A DIFFICULTY IN USING THE BRITISH NATIONAL ROAD ACCIDENT DATA IN THE ANALYSIS OF INJURY SEVERITY IS THAT SOME SYSTEMATIC VARIATIONS EXIST IN THE GRADING OF INJURY SEVERITY. A STATISTICAL METHOD OF ALLOWING FOR THIS VARIABILITY IS BEING DEVELOPED FOR USE WHEN ANALYSING THE SEVERITY OF INJURY SUSTAINED IN DIFFERENT CIRCUMSTANCES AND THE EFFECT OF INJURY-REDUCING MEASURES ON SEVERITY, WITH THE AIM OF INCREASING THE SENSITIVITY OF SUCH ANALYSES. THE METHOD BEING DEVELOPED IS BASED ON THE EXPONENTIAL MODEL OF DISTRIBUTION OF SEVERITY PROPOSED BY HUTCHINSON. APPLICATIONS OF THE METHOD TO SEVERAL KINDS OF ACCIDENTS NOT INVOLVING PEDESTRIANS ARE PRESENTED: THOSE INVOLVING A SINGLE VEHICLE, TWO VEHICLES HEAD-ON, A MOTOR-CYCLE ONLY AND A MOTOR-CYCLE WITH ONE OR TWO OTHER VEHICLES. THE SEVERITY OF THE RESULTING INJURIES UNDER FOUR SETS OF CIRCUMSTANCES - DAY-URBAN, DAY-RURAL, NIGHT-URBAN AND NIGHT-RURAL - ARE INVESTIGATED. THE DATA FOR 1972 WERE USED AND FIVE POLICE FORCE AREAS WERE CHOSEN. THE RESULTS FROM THIS METHOD ARE THEN COMPARED WITH THOSE GIVEN BY ANALYSIS OF SOME CONVENTIONAL MEASURES OF INJURY SEVERITY.(Author/publisher).
Abstract