The effects of graduated driver licensing on hospitalization rates and charges for 16-and 17-year-olds in North Carolina

OBJECTIVE: To mitigate the high risk of motor vehicle crashes for young beginning drivers, over 40 states and the District of Columbia have implemented graduated driver licensing (GDL) systems that gradually and systematically ease teen drivers into higher risk driving conditions. Evaluations of GDL programs using motor vehicle crash data have demonstrated marked declines in crashes. The objective of this study is to examine the association between the implementation of the North Carolina GDL program and the rate of hospitalization, as well as hospital charges, for 16-and 17-year-old drivers. METHODS: Data were obtained from the North Carolina Hospital Discharge Database for the 26 months before and 46 months after the December 1, 1997, implementation of the GDL program. ARIMA interrupted time series analyses were used to model monthly hospitalization rates, controlling for the hospitalization rates of 25-to 54-year-old drivers. ARIMA analyses were also used to determine whether changes occurred in monthly total hospital charges. RESULTS: Among the 568 16-year-old hospitalized drivers, GDL was associated with a 36.5% decline in the hospitalization rate per population and a 31.2% decline in the total monthly driver hospitalization charges. Although a 12% reduction in the rate of hospitalizations was observed among the 615 17-year-old drivers, the analysis lacked sufficient power to be statistically reliable. No consistent change was observed in the 16-year-old driver total monthly hospital charges. CONCLUSIONS: The North Carolina GDL program was associated with a marked decline in the rate of hospitalizations and hospital charges for 16-year-old drivers. Following the implementation of GDL, over $650,000 in hospital charges have been averted each year for 16-year-old drivers. Analyses suggest these reductions were primarily the result of reduced exposure rather than an improvement in teen driving.