Citation

Wang X, Liu J, Khattak AJ, Clarke D. Safety Sci. 2016; 82: 44-55.

Copyright

(Copyright © 2016, Elsevier Publishing)

DOI

10.1016/j.ssci.2015.08.017

PMID

unavailable

Abstract

Transportation professionals have long recognized the harm of trespassing along railway rights-of-way. However, the non-crossing rail trespassing issue has received less attention compared with highway-rail grade crossing crashes, despite the fact that nearly 8800 rail-trespassing crashes occurred on non-crossing rail tracks during the past decade with a large number of them resulting in fatality. Also, geographic and socio-demographic diversity within the US implies that trespassing crash severity and its correlates may vary across geographic entities or regions. The purpose of this paper is to investigate these issues using rail-trespasser crash data maintained by Federal Railroad Administration (N = 8794 over 2004-2013). The unique aspects of the study are the development of a framework that explores correlates of injury severity in crashes and applies appropriate analysis methods. Specifically, using rigorous spatial modeling methods (Geographically Weighted Regression), the study uncovers spatial variations in correlates between rail-trespassing injury and revealed contributing factors. The factors include personal attributes of individuals, environmental and location attributes, time of crash and pre-crash behaviors. The results show that non-crossing trespass crashes are generally severe with 52.1% involving a fatality. Pre-crash behaviors were found to be key factors showing significant associations with the probability of rail-trespassing injury, especially lying or sleeping (on or near tracks). Fundamentally, the basic assumption of spatial stationarity in traditional regression models does not fully hold in the situation explored. Correlates of injury severity are found to be non-stationary across space. Therefore, regional considerations in specific situations should guide the implementation of treatments and policies.