Background: Several epidemiological and experimental studies have found a positive association between the risk of motor vehicle accidents (MVAs) and use of zopiclone and benzodiazepines. There is, however, little evidence of any risk of MVA attributable to the use of zolpidem 1 day before such accidents. We attempted to determine whether the use of zolpidem 1 day before is associated with an increased risk of an MVA.
Methods: Using a 1-million-person randomly sampled cohort from the Taiwan National Health Insurance reimbursement database, 12 929 subjects were identified as having been hospitalized between 1998 and 2004 due to an MVA. Using a case–crossover design, we selected the day before an MVA as the case period for each subject, and the 91st, 182nd, and 273rd days before the case period as 3 retrospective control periods. Conditional logistical regression models were constructed to calculate the odds ratio (OR) of having an MVA and the exposure of zolpidem 1 day before. We calculated doses of benzodiazepines, zopiclone, and zolpidem based on their defined daily dose.
Results: The adjusted OR for involvement in an MVA after taking 1 defined daily dose of zolpidem was 1.74 (95% confidence interval: 1.25–2.43). There were also positive effects for different washout periods and cumulative doses at 7, 14, 21, and 28 days before the occurrence of an MVA.
Conclusions: Use of zolpidem 1 day before might be associated with an increased risk of MVA. Thus, precautionary warnings should be provided when prescribing zolpidem.