No restraint? Understanding differences in seat belt use between fatal crashes and observational surveys

Abstract

Observational surveys of restraint use in South Australia have reported vehicle occupant wearing rates somewhere in the order of 97%, however these rates drop below 70% for crashes where vehicle occupants are killed or seriously injured. In order to seek some understanding of why the prevalence of seat belt use varies between observational surveys and crash statistics a review of published international research and an analysis of a sample of fatal crashes in South Australia were undertaken. The literature review indicated that individuals less likely to wear seat belts were also most likely to be involved in crashes resulting in death or serious injury. A review South Australian Coroner’s data for fatal crashes in 2008 revealed that 37% of vehicle occupants killed in a crash were unrestrained. Further analysis indicated that those least likely to be restrained were younger, more likely to have tested positive to drugs and were more likely to have engaged in extreme behaviour than those who were restrained. Restraint use was also less common amongst fatalities in rollover crashes and amongst those ejected from the vehicle. Restraint use was also identified as an important issue for Indigenous Australians and people in regional and remote areas. Examination of the characteristics of fatal crashes revealed that observational survey methodologies have a limited capacity to detect those least likely to wear seat belts. Evidence of a selective recruitment effect was also observed. The findings are discussed in relation to potential countermeasures to increase restraint use.