Abstract
IMPORTANCE: A better understanding of how health-related social needs (HRSNs) are associated with health care utilization may help health systems prioritize efforts to address HRSNs. OBJECTIVES: To examine the association of HRSNs with inpatient (IP) and emergency department (ED) utilization and whether resolution of HRSNs is associated with a reduction in utilization. DESIGN, SETTING, AND PARTICIPANTS: This retrospective cohort study included 166 682 Medicare and Medicaid recipients in a single health system who were screened for HRSNs under the Accountable Health Communities (AHC) program from June 5, 2018, to January 31, 2022. A follow-up sample included the subset of patients with multiple screenings in the study period. Statistical analysis was performed from June 2024 to September 2025. EXPOSURES: Health-related social needs assessed under the AHC program included housing stability, housing quality, food security, transportation, utilities, and interpersonal safety. Measures included the presence of individual HRSNs, number of needs, and resolution of need at follow-up. MAIN OUTCOMES AND MEASURES: IP admissions and ED visits were ascertained for 6 months before baseline screening and 6 months after follow-up. Models assessed the association of baseline HRSNs and IP and ED admissions in the 6 months prior to screening, and resolution of HRSNs with IP and ED admissions in the 6 months after follow-up. Models were mutually adjusted for HRSNs as well as demographic characteristics and comorbidities. RESULTS: There were 166 682 patients (43.4% aged ≥65 years; 60.3% women) with baseline screening in the sample. All 6 baseline HRSNs were associated with ED use after adjustment for all covariates: 5 with higher odds of ED visits and 1 (housing quality) with lower odds of ED use; point estimates were largest for housing stability (odds ratio [OR], 1.25; 95% CI, 1.18-1.33) and transportation (OR, 1.31; 95% CI, 1.25-1.38). Emergency department visits occurred for 26.9% of patients with housing stability needs and 25.6% of those with transportation needs. After adjustment for other HRSNs, demographic characteristics, and health conditions, only 2 HRSNs retained an association with an increase in IP admissions: housing stability (OR, 1.34; 95% CI, 1.23-1.47) and transportation (OR, 1.16; 95% CI, 1.08-1.24). In the subsample with follow-up screening (n = 15 139), resolution of HRSNs was not associated with utilization after adjustment for other HRSNs, demographic characteristics, and comorbidities. CONCLUSIONS AND RELEVANCE: In this cohort study, most HRSNs were associated with ED visits, with strongest associations for housing instability and transportation. Associations were not present for adjusted models for resolution of HRSNs. Further research on the fluctuation of individual HRSNs over time is warranted to better understand if resolution of needs is associated with reduced health care utilization.