Institute for Mobility, Activity and Participation, College of Public Health and Health Professions, University of Florida, Gainesville, FL 32610, USA.
Epilepsy Behav. 2012 Feb;23(2):103-12. doi: 10.1016/j.yebeh.2011.11.015. Epub 2012 Jan 9.
The aim of this study was to synopsize the evidence on predictors of crashes and driving status in people with epilepsy (PWE).
Evidence-based review of the published English literature was the method used. We searched various databases and extracted data from 16 (of 77) primary studies. On the basis of American Academy of Neurology criteria, we assigned each study a class of evidence (I-IV, where I indicates the highest level of evidence) and made recommendations (Level A: predictive or not; Level B: probably predictive or not; Level C: possibly predictive or not; Level U: no recommendations).
For PWE, the following characteristics are considered useful: For identifying crash risk, epilepsy (level B) and short seizure-free intervals (≥3 months) (Level C) are not predictive of motor vehicle crash (MVC). For self/proxy-reported crash risk, epilepsy surgery (Level B), seizure-free intervals (6-12 months) (Level B), few prior non-seizure-related crashes (Level B), and regular antiepileptic drug adjustments (Level B) are protective against crashes; seizures contribute to MVCs (Level C); mandatory reporting does not contribute to reduced crashes (Level C). No recommendations for reliable auras, age, and gender (Level U), as data are inadequate to make determinations. For self-reported driving or licensure status, employment and epilepsy surgery are predictive of driving (Level C); there are no recommendations for antiepileptic drug use, self-reported driving, gender, age, receiving employment benefits, or having reduced seizure frequency (Level U).
Limitations, that is, heterogeneity among studies, examining the English literature from 1994 to 2010, must be considered. Yet, this is the first evidence-based review to synopsize the current PWE and driving literature and to provide recommendation(s) to clinicians and policy makers. Class I studies, matched for age and gender, yielding Level A recommendations are urgently needed to define the risks, benefits, and causal factors underlying driving performance issues in PWE.
本研究旨在综合有关癫痫患者（PWE）发生车祸和驾驶状况的预测因素的证据。
采用已发表的英文文献的循证综述方法。我们搜索了各种数据库，并从 16 项（77 项中的 16 项）原始研究中提取数据。根据美国神经病学学会的标准，我们为每项研究分配了一个证据等级（I-IV，其中 I 表示最高水平的证据）并提出了建议（A级：有预测性或无预测性；B 级：可能有预测性或无预测性；C 级：可能有预测性或无预测性；U 级：无建议）。
对于 PWE，以下特征被认为具有预测价值：对于识别车祸风险，癫痫（B 级）和短无发作间隔（≥3 个月）（C 级）不能预测机动车事故（MVC）。对于自我/代理报告的车祸风险，癫痫手术（B 级）、无发作间隔（6-12 个月）（B 级）、较少的非癫痫相关车祸（B 级）和定期调整抗癫痫药物（B 级）可预防车祸；癫痫发作可导致 MVC（C 级）；强制性报告并不能降低车祸发生率（C 级）。对于可靠的先兆、年龄和性别，没有建议（U 级），因为数据不足以做出决定。对于自我报告的驾驶或驾驶执照状况，就业和癫痫手术可预测驾驶（C 级）；对于抗癫痫药物使用、自我报告的驾驶、性别、年龄、获得就业福利或减少发作频率，没有建议（U 级）。
必须考虑到研究之间的局限性，即异质性，这是对 1994 年至 2010 年的英文文献进行的首次基于证据的综述，旨在综合目前有关 PWE 和驾驶文献的证据，并为临床医生和决策者提供建议。迫切需要匹配年龄和性别进行 I 级研究，以确定 PWE 驾驶表现问题的风险、收益和因果因素。
