Social vulnerability index predictive of being 'left without being seen' in the emergency department

This study assesses whether the Social Vulnerability Index (SVI) predicts the likelihood of being 'Left Without Being Seen' (LWBS) in the emergency department (ED). A cohort of 73,044 patient encounters between May 1, 2022 and April 30, 2023 in an academic, urban, tertiary care ED was analyzed using logistic regression to evaluate for SVI as a predictor of LWBS disposition.

Patient addresses were geocoded using ArcGIS Pro 3.3 to obtain SVI data at the census tract level. Overall SVI and its four component themes (socioeconomic status, household characteristics, racial and ethnic minority status, housing type and transportation) were analyzed for predictive value. Statistical models assessed LWBS likelihood for every 0.1-unit increase in SVI.

LWBS patients accounted for 5.4 % (3880) of total patient encounters, with higher mean SVI scores (0.73, SD = 0.27) compared to seen patients (0.67, SD = 0.30, p < 0.001). Each 0.1-unit increase in SVI was predictive of a 7 % increase in the likelihood of being LNS (p < 0.01). Among component themes, racial and ethnic minority status showed the strongest correlation, with a 13 % increase in the likelihood of being LWBS per 0.1-unit increase in SVI.

Higher SVI scores strongly predicted being LWBS overall and across all themes. SVI may be used to predict a person's risk of limited access to emergency care, as represented by arriving at the ED but leaving before being seen. Integrating SVI into ED and health system workflows could aid in identifying at-risk populations, addressing systemic barriers, and improving equitable access to care.