Confounding Factors of Commercial Motor Vehicles in Safety Critical Events

Abstract

Recent quasi-experimental commercial motor vehicle (CMV) driver hours-of-service (HOS) studies published through the Federal Motor Carrier Safety Administration (FMCSA) in 2011 readily identified consistent increases in crash odds as driving time increased. These studies identified time-on-task as a significant indicator of the potential for a safety critical event (SCE) (crash, near crash, or crash-relevant event). However, while these studies may have provided indication of a relationship between HOS and the probability of an SCE, they largely failed to account for many potential confounding factors. The HOS relationship is frequently attributable to the fatigue of the driver. Confounding factors however, are those factors that may also contribute to the likelihood of an incident and potentially create a systematic bias or contribute to measured error. This study sought to uncover existing relationships between the HOS observations and a set of potential confounding factors related to time of day. These relationships were addressed by controlling for confounds. Conclusions drawn from HOS-related studies, such as those mentioned herein affect millions of people and have economic impacts in the billions. Faulty scientific inferences from these studies can have high human and economic costs. Therefore, the work described in this study was needed to validate the findings of these studies, as well as other studies using similar designs and variables. In addition, the work described could lead to deeper insights into commercial motor vehicle crash risk and causation, with safety implications and applications beyond HOS regulations. We foresee the following to benefit from this work: federal regulators, trucking industry groups, carriers, academia, insurance companies; anyone interested in understanding crash causation.