LAUSR

INTRODUCTION This study aimed to compare the clinical documentation of social determinants of health (SDoH) data and its associations with COVID-19 clinical outcomes among people with HIV (PWH) and people without HIV (PWoH) in the United States (US) using the nationally sampled electronic health records (EHR) repository called the National COVID Cohort Collaborative (N3C). METHODS Information on all adult COVID-19 cases was retrieved in the N3C platform from 2020 through 2023. Under the Super-Utilizer Model, we categorized SDoH measures (identified by ICD-10-CM diagnosis codes ["Z codes"]) into different domains, i.e., community/social domain, socioeconomic domain, and environmental domain. Logistic regression models were employed for analysis. RESULTS Among 41,759 PWH and 6,795,369 PWoH with a COVID-19 infection, 49.8% and 24.0% had at least one Z code documented in the EHR, respectively. For both PWH and PWoH, problems related to employment and housing in the socioeconomic domain and problems related to legal and psychosocial circumstances in the community/social domain were the most commonly documented Z codes associated with severe COVID-19 outcomes; while problems related to upbringing in community/social domain and economic issues in socioeconomic domain were associated with reduced risk of severe COVID-19 outcomes. CONCLUSION While documented Z code associations with severe COVID-19 outcomes were similar between PWH or PWoH, they varied between SDoH factors. Identification of Z code associations with severe COVID-19 related outcomes allows for intervention at the population health level, and even more broadly for tracking and resource allocation by public health officials or policymakers.