 1998 352: 1331-6.

previous or next story in this issue