354 - Association of Nonemergent Pediatric Emergency Department Visit Rate and Childhood Opportunity Index Score by Census Tract

Poster Number: 354
Publication Number: 354.313

Mark R. Hincapie, UPMC Children's Hospital of Pittsburgh, Pittsburgh, PA, United States; Montserrat A. Corbera-Hincapie, UPMC Children's Hospital of Pittsburgh, Pittsburgh, PA, United States; Srinivasan Suresh, UPMC Children's Hospital of Pittsburgh/University of Pittsburgh, Wexford, PA, United States; Gabriella A. Butler, UPMC Childrens Hospital of Pittsburgh, Canonsburg, PA, United States; Kristin Ray, University of Pittsburgh, Pittsburgh, PA, United States

Background: Over half of emergency department (ED) visits are deemed nonemergent. Prior approaches to reduce nonemergent ED visits have often focused on individual factors and interventions. Emerging data are identifying important associations between ED visits and structural and community factors, warranting further examination. The Childhood Opportunity Index 2.0 (COI) offers a metric for assessing multiple community factors, including census tract early childhood education, housing, and other neighborhood resources.

Objective: To investigate the association of census tract community resources and census tract nonemergent pediatric ED (PED) rates

Design/Methods: We identified nonemergent PED visits during a three-year period (01/01/18-12/31/20) by using PED electronic health record data from a tertiary children’s hospital. Encounters with patient address outside of the hospital’s county were excluded. The remaining encounters were successfully geocoded using ArcGIS Pro 2.7 (ESRI, Inc) and assigned to census tracts; 7% of encounters were unable to be geocoded and were excluded. We divided the number of visits in each census tract by census tract child population to estimate census tract nonemergent ED visits per 100 children. We obtained COI values for each census tract, including its domains: education, health and environment, and social and economic. We examined the association between census tract ED visits per 100 children and census tract COI score by using Pearson correlation coefficient. We then grouped census tracts into quartiles of ED visits per 100 children and compared COI across quartiles using one-way ANOVA.

Results: 75,372 nonemergent PED visits occurred across 393 census tracts in Allegheny County during the study period. Census tract nonemergent ED visits per 100 children ranged from 0 to 102 and the overall COI ranged from 1 to 100 (Table 1; Figure 1). Census tract nonemergent ED visits per 100 children and census tract COI were negatively correlated in 2018 (-0.61), 2019 (-0.61), and 2020 (-0.59) (p < 0.001 for all years). Overall COI and COI domains each declined across quartiles with increasing nonemergent ED visit rates (Table 1). For example, among census tracts in the lowest nonemergent ED visit rate quartile, overall COI mean score in 2018 was 73.6 and among census tracts in the highest nonemergent ED visit rate quartile, overall COI score was 20.5 (p < 0.001).Conclusion(s): Census tracts with higher nonemergent ED visit rates had fewer community resources, as indicated by the COI. These results indicate the potential for structural community-based approaches to reduce nonemergent ED visits.
Curriculum Vitae - Mark HincapieCurriculum Vitae - Hincapie.pdf.pdf
Figure 1. Map of County COI and Quartiles of Census Tract ED Rates