Longitudinal Study of Adverse Driving Outcomes Among Newly Licensed Adolescents with Autism

Background: The ability to drive independently enables opportunities for employment, education, and engagement in social activities; barriers to these during the period of transition to adulthood can be particularly limiting for autistic individuals. In a prior study we estimated that one-third of autistic teens successfully achieve licensure by age 21 (Curry et al., 2018). However, preliminary evidence also suggests that autistic adolescents may be at heightened risk for motor vehicle crashes.

Objectives: We conducted a retrospective cohort study to compare adverse driving outcomes—crashes, moving violations, and license suspensions—for newly licensed autistic and non-autistic adolescents.

Methods: This cohort included New Jersey residents who were born 1987-1997 and were patients of the Children’s Hospital of Philadelphia pediatric healthcare network. Electronic health records were linked with NJ’s statewide driver licensing and crash data. Autism status was classified via ICD-9-CM diagnosis codes; those with intellectual disability were excluded. The current study included a total of 53,320 licensed drivers, including 261 (0.5%) autistic and 53,059 (99.5%) non-autistic patients. Drivers were followed for up to 48 months post-licensure.

We calculated rates of overall and specific types of crashes—including injury, at-fault, nighttime, and involving peer passengers—as well as moving violations and suspensions. Further, we examined the proportion of crashes that were attributed to specific driver actions, such as unsafe speed, inattention, and failure to yield. Generalized estimating equation models were used to estimate adjusted rate ratios (aRR) and prevalence ratios (PR) comparing autistic and non-autistic drivers.

Results: In their first 48 months of licensure, 30.7% of autistic drivers and 30.5% of non-autistic drivers were involved in a police-reported crash. The average monthly overall crash rate over the study period was similar among autistic and non-autistic drivers (129.7 vs 109.2 per 10,000 drivers; aRR: 1.14 [0.92, 1.42]); rates of crash subgroups also did not differ among driver groups. Autistic drivers had lower moving violation rates (aRR: 0.68 [0.53, 0.86]) and half the rate of license suspensions (aRR: 0.45 [0.26, 0.78]).

The most common driver action for crashes among both autistic and non-autistic drivers was inattention (45.0% vs 39.4%; PR: 1.14 (0.93, 1.40]). Second most common among autistic drivers was failure to yield to the right-of-way (19.8% of all crashes), which was 1.6 times higher than among non-autistic drivers (8.6%; PR: 2.31; [1.56, 3.41]).

Conclusions: We found that newly licensed autistic drivers had similar crash rates and lower moving violation rates compared with non-autistic drivers; however, some characteristics of these crashes—including the specific crash-contributing driver actions—differed. In particular, crash-involved autistic drivers were more likely to have failed to yield to pedestrians and other vehicles; this finding is consistent with previous studies demonstrating that autistic teens more often divert their gaze away from the roadway compared with other teens. Results from this study may help inform driving interventions for newly licensed autistic adolescents to reduce risk of adverse driving outcomes and increase independence.