K. Bressler1, H. Roberts1, D. Schulz1, J. Feliciano2, D. Karev2, S. DiRusso1,2 1New York College Of Osteopathic Medicine, Old Westbury, NY, USA 2St Barnabas Hospital, Bronx, NY, USA
Introduction: The COVID-19 pandemic response of stay-at-home orders resulted in reduced vehicle traffic and changes in pedestrian activity. Despite these changes, the number pedestrian struck (PS) traumas has remained persistent. Geotemporal mapping has long been used to track where these incidents occur and have helped shape injury prevention plans. A NYC inner city level II trauma center was used to analyze demographics, injury severity and Geotemporal patterns of PS traumas throughout the pandemic.
Methods: This level II trauma registry included 3,001 patients of which 317 were pedestrian struck injuries and 247 had corresponding location data. Pandemic timeframe cohorts were PreCOVID (2019, N=57), COVID (2020, N=67), 1yrPostCOVID (2021, N=79), and 2yrsPostCOVID (2022, N=44). Patient demographics and injury severity were analyzed using SPSS Statistics v27 with p<0.05 indicating significance. Kernel density estimation (KDE) was done using ArcGIS to determine Geotemporal locations with highest PS density.
Results: Pedestrian struck trauma admissions and injury severity measures did not significantly change throughout the pandemic. Demographic analysis showed that PS patients ³ 65 years old significantly decreased from PreCOVID to COVID and 1yrPostCOVID [18(31.5%) vs 10(14.9%) and 11(13.9%)]. Black Non-Hispanic PS patients significantly increased from PreCOVID to 1yrPostCOVID [21(36.8%) vs 43(54.4%)]. White Non-Hispanics significantly decreased from PreCOVID to 1yrPostCOVID [16(28.1%) vs 11(13.9%)]. Kernel density analysis revealed smaller more confined clusters of PS incidents during COVID compared to all other time periods. 1yrPostCOVID shows just one distinct cluster compared to three distinct clusters in PreCOVID and COVID. This cluster then shifted location and became more spread out during 2yrsPostCOVID (Figure 1).
Conclusion: Although the pandemic did not significantly change the number of PS trauma admissions, it was associated with a change in the demographics of those admitted. Locations of PS incidents became more condensed during COVID and then expanded back out in the years after but never returned to PreCOVID clusters. This use of kernel density estimation to analyze incident location density demonstrates this methodology is a useful tool to help shape injury prevention and road traffic changes for the future.