Oak Ridge Institute for Science and Education, US Department of Transportation, Office of Policy Development, Strategic Planning and Performance, 1200 New Jersey Avenue, Washington, DC 20590, United States.
US Department of Transportation, Office of the Chief Information Officer, 1200 New Jersey Avenue, Washington, DC 20590, United States.
Accid Anal Prev. 2018 Dec;121:166-176. doi: 10.1016/j.aap.2018.06.018. Epub 2018 Sep 22.
Characteristics of the transportation system and built environment contribute to pedestrian fatality risks, including vehicular traffic and land-use characteristics associated with higher pedestrian activity. We combined data from FHWA, NHTSA, EPA, and the Census Bureau and performed regression modeling to explore associations between transportation system and built environment characteristics and pedestrian fatalities between 2012 and 2016 at the Census tract scale across the United States. In urban tracts, we found especially strong associations between traffic on non-access-controlled principal arterial and minor arterial roadways and pedestrian fatalities (0.91 and 0.68 additional annual pedestrian fatalities per 100,000 persons per 10,000 VMT/mi increase in traffic density, respectively). In both urban and rural tracts, we also found strong associations between employment density in the retail sector and pedestrian fatalities. Finally, we compared our model to the High Injury Network in Los Angeles, CA. Nearly half (43%) of observed fatalities were identified by both methods, while some fatalities were identified by only one (19% by our model and 23% by the High Injury Network). This work shows that traffic on certain roadway facility types and employment in certain sectors have especially strong associations with pedestrian fatality risk. More broadly, we illustrate how leveraging cross-disciplinary data in novel ways can support prospective, risk-based assessments of pedestrian fatality risks and support integrated and systemic approaches to transportation safety.
交通系统和建成环境的特点是导致行人死亡的风险因素，包括与行人活动量较大相关的机动车交通和土地利用特征。我们结合了 FHWA、NHTSA、EPA 和人口普查局的数据，并进行了回归建模，以探索 2012 年至 2016 年期间美国普查区范围内交通系统和建成环境特征与行人死亡之间的关系。在城市普查区，我们发现非控制主要干道和次要干道上的交通与行人死亡之间存在特别强的关联（交通密度每增加 10,000 英里/每 10,000 辆 VMT，交通密度分别增加 0.91 和 0.68 例行人每年额外死亡）。在城市和农村普查区，我们还发现零售部门就业密度与行人死亡之间存在很强的关联。最后，我们将我们的模型与加利福尼亚州洛杉矶的高伤害网络进行了比较。两种方法都识别出了近一半（43%）的观察到的死亡事件，而有些死亡事件仅被一种方法识别（我们的模型识别出 19%，高伤害网络识别出 23%）。这项工作表明，某些道路设施类型的交通和某些部门的就业与行人死亡风险有特别强的关联。更广泛地说，我们展示了如何以新颖的方式利用跨学科数据，以支持行人死亡风险的前瞻性、基于风险的评估，并支持交通安全性的综合和系统方法。
