Social Emergency Medicine
Background: In 2017, Yale New Haven Hospital (YNHH) was selected to participate in the Centers for Medicare and Medicaid’s Accountable Health Communities Model to provide social services navigation for patients with high unmet social needs in the Greater New Haven Area. This study examines how prevalence of structural determinants of health inequity is associated with increased emergency department (ED) utilization, which could be addressed by a hospital social services navigation program.
Methods: This is a cross-sectional study of YNHH ED visits in the Greater New Haven Area from January 2014 to December 2015. Frequent ED utilizers are defined as patients had four or more ED visits during the study year. Records were geocoded and linked to data from a community-based social needs survey conducted by DataHaven. Chi-square and t-test analyses were done to compare the frequency of ED utilization and social determinants of health, specifically food insecurity, housing instability, and transportation. Multivariate logistic regression was subsequently conducted to look at factors impacting frequency of ED utilization.
Results: A total of 139,383 patient encounters were included in the analysis. Several person-level factors—age of sixty years and older (OR: 3.57; CI [3.38-3.77]), female gender (OR: 1.40; CI [1.32-1.48]), and history of chronic pain (OR: 1.30; CI [1.13-1.50])—significantly increased the likelihood of being having and ED visit for an ACSC. Both person-level factors—homelessness (OR: 3.74; CI [2.35-5.95]), history of abuse (OR: 1.79; CI [1.54-2.09]), and history of substance use disorder (OR: 1.53; CI [1.37- 1.69])—and neighborhood-level housing instability (1.36; CI [1.33-1.39]) were associated with frequent ED utilization.
Conclusion: Being Black or African American, publicly insured, and unemployed were independently associated with increased rates of frequent ED utilization. Residence in an area with greater prevalence of housing instability, was also found to be associated with increased ED utilization. Our research can support endeavors to systematically direct services toward the most vulnerable patient populations by geographical location.