Elizabeth A. Walshe is a research scientist at the Center for Injury Research and Prevention at the Children's Hospital of Philadelphia, in Philadelphia, Pennsylvania.
Daniel Romer is the research director in the Annenberg Public Policy Center, University of Pennsylvania, in Philadelphia, Pennsylvania.
Health Aff (Millwood). 2020 Oct;39(10):1792-1798. doi: 10.1377/hlthaff.2020.00802.
Motor vehicle crashes remain the leading cause of adolescent mortality and injury in the United States. For young drivers, crash risk peaks immediately after licensure and declines during the next two years, making the point of licensure an important safety intervention opportunity. Legislation in Ohio established a unique health-transportation partnership among the State of Ohio, Children's Hospital of Philadelphia, and Diagnostic Driving, Inc., to identify underprepared driver license applicants through a virtual driving assessment system. The system, a computer-based virtual driving test, exposes drivers to common serious crash scenarios to identify critical skill deficits and is delivered in testing centers immediately before the on-road examination. A pilot study of license applicants who completed it showed that the virtual driving assessment system accurately predicted which drivers would fail the on-road examination and provided automated feedback that informed drivers on their skill deficits. At this time, the partnership's work is informing policy changes around integrating the virtual driving assessment system into licensing and driver training with the aim of reducing crashes in the first months of independent driving. The system can be developed to identify deficits in safety-critical skills that lead to crashes in new drivers and to address challenges that the coronavirus disease 2019 pandemic has introduced to driver testing and training.
机动车事故仍然是美国导致青少年死亡和受伤的主要原因。对于年轻司机来说，事故风险在获得驾照后立即达到峰值，并在接下来的两年内下降，因此获得驾照是一个重要的安全干预机会。俄亥俄州的立法在俄亥俄州、费城儿童医院和 Diagnostic Driving, Inc. 之间建立了一个独特的健康交通伙伴关系，通过虚拟驾驶评估系统识别准备不足的驾驶员申请人。该系统是一个基于计算机的虚拟驾驶测试，通过模拟常见的严重撞车场景，发现关键技能缺陷，并在路考前在测试中心进行。对完成该系统的驾照申请人的试点研究表明，虚拟驾驶评估系统能够准确预测哪些驾驶员会在路考中失败，并提供自动反馈，告知驾驶员他们的技能缺陷。目前，该伙伴关系的工作正在推动将虚拟驾驶评估系统纳入许可和驾驶员培训的政策改革，以减少新手司机在独立驾驶的头几个月发生的事故。该系统可以开发出来，以识别导致新司机撞车的安全关键技能缺陷，并解决新冠疫情对驾驶员测试和培训带来的挑战。
