Article Text

Download PDFPDF
116 Variation in drivers’ seat belt use by community-level social vulnerability
  1. Emma Sartin,
  2. Leah Lombardi,
  3. Kristina Metzger,
  4. Allison Curry
  1. Children’s Hospital of Philadelphia, Philadelphia, USA


Statement of Purpose Despite seat belt legislation, many drivers are not restrained every trip. Health behaviors with documented disparities—including restraint use—are associated with community-level factors. Identifying at-risk communities ensures supportive initiatives are allocated to communities who need them most. We utilized novel methods to compare prevalence of drivers’ restraint use by community-level measures of social vulnerability.

Methods/Approach Our New Jersey Safety and Health Outcomes warehouse contains crash and geocoded residential address data for NJ drivers (2010–2017). We borrowed the fundamental assumption of quasi-induced exposure methodology: that non-responsible drivers in multi-vehicle ‘clean’ crashes (only one responsible driver) reasonably represent drivers on the road (N=1.1M non-responsible drivers; Curry, 2017). Using the Census Bureau’s Community Resilience Estimates (CRE), we categorized Census tracts into quintiles based on the proportion of the population with 3+ (of 11) indicators of social vulnerability (e.g., no Internet access). We compared trip-level prevalence of restraint use across CRE quintiles.

Results Overall, NJ drivers were unbelted during 1.7% of trips. However, drivers living in Census tracts with the largest proportion of the population with 3+ indicators of social vulnerability were twice as likely to be unbelted as those in Census tracts with the smallest proportion of the population with 3+ indicators (2.5% vs. 1.2% of trips unbelted; PR: 2.02 [1.93, 2.12]).

Conclusion Although drivers were restrained for most trips, based on average travel patterns we estimate that NJ drivers were unbelted on up to 416,000 trips in 2017. Our results indicate that a disproportionate number of unbelted trips were taken by drivers living in communities with more socially vulnerable households, placing them at substantially higher risk for sustaining crash injuries. These novel methods can be utilized to identify at-risk communities that should be prioritized for prevention efforts and resources aimed at reducing the burden of motor vehicle crashes.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.