Identifying Novel Targets for Child Passenger Safety Intervention Using Geospatial Analysis and Data Science - PROJECT SUMMARY/ABSTRACT Unintentional injuries continue to be the leading cause of mortality for children and youth. Motor vehicle collision (MVC)-related injuries lead to the majority of these deaths. When children experience non-fatal injuries in MVCs, they can suffer life-long chronic disease from traumatic brain injury, spinal cord injury, fractures, and internal organ damage. While appropriate use of child passenger restraint systems (CRS) is a proven life-saving measure, benefits are compromised by selection of a less protective CRS, installation errors and misuse. Child passenger safety is reduced further when drivers allow children to sit in the front seat or travel unrestrained altogether. CRS deployment and education (pre-event)—targeted to specific family, community, and societal crash factors (event) as well as available triage and treatment after an MVC (post- event)—fill an important gap where caregiver education and state laws have not resulted in adherence to CRS best practices. Despite what is known about the benefits of CRS usage, limited information exists on where to deploy scarce resources in the US to increase adherence to guidelines and laws. Additionally, few studies have evaluated suboptimal child passenger safety behaviors and crash outcomes within the context of family (size, income, driver, vehicle), community (child passenger safety technician [CPST] availability, seat check utilization), and societal (Child Opportunity Index [COI], pediatric injury risk, and traffic laws) factors—as well as overall contributors and countermeasures. I aim to systematically identify factors impacting MVC injury outcomes from pre- to post-injury and inform interventions that strategically deploy effective countermeasures to areas of greatest need. First, I will use national car seat check datasets and geospatial techniques in ArcGIS Pro to identify CPST deserts and hotspots for selection of less protective CRS, installation errors, and misuse. Next, I will use national crash data to identify suboptimal CRS practice hotspots and employ a logistic regression model to detect salient family, community, and societal risk factors. Finally, I will conduct a road network distance analysis using the ArcGIS Pro Network Analysis Toolbox to identify populations at highest likelihood of poor injury outcomes due to being over 60 minutes from a children’s hospital or trauma center. I will triangulate these analyses and findings together into a virtual geodatabase from which real-time maps can be created and overlaid to identify county-level needs and subsequent intervention packages. Anticipated downstream interventions informed by this work may include deployment of CPSTs, CRS, and specific education; state-level refinement to traffic laws; and rural telehealth and triage support. Through the proposed research, I will acquire skills in data science and advanced statistical and spatiotemporal analyses to systematically identify factors, contributors, and countermeasures impacting MVC-related injury outcomes. Training in geographic information systems analysis will help launch my career in precision spatiotemporal injury intervention targeting as an independent pediatric physician-scientist.
