Teenage driver involvement in injury motor vehicle crashes by strength of graduated driver licensing law

Introduction: Motor vehicle crashes (MVC) are a leading cause of morbidity and mortality for teenagers. Graduated driver licensing (GDL) programs are a common intervention designed to reduce this burden on newly licensed teens. Objective: Compare rates of teenage driver involvement in injury producing MVCs by the Insurance Institute for Highway Safety rating of GDL programs. Methods: 2005-2008 MVC data probabilistically linked with emergency department and hospital discharge data from the Crash Outcome Data Evaluation System (CODES) Data Network were used to identify 16-18 year-old driver MVCs. GDL programs were classified as Good or Fair/Marginal. Intercensal estimates were used to derive the number of 16-18 years-olds in each state and per capita MVC rates. An injury MVC was defined as one in which at least one occupant sustained a MAIS≥2 injury or died. Rate ratios were estimated using Poisson regression models with generalized estimating equations. Results: 519,094 teenage drivers were involved in MVCs: 22.0% were 16; 36.6% were 17; and 41.4% were 18. The rate of teenage driver involvement in injury MVCs for GDL programs rated as Good was 38% lower than GDL programs rated as Marginal/Fair. Good-rated GDL programs were also associated with lower rates of nighttime injury MVCs, lower rates of involvement by unbelted teenage drivers and teenage drivers suspected of alcohol/drug use. The largest risk reductions were associated with 16 year-old drivers. Risk reductions diminished with increasing age. Conclusion: Components of GDL incorporated in Good-rated programs were associated with risk reductions of teenage driver involvement in injury MVCs.