|
Abstract
|
In contrast to most other countries with Graduated Driver Licensing (GDL) systems, only a few U.S. jurisdictions currently extend GDL restrictions to drivers licensed at age 18 or older. However, several recent studies have found that a substantial proportion of drivers are getting licensed at older ages, and further, that later licensure occurs disproportionately among low-income and minority drivers. A current topic of discussion in the traffic safety community is whether standard GDL policies that are applied in the U.S. for younger novice drivers should be applied to older novices not presently covered by them. One step in evaluating this issue is to determine whether there is epidemiologic evidence to support—from a crash reduction perspective—adoption of extended-age GDL policies, and in particular passenger and nighttime restrictions. However, the current body of literature that directly addresses whether GDL restrictions might reduce the burden of crashes if applied to older novices is very limited. This study examines how overall, injury, nighttime, and passenger crashes over the initial four years of licensure differ for novice drivers licensed at different ages.
Using the New Jersey Traffic Safety Outcomes (NJ-TSO) Program database, the research team constructed an analytic database of New Jersey drivers who obtained their initial intermediate driver's license from January 2006 through December 2014 and had at least one month of follow-up from the date of licensure to study end or death (n=1,034,835). Novice drivers were grouped based on age at licensure: (1) age 17 (subject to full GDL restrictions); (2) age 18–20 (subject to full GDL restrictions in New Jersey but not in the vast majority of other U.S. states); (3) age 21–24 (not subject to nighttime or passenger restrictions in U.S. states); and (4) age 25 or older (not subject to nighttime or passenger restrictions in U.S. states). Monthly rates were estimated for overall crashes (per 10,000 licensed drivers) as well as: injury crashes, late night crashes (11:01 p.m.–4:59 a.m.), early night crashes (9 p.m.–11 p.m.), and multiple-passenger crashes (two or more passengers present). Average monthly rates were calculated for specific relevant time periods since licensure (0–3 months, 4–6 months, 7–12 months, 13–18 months, 19–24 months, 25–36 months, and 37–48 months) and were compared via estimation of rate ratios and 95% confidence intervals using Poisson regression models. Rate ratios, adjusted for sex, were estimated to compare rates: (1) between novice driver groups with the same time since licensure; (2) over the first 48 months of licensure within each novice driver group; and (3) between same-aged 21-year-old drivers with varying lengths of licensure.
Main findings for overall, nighttime, and multiple-passenger crashes were as follows:
Although the initial three-month crash rates of novice New Jersey drivers age 21 and older were higher than rates of same-aged experienced drivers, they were substantially lower than initial three-month rates for 17- to 20-year-old novice drivers, who were subject to full GDL policies upon licensure in New Jersey. This was particularly true for novice drivers aged 25 and older, whose average crash rates over the first three months of licensure were more than 50% lower than for those licensed at age 17. Moreover, novice drivers aged 21 and older experienced much less steep crash reductions over the first year of licensure than younger novice drivers.
Nighttime crash rates among the groups licensed at ages 21 to 24 and at ages 25 and older were stable over the first year of licensure. For novice drivers licensed before age 21, early night crash rates during the 9 p.m.–11 p.m. period declined rapidly over the course of licensure—significantly more rapidly than overall crashes did—while changes in rates of late night crashes were much smaller. First-year multiple-passenger crash rates were highest for drivers licensed at ages 18 to 20. Novice driver groups experienced varying amounts of reduction in multiple-passenger crashes over time. Among drivers licensed at age 17, the rate of multiple-passenger crashes was 35% lower at 7 – 12 months post-licensure compared with the first three months post - licensure. Drivers licensed at ages 21 and older experienced similar reductions to drivers licensed at ages 18-20, with 17% to 23% lower rates over the same time period. In summary, findings support New Jersey's current GDL policies for 17- to 20-year-old novice drivers and the potential for added benefits from beginning the restrictions at 9 p.m. There is a lack of compelling evidence for extending GDL policies currently applied to younger novices to drivers licensed at ages 21 – 24. The lower initial crash rates for these drivers compared with younger novice drivers and the lack of steep reductions in their crashes over time call into question the potential for GDL provisions to significantly reduce the crash rates of these drivers. There is no evidence to indicate a need for additional GDL policies for New Jersey novices aged 25 years and older. |