Housing status, patient characteristics, and ED utilization associated with medication prescribing at ED discharge among homeless and nonhomeless adults in urban hospitals in the United States

This cross-sectional study used a weighted sample of ED visits contained in the 2010-2015 years of the National Hospital Ambulatory Care Survey-Emergency Department (NHAMCS-ED) dataset. The purpose of this study was to: 1) identify differences in predisposing, enabling, and need characteristics, and ED use and medication prescribing characteristics between homeless and nonhomeless ED users; 2) assess the association between housing status and medication prescribing at ED discharge, and identify variables contributing to the disparity in medication prescribing between homeless and nonhomeless ED users; and 3) assess the predisposing, enabling, need, and ED use characteristics that predict medication prescribing at ED discharge among homeless ED users. This research is guided by the Andersen-Gelberg Behavioral Model for Vulnerable Populations. There were a total of 502,614,359 visits to EDs located within a MSA made by homeless and nonhomeless adults 18 years of age and older. About 0.9% of these visits were made by homeless individuals. Age, mental health diagnosis, substance use diagnosis, primary payer, and patient-reported pain differed significantly between homeless and nonhomeless ED users. A significantly greater proportion of homeless ED users arrived to the ED via ambulance, and was seen in the last 72 hours. Homeless ED users tended to have longer ED visits, and ED disposition differed significantly between homeless and nonhomeless ED users. A significantly smaller proportion of homeless ED users were prescribed a medication at ED discharge, and an opioid medication at ED discharge. There was no difference in the likelihood of medication prescribing at ED discharge between homeless and nonhomeless ED users after controlling for predisposing, enabling, need, and ED use characteristics. ED diagnosis was the greatest contributor to the disparity in medication prescribing at ED discharge between homeless and nonhomeless ED users. Among homeless ED users, visits covered by Medicare and other payers were significantly more likely to result in medication prescribing at ED discharge compared to nonhomeless ED users covered by private insurance. Homeless ED users with no substance use condition diagnosis were significantly more likely to be prescribed a medication at ED discharge compared to those with a substance use condition diagnosis