Medication use and the risk of motor vehicle collisions among licensed drivers: A systematic review

Abstract

Objectives: Driving under the influence of prescription and over-the-counter medication is a growing public health concern. A systematic review of the literature was performed to investigate which specific medications were associated with increased risk of motor vehicle collision (MVC).

Methods: The a priori inclusion criteria were: (1) studies published from English-language sources on or after January 1, 1960, (2) licensed drivers 15 years of age and older, (3) peer-reviewed publications, master's theses, doctoral dissertations, and conference papers, (4) studies limited to randomized control trials, cohort studies, case-control studies, or case-control type studies (5) outcome measure reported for at least one specific medication, (6) outcome measure reported as the odds or risk of a motor vehicle collision. Fourteen databases were examined along with hand-searching. Independent, dual selection of studies and data abstraction was performed.

Results: Fifty-three medications were investigated by 27 studies included in the review. Fifteen (28.3%) were associated with an increased risk of MVC. These included Buprenorphine, Codeine, Dihydrocodeine, Methadone, Tramadol, Levocitirizine, Diazepam, Flunitrazepam, Flurazepam, Lorazepam, Temazepam, Triazolam, Carisoprodol, Zolpidem, and Zopiclone.

Conclusions: Several medications were associated with an increased risk of MVC and decreased driving ability. The associations between specific medication use and the increased risk of MVC and/or affected driving ability are complex. Future research opportunities are plentiful and worthy of such investigation.

Keywords: Ability; Driving; Drugs; Prescription; Review.

Figure 1

Flow chart for the selection of studies.*, studies could have been excluded for multiple reasons

Figure 2

Odds or risk ratios with corresponding 95% confidence intervals from included studies that investigated the association between specific analgesic medications and MVC. Estimates by Foley et al were adjusted for age and gender; all other estimates presented are unadjusted.

Figure 3

Odds or risk ratios with corresponding 95% confidence intervals from included studies that investigated the association between specific anticonvulsant medications and MVC. The study by Etminan et al. was adjusted for age, sex, residence, previous MVC, chronic disease score, and exposure to antidepressants, antiepileptic, benzodiazepines, antipsychotics, antimigraine, muscle relaxants, and/or narcotic analgesics. All other estimates are unadjusted.

Figure 4

Odds or risk ratios with corresponding 95% confidence intervals from included studies that investigated the association between specific antidepressant medications and MVC. The study by Hansen provided two estimates. The first estimate (81a) corresponds to new users of Trazodone while the second (81b) corresponds to prevalent users of Trazodone.

Figure 5

Odds or risk ratios with corresponding 95% confidence intervals from included studies that investigated the association between specific antihistamine medications and MVC

Figure 6

Odds or risk ratios with corresponding 95% confidence intervals from included studies that investigated the association between specific antihyperglycemic medications and MVC. The study by Lonnen et al provided five estimates of risk (71a-e). These correspond to the risk of MVC for the following age groups respectively, <25, 25-44, 45-64, 65-74, and 75-84 years of age. The study by Mcgwin et al provided two estimates of risk (84a-b), which correspond with the odds of having any crash or a not-at-fault crash, respectively. The study by Mcgwin et al was adjusted for age, sex, and annual miles driven. The study by Sims et al was adjusted for sex, age, and days driven per week. The study by Hemmelgarn et al. was adjusted for age, gender, previous MVC, and place of residence. All other estimates are unadjusted.

Figure 7

Odds or risk ratios with corresponding 95% confidence intervals from included studies that investigated the association between specific benzodiazepine medications and MVC. Hansen provided two estimates of risk (81a-b) which correspond to new and prevalent users of Temazepam, respectively.

Figure 8

Odds or risk ratios with corresponding 95% confidence intervals from included studies that investigated the association between specific sleep-promoting medications and MVC. Hansen provided two estimates of risk (81a-b) which correspond to new and prevalent users of Zolpidem, respectively. The estimates provided by Yang et al. were adjusted for concomitant use of the following medications: Zolpidem, Zopiclone, long and short acting benzodiazepines, antihistamines, anticonvulsants, antidepressants, other sedatives/hypnotics, other psychoactive drugs, muscle relaxants, and opioid analgesics. All other estimates are unadjusted.

Figure 9

Odds or risk ratios with corresponding 95% confidence intervals from included studies that investigated the association between MVC and specific medications that could not be grouped into the other drug categories. Delaney et al provided two estimates of risk (87a-b) which correspond to both new and prevalent users, respectively, of Warfarin. The study by Sims et al. was adjusted for sex, age, and days driven per week. The study by Etminan et al. was adjusted for age, sex, residence, previous MVC, chronic disease score, and exposure to antidepressants, antiepileptic, benzodiazepines, antipsychotics, antimigraine, muscle relaxants, and/or narcotic analgesics. The study by Delaney et al. was adjusted for cardiac or stroke events within the past year and the following drug classes in the previous 60 days: antidepressants, anti-epileptics, benzodiazepines, antipsychotics, anti-migraine, narcotic analgesics, muscle relaxants. All other estimates are unadjusted.