

Predictors of Emergency Department Return Among Patients Discharged After Treatment for Alcohol Withdrawal Syndrome
Wednesday, May 20, 2026 12:40 PM to 12:48 PM · 8 min. (America/New_York)
L506 - L507: Level L
Abstracts
Substance Abuse/Toxicology
Information
Abstract Number
428
Background and Objectives
Alcohol withdrawal syndrome (AWS) is a frequent cause of emergency department (ED) visits, and some patients return soon after discharge. Identifying clinical and social risk factors for ED short-term recidivism may inform targeted interventions and improve post-ED outcomes.
Methods
We conducted a retrospective review of 318 ED encounters for AWS at a single academic medical center from January 1, 2020, and December 31, 2023. Demographic, psychiatric, housing, insurance, and discharge pharmacology variables were collected. The primary outcomes were 72-hour ED return and 30-day hospital admission. Group comparisons were performed using chi-square, Shapiro-Wilk, Fisher’s exact, and Mann–Whitney U tests. Variables with significant univariate associations were entered into multivariable logistic regression, with results reported as odds ratios (ORs) with 95% confidence intervals (CIs).
Results
On multivariable logistic regression, schizophrenia (OR 8.59, 95% CI 2.07–35.61), Medicaid insurance (OR 4.43, 95% CI 1.39–14.14), and anxiety (OR 3.53, 95% CI 1.10–11.39) were independently associated with increased 72-hour ED return . Homelessness was independently associated with higher 30-day admission (OR 3.19, 95% CI 1.26–8.07) . Depression was not independently associated with 30-day admission. Discharge tapers and ED treatment strategy were not associated with either 72-hour return or 30-day admission.
Conclusion
Among patients discharged after ED treatment for AWS, Medicaid insurance, anxiety, schizophrenia, and homelessness were independently associated with short-term return or admission. These findings suggest that social and psychiatric factors may play a greater role in post-discharge outcomes than pharmacologic regimen, highlighting opportunities to strengthen follow-up pathways and address social determinants of health. Larger, multicenter studies with adequately powered multivariable models are warranted to validate these associations.
CME
0.75
Disclosures
Access the following link to view disclosures of session presenters, presenting authors, organizers, moderators, and planners:

