A deterministic algorithm was developed which allowed data from Department of Transportation motor vehicle crash records, state mortality registry records, and hospital admission and emergency department records to be linked for analysis of the financial and medical impact of motor vehicle crashes in Connecticut in 1995. There were 132,918 motor vehicle crash records (individual vehicles or pedestrians) involving 183,358 persons, 18.9% of whom were associated with one or more linked medical records, resulting in a total of 34,778 hospital visits. Of these, 91.7% were treated and released from the emergency department, with the rest being admitted to hospital with median length of stay of 3 days. About 77% of the persons treated in the emergency department or admitted as hospital inpatients had been identified as injured by the traffic safety officer at the scene of the crash. Median total hospital charges for these hospital visits was $405. Home discharge constituted 97% of the hospital discharges; mean total charges for those discharged to skilled nursing or intensive care facilities were 10 times greater than for those discharged home. There were 329 mortality registry records linked with these crashes, of which 46.8% occurred at the crash site. Logistic regression analysis determined that higher risk of fatal or serious injury was associated with head on collisions, motorcycle riding, driver illness, violating traffic control, and being at fault (according to the investigating traffic officer). Also associated with higher risk of fatal or serious injury were striking a tree, utility pole, or other object off the road, and multi vehicle collisions. Weather conditions associated with increased risk of fatal or serious injury were blowing sand, soil, or snow, or no adverse weather conditions. (A)
Abstract