Spatial heterogeneity in the impacts of Ohio’s enhanced graduated driver’s licensing law on teen motor vehicle crashes

Background: Graduated Driver’s Licensing (GDL) policies create an intermediate licensure phase for young novice drivers, and previous studies suggested that they reduce teen motor- vehicle crashes (MVCs). Multiple studies have shown that the effects of GDL laws vary in association with demographic factors and location, motivating estimation of sub-state policy effects. The present study estimates county-level effects of Ohio’s 2007 enhanced GDL law on MVCs among 16–17-year-olds. Methods: We analyzed the rates of non-rear-struck MVCs per licensed driver for 16–17-year-old drivers, who were targeted by the law, and 30–34-year-old drivers, who acted as a control group. Our primary analysis consisted of a spatio-temporal Bayesian disease mapping model that allowed parameters to vary by county while also capturing spatial dependence in the data and policy effects. We performed a series of post-hoc analyses to explore the policy effects associated with county-level factors like population, rurality, and annual vehicle miles traveled. Results: Both age groups experienced reductions in MVC rates after Ohio’s enhanced GDL law was enacted. However, the reductions were more extreme among 16–17-year-olds. The pre- to post-law MVC rate ratio was 18% more extreme (ratio of MVC rate ratios 0.82; 95% CI 0.81, 0.83) for 16–17-year-olds than it was for 30–34-year-olds. Furthermore, while we found some cursory evidence that the policy effect varied by county, Ohio’s enhanced GDL law was associated with reductions in MVC rates in all 88 counties. Conclusions: Our study suggests that Ohio’s enhanced GDL law was associated with reductions in MVC crash rate among 16–17-year-old drivers relative to 30–34-year-olds. Practical Applications: Our results support that Ohio’s enhanced GDL law was effective not only statewide, but also within each of Ohio’s 88 counties.