Changes in the prevalence of behavioural factors including police-reported fatigue and alcohol intoxication, as well as self-reported seatbelt use, and their effect on hospitalization or death after a motor vehicle crash were assessed. Probabilistic linkage was used to match drivers in motor vehicle crashes with hospital discharge records for the years 1992-7. Frequencies of specific behavioural factors were evaluated using the Cochran-Armitage test for trend. Odds ratios and corresponding 95% confidence intervals were calculated using generalized estimating equations (GEEs) with crash and driver characteristics as independent variables and hospitalization or death as the dependent variable. The analysis database consisted of 450,286 crash driver records, which linked to 4219 (0.9%) hospitalizations or deaths. There was an increasing trend for self-reported seatbelt use among crash-involved drivers from 80.5% in 1992 to 89.3% in 1997 (P less than 0.001). Police-reported alcohol intoxication among crash-involved drivers showed a decreasing trend from 2.4% in 1992 to 1.5% in 1997 (P less than 0.001). There was no trend for police-reported fatigue-related crashes. Odds ratios of hospitalization or death for seatbelt use, alcohol involvement, and fatigue were significant and did not fluctuate considerably between 1992 and 1997. Seatbelt use offered a protective effect from hospitalization or death, while alcohol intoxication and fatigue contributed to increased likelihood of hospitalization or death. These results suggest that while some improvement has been made in decreasing seatbelt non-use and driver alcohol intoxication among crash-involved drivers, no improvement has been made in reducing fatigue-related crashes. (A) "Reprinted with permission from Elsevier".
Abstract