Individual Differences and the "High-Risk" Commercial Driver (2004)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

5CHAPTER 1 INTRODUCTION 1.1 BACKGROUND: ILLUSTRATIVE EXAMPLE In an instrumented vehicle study of local/short-haul (LSH) truck driving sponsored by the Federal Motor Carrier Safety Administration (FMCSA), Hanowski et al. (2000) observed 42 truck drivers driving a total of 28,000 vehicle miles. The study identified 249 critical incidents (CIs), which were defined as significant unsafe driver actions or “near-crashes.” Of these 249 CIs, 77 were related primarily to the actions and errors of truck drivers. Common critical incidents included running late yellow or red lights and crossing traffic with insuf- ficient gaps (i.e., approaching vehicles too close for safe cross- ing). The 42 truck drivers initiated 77 CIs in 1,376 hours of driving, yielding an average rate of 0.06 truck driver–initiated CIs per hour. Figure 1 shows the frequency distribution of CI/hour rates among the 42 drivers. Of the 42 truck drivers, 6 drivers had CI/hour rates greater than 0.15. These 6 drivers drove 12% of the total driving hours of the study but were responsible for 38% of all the truck driver–initiated CIs (29 of 77). In contrast, the 25 “best” of the 42 drivers (the first two bars in Figure 1) drove 63% of the driving hours but were responsible for 16% of the CIs. Figure 2 illustrates these exposure-risk relationships for the “worst” and “best” LSH drivers in terms of CI initiation. The study also assessed driver alertness level, using a 5-point Observer Rating of Drowsiness (ORD) scale, which had previously been validated against physiological alert- ness measures. Levels 4 and 5 corresponded to “very” and “extremely” drowsy. The equipment malfunctioned for one driver, so there were 41 drivers in this sample. The 41 drivers had a total of 285 time episodes of Level 4 or 5 on the ORD scale over 1,348 hours of driving, for an average rate of 0.21 high-drowsiness episodes per hour. Figure 3 shows the fre- quency distribution of high-drowsiness episodes for the 41 drivers. Four drivers had rates of more than 0.75 high-drowsiness episodes per hour. These four drivers drove 7% of the total driving hours but were responsible for 39% of all observed high-drowsiness episodes (112 of 285). There was also a moderate-risk group (10 drivers) who had 29% of the ex- posure and 47% of the drowsy episodes. In contrast, the 27 most alert drivers (the first two bars in Figure 3) drove 64% of the driving hours but were responsible for only 14% of high-drowsiness episodes. Figure 4 illustrates the exposure- risk relationships for the high-, moderate-, and low-risk subgroups of LSH drivers. The risk-exposure odds ratios between the worst and best groups of drivers identified here were 12.5 for CIs and 25.5 for high-drowsiness episodes. In other words, on average, each high-risk driver in Figure 2 was 12.5 times more likely to be involved in CIs than the low-risk drivers. In Figure 4, the high-risk drivers were 25.5 times more likely to have drowsy episodes than were the low-risk drivers. There was only a small positive relationship between the rate of CIs and the rate of drowsiness among the 41 drivers for which both types of data were available. This suggests that drowsiness was a factor in CI involvement, but it was not a predominant factor. Only one of the six high-CI drivers was among the four high-drowsiness drivers. 1.2 SCOPE These LSH study statistics were presented to introduce and demonstrate the phenomenon of high-risk commercial drivers. Although commercial drivers generally drive responsibly and exhibit lower rates of most types of incident and crash involve- ment than drivers in general (FMCSA 2003, Craft 2004, Wang, Knipling, and Blincoe, 1999), it appears that there are significant safety-related individual differences among groups of drivers, and that a few commercial drivers have signifi- cantly elevated risk compared with their peers. This synthesis will explore individual differences among commercial drivers in general and high-risk commercial driv- ers in particular. It will identify dimensions and factors relat- ing to differences in commercial driver crash risk and assess ways that the high-risk driver can be targeted by various safety programs and practices, at both fleet- and industry-wide levels. Specifically, the synthesis will • Summarize available information on individual differ- ences in commercial driver safety performance and alertness. • Examine various metrics and tests that might be used to hire better drivers and avoid hiring high-risk drivers. • Identify safety management techniques that are currently used by commercial vehicle carriers to target problem drivers and their specific risky behaviors.

60 2 4 6 8 10 12 14 16 .0 .01-.05 .06-.10 .11-.15 .16+ Critical Incident/Hour Groupings N um be r o f D riv er s (N =4 2) Figure 1. Frequency distribution of LSH truck driver critical incident rate. Exposure: Hours of Driving 63% 25% 12% Risk: Critical Incidents 16% 46% 38% High Risk Moderate Risk Low Risk Figure 2. Relationship between exposure and CI risk for high-risk, moderate-risk, and low-risk groups of drivers in the Hanowski et al. (2000) LSH truck driver study. 0 2 4 6 8 10 12 14 16 18 .0 .01-.25 .26-.50 .51-.75 .76+ High-Drowsiness Episodes/Hour Groupings N um be r o f D riv er s (N =4 1) Figure 3. Frequency distribution of LSH truck driver high-drowsiness episodes.

Although the synthesis focuses primarily on heavy truck drivers, it will also address long-haul motor coach drivers and, to a limited extent, the phenomenon of high-risk transportation operators in general. Most safety interventions described in the synthesis are carrier management related, and carrier safety managers are the major information source and intended audience for the synthesis. However, a few of the safety approaches discussed are related to national regulatory and enforcement issues. Appendix B reproduces the research project’s Statement of Work. The reasons and causes behind differential driver risk will be explored in this synthesis. In the LSH study, for example, each driver was observed for only 1 week. It is possible that the differences observed were related to long-term constitutional factors (e.g., enduring individual differences in aggression, risk-taking, health), short-term personal factors (e.g., tempo- rary illness, lack of sleep due to a new baby at home), or sit- uational factors (e.g., weather, traffic variations, new delivery routes). The study collected no longitudinal data on drivers, and few truck safety studies seem to have done so. 1.3 APPROACH Information on high-risk commercial drivers and poten- tial management solutions was obtained through several major approaches. The primary vehicle for obtaining information was surveys. Two parallel survey forms were employed: one for current CMV fleet safety managers and one for other experts in motor carrier safety. The safety manager and other expert survey forms were parallel in their questions and con- tent, but there was a key difference in the way the management effectiveness-related questions were asked. Safety managers were asked if they currently used the method and then, if “yes,” were asked to rate its effectiveness. Thus, these questions yield 7 data on the prevalence of industry use of the methods as well as subjective evaluations of them. The other expert survey provided data only on expert opinion because the questions regarding use of the methods were not applicable. Findings from these surveys will be the centerpiece of this synthesis, although the authors concede that opinion surveys are not a substitute for empirical data. To supplement the expert survey, a focus group was con- ducted with staff members of the FMCSA, the major fed- eral agency overseeing truck and bus safety. Supporting the survey and focus group findings is information obtained from literature reviews relating to the various personal cor- relates of driver risk and to safety management methods. The literature reviews employed Transportation Research Information System (TRIS) and other reference systems to identify relevant publications in the transportation liter- ature. Also reviewed were FMCSA research publications and research journals on traffic safety, psychometrics, and industrial safety management. The remainder of this synthesis will present the survey methodology and results, define and characterize the basic phenomenon of high-risk commercial drivers, explore fac- tors related to driver risk, and discuss operational safety management methods for addressing the problem. The syn- thesis concludes with recommendations for R&D that might be performed to address the many unanswered questions relating to this issue, deepen current knowledge, and create tools that industry can apply to ameliorate the problem. R&D is conceived broadly and may include many different types of initiatives undertaken by various parties and stakeholders involved in motor carrier safety. If it is true that approxi- mately 10% of commercial drivers are associated with one- third or more of safety risk caused by commercial drivers, there is clearly an opportunity to significantly reduce crash loss through focused efforts on these drivers. Exposure: Hours of Driving 64% 29% 7% Risk: High-Drowsiness Episodes 14% 47% 39% High Risk Moderate Risk Low Risk Figure 4. Relationship between exposure and drowsiness risk for high-risk, moderate-risk, and low-risk groups of drivers in the Hanowski et al. (2000) LSH truck driver study.