Traffic Injuries in Ireland: a neglected problem

Abstract

Road safety policy has focussed heavily on the reduction of the number of fatalities. However injuries are also a serious consequence of traffic collisions and they have received much less research or policy attention. This paper examines the extent of the traffic injury problem, the costs for society and appropriate policies to reduce these costs. The available data from a variety of sources are examined and assessed. The analysis then concentrates on three main data sources in Ireland: police, hospitals and the Injuries Board, from each of which anonymised datasets were available. These three sources provide different perspectives on aspects of the injury problem. All three have specific data problems and suggestions are made for improvements in the data collection and analysis processes. One conclusion is that no single data source can adequately reflect the extent and variety of the traffic injury problem. The data are analysed first as separate datasets and then together using probabilistic record linkage techniques. The results indicate that less than 20% of police-reported injuries are matched with hospital patients and less than 30% of hospital patients are matched with a police-reported injury. The linkage shows how police-recorded injuries understate the total number of injuries in the two sources by about 50% . Almost 36,000 people were hospitalised for transport accidents over the nine-year period but were not recorded by the police. Of these, over 6,400 were clinically assessed as being seriously injured, underlining how official data understate the true extent of the injury problem. The linkage also confirms that the definition of a serious injury used by police is neither an accurate nor a consistent measure. Linkage involving the three data sets shows lower matching rates than between hospitals and police, and indicates a large number of injuries in addition to those identified by police and hospital data. The total number of individual injuries in the three sources combined is more than three times the official police-reported number. Present social cost estimates understate by at least €500m annually the true social costs of injuries. These costs now exceed the social costs of fatalities. The analysis also shows how cost-benefit analysis parameters for injuries should be revised to better account for injury underestimation. These findings have significant policy implications. These include firstly, the need for improved data to better understand injuries, secondly the need for greater emphasis on injuries in the national safety strategy, and thirdly the need for specific policies to deal with groups with high injury rates like cyclists and other vulnerable users.