Michael Keall1, Stuart Newstead. 1. Wellington School of Medicine and Health Sciences, Otago University, Wellington, New Zealand. Michael.keall@otago.ac.nz
Abstract
OBJECTIVES: The objective of this study was to find a comparison crash type that best represented exposure on the road and to identify situations where the induced exposure risk estimates were likely to be biased. METHODS: Counts of crash involvements were compared with distance driven estimates derived from a register of licensed motor vehicles to identify the most appropriate comparison crash type for induced exposure estimation, which is the crash type whose counts are best correlated with vehicle distance driven. RESULTS: The best sets of comparison crashes for disaggregations by driver age and gender and vehicle type were found to be multi-vehicle crashes in which the vehicle was damaged in the rear or multi-vehicle crashes in which the driver was adjudged to be not at fault. Likely bias of induced exposure risk estimates was identified, even for these best sets of comparison crashes, according to vehicle size (with large vehicles underrepresented) and owner age and gender (with young owners and female owners overrepresented). CONCLUSIONS: This research identified some important features of crash occurrence useful for making choices of comparison crash types when controlling for exposure. None of the crash types considered as comparison crashes performed perfectly. Even the crash types that seemed to best reflect exposure on the road still appeared to over- or underestimate distance driven according to owner age group, gender, and vehicle size.
OBJECTIVES: The objective of this study was to find a comparison crash type that best represented exposure on the road and to identify situations where the induced exposure risk estimates were likely to be biased. METHODS: Counts of crash involvements were compared with distance driven estimates derived from a register of licensed motor vehicles to identify the most appropriate comparison crash type for induced exposure estimation, which is the crash type whose counts are best correlated with vehicle distance driven. RESULTS: The best sets of comparison crashes for disaggregations by driver age and gender and vehicle type were found to be multi-vehicle crashes in which the vehicle was damaged in the rear or multi-vehicle crashes in which the driver was adjudged to be not at fault. Likely bias of induced exposure risk estimates was identified, even for these best sets of comparison crashes, according to vehicle size (with large vehicles underrepresented) and owner age and gender (with young owners and female owners overrepresented). CONCLUSIONS: This research identified some important features of crash occurrence useful for making choices of comparison crash types when controlling for exposure. None of the crash types considered as comparison crashes performed perfectly. Even the crash types that seemed to best reflect exposure on the road still appeared to over- or underestimate distance driven according to owner age group, gender, and vehicle size.