Department of Emergency Medicine, The University of British Columbia, Vancouver, BC, Canada.
Department of Emergency Medicine, The University of British Columbia, Vancouver, BC, Canada.
Lancet Public Health. 2021 Jun;6(6):e374-e385. doi: 10.1016/S2468-2667(21)00027-X. Epub 2021 Apr 20.
Many medications impair driving skills yet their influence on collision risk remains uncertain. We aimed to systematically investigate the risk of collision responsibility associated with common classes of prescription medications.
In this population-based case-control study we analysed linked driving and health records in British Columbia, Canada from Jan 1, 1997, to Dec 31, 2016. The study cohort included all drivers involved in an incident collision (defined as first collision after 3 collision-free years) that resulted in a police report. We scored police collision reports and classified drivers as responsible for the collision (cases) or not responsible (controls); drivers with indeterminate scores were excluded. We used logistic regression to determine odds of collision responsibility in drivers with current prescriptions for medications of interest versus drivers without prescriptions. To explore whether risk of collision responsibility was related to medication effect or driver factors, we compared risk in current medication users versus past users. To study whether drivers developed tolerance to medication effects, we compared risk in new (first 30 days of a prescription) versus established users.
During the study period, 4 906 925 drivers had their driving licence linked to health records; of these drivers, 747 662 unique drivers were involved in 837 919 incident collisions between Jan 1, 2000, and Dec 31, 2016. 382 685 drivers responsible for the collision (cases) and 332 259 drivers not responsible (controls) were included in the final analysis; 122 975 drivers with indeterminate responsibility were excluded. We found increased risk of collision responsibility in drivers prescribed sedating antipsychotics (adjusted odds ratio [aOR] 1·35 [98·75% CI 1·25-1·46]), long-acting benzodiazepines (aOR 1·30 [1·22-1·38]), short-acting benzodiazepines (aOR 1·25 [1·20-1·31]), and high-potency opioids (aOR 1·24 [1·17-1·30]). Among medications used for medical indications, the highest risk was seen in drivers prescribed neurological medications: cholinergic drugs (aOR 1·83 [1·39-2·40]), anticholinergic agents for Parkinson's disease (aOR 1·45 [1·08-1·96]), dopaminergic agents (aOR 1·20 [1·04-1·38]), and anticonvulsants (aOR 1·20 [1·14-1·26]). People currently taking benzodiazepines, non-sedating antidepressants, high-potency opioids, and anticonvulsants had increased risk compared with past users, and we did not find increased risk in new compared with established users of these drugs.
Drivers prescribed benzodiazepines or high-potency opioids are at increased risk of being responsible for collisions and this risk does not decrease over time. Several other classes of medications are associated with increased risk, but this association might be independent of medication effect. These findings can guide medication warnings and prescription choices and inform public education campaigns targeting impaired driving.
Canadian Institutes of Health Research.
许多药物会损害驾驶技能，但它们对碰撞风险的影响仍不确定。我们旨在系统地研究与常见处方药物类别相关的碰撞责任风险。
在这项基于人群的病例对照研究中，我们分析了加拿大不列颠哥伦比亚省从 1997 年 1 月 1 日至 2016 年 12 月 31 日的驾驶和健康记录的关联。研究队列包括所有因第一次碰撞前连续 3 年无碰撞而导致警方报告的事故碰撞（定义为首次碰撞）的驾驶员。我们对警方碰撞报告进行评分，并将驾驶员分类为对碰撞负责（病例）或不负责任（对照）；得分不确定的驾驶员被排除在外。我们使用逻辑回归来确定当前使用有研究意义的药物的驾驶员与无处方的驾驶员相比，发生碰撞责任的可能性。为了探究碰撞责任风险是否与药物作用或驾驶员因素有关，我们比较了当前药物使用者与过去使用者之间的风险。为了研究驾驶员是否对药物作用产生耐受性，我们比较了新处方（处方的前 30 天）使用者与已建立使用者之间的风险。
在研究期间，有 4906925 名驾驶员的驾驶执照与健康记录相关联；在这些驾驶员中，有 747662 名唯一驾驶员在 2000 年 1 月 1 日至 2016 年 12 月 31 日期间发生了 837919 起事故碰撞。382685 名负责碰撞的驾驶员（病例）和 332259 名无责任驾驶员（对照）被纳入最终分析；122975 名责任不确定的驾驶员被排除在外。我们发现，服用镇静抗精神病药物（调整后的优势比[aOR]1.35[98.75%CI 1.25-1.46]）、长效苯二氮䓬类药物（aOR 1.30[1.22-1.38]）、短效苯二氮䓬类药物（aOR 1.25[1.20-1.31]）和高效阿片类药物（aOR 1.24[1.17-1.30]）的驾驶员发生碰撞责任的风险增加。在用于医疗适应症的药物中，我们发现服用神经科药物的驾驶员风险最高：胆碱能药物（aOR 1.83[1.39-2.40]）、帕金森病的抗胆碱能药物（aOR 1.45[1.08-1.96]）、多巴胺能药物（aOR 1.20[1.04-1.38]）和抗惊厥药（aOR 1.20[1.14-1.26]）。与过去使用者相比，目前服用苯二氮䓬类药物、非镇静性抗抑郁药、高效阿片类药物和抗惊厥药的驾驶员发生碰撞的风险增加，我们没有发现这些药物的新使用者与已建立使用者之间的风险增加。
服用苯二氮䓬类药物或高效阿片类药物的驾驶员发生碰撞的风险增加，而且这种风险不会随着时间的推移而降低。还有其他几类药物与增加的风险相关，但这种关联可能与药物作用无关。这些发现可以指导药物警示和处方选择，并为针对驾驶障碍的公众教育活动提供信息。
加拿大卫生研究院。
