(71) Race, Ethnicity & Language Differences in Medications for Alcohol Use Disorder Prescribing

*Background & Introduction: Alcohol use disorder (AUD) is the most prevalent substance use disorder, and recent research has highlighted disparities in alcohol-related morbidity and mortality. Specifically, Hispanic ethnicity is associated with higher age-adjusted death rates, higher rates of alcohol liver disease and lower rates of liver transplantation (1,2). Furthermore, data from the National Survey on Drug Use and Health reveal disparities in AUD treatment at the intersection race, ethnicity, gender and insurance status (3).

Hospital presentations can be reachable moments for AUD care engagement. Hospital-based prescribing of medications for AUD (MAUD) is associated with reduced risk of hospital readmission in multiple settings (4,5). However, differences in MAUD prescribing in hospital settings by patient race, ethnicity and preferred language are not well understood. Understanding these differences is needed to inform hospital-based interventions to promote equitable access to AUD treatment.

*

Methods: This was a single-site, retrospective cohort study using electronic medical record (EMR) data from adult emergency department (ED) and inpatient encounters from 2016 to 2024 with ICD-9/10 diagnosis codes for alcohol withdrawal syndrome (AWS). This large academic medical center averages over 50,000 annual hospitalizations and has an addiction consult team, established in 2014. Patients without an encounter in the health system (14 hospitals, 1.5 million annual patients) within the preceding 12 months were excluded.

First, we determined the prevalence of MAUD prescriptions (oral/injectable naltrexone, acamprosate, disulfiram) on discharge for emergency department (ED) and inpatient encounters. Second, for inpatient encounters, we used generalized linear models (GLM) adjusted for clustering within patient to identify provider (addiction consultation, primary team), patient [demographics, comorbidities, AUD severity, recent (within 90 day) MAUD prescription] and hospitalization characteristics (length of stay, ICU stay, year) associated with MAUD prescription on discharge. Third, we identified demographics associated with recent MAUD prescription. Length of stay and year of presentation were continuous variables, while the remainder were categorical variables.

*

Results: This cohort included 7946 ED visits and inpatient encounters, representing 3058 unique patients. Prevalence of MAUD prescriptions at discharge was 2.0% for ED-only encounters (2395 encounters, 1024 patients) and 35% for inpatient encounters (5551 encounters, 2564 unique patients). Among inpatient encounters, 77% received an addiction consultation; demographics were 24% female, 6% Black non-Hispanic, 16% Hispanic, 1% American Indian/Alaska Native, 74% White non-Hispanic, 11% non-English language preference (NELP) and 21% experiencing homelessness.

In GLM analysis, elevated rates of discharge MAUD prescriptions were associated with Hispanic ethnicity (aRR 1.22, 95% CI [1.05, 1.41], vs White non-Hispanic), Black non-Hispanic race/ethnicity (aRR 1.25, 95% CI [1.05, 1.48], vs White non-Hispanic), NELP (aRR 1.38, 95% CI [1.20, 1.59]) and addiction consultation (aRR 2.38, 95% CI [2.04, 2.77]). Lower rates were associated with prescription opioid use (aRR 0.82, 95% CI [0.75, 0.90]) and opioid use disorder (aRR 0.59, 95% [0.51, 0.69]).

For the outcome of recent MAUD prescription prior to AWS hospitalization, Hispanic ethnicity (aRR 1.32, 95% [1.07, 1.73], vs White non-Hispanic) was associated with higher rates of MAUD prescription, adjusted for demographics, opioid use disorder and prescription opioid use.

*Conclusion & Discussion: In this single-center cohort of patients hospitalized with AWS, MAUD prescriptions from the ED were rare, posing a missed opportunity. Among inpatient stays, supported by a robust addiction consult team, 35% included a MAUD prescription at discharge, and rates of discharge MAUD prescription were higher among racial/ethnic minorities and those with NELP. Hispanic ethnicity was associated with higher rates of recent MAUD prescription, prior to hospitalization for AWS, suggesting the importance of pharmacologic and non-pharmacologic interventions for equitable treatment of AUD.

Findings are limited by the use of medication data from a single health system, with variables defined by ICD codes and other EMR data. Furthermore, American Indian/Alaska Native peoples face significant healthcare disparities but were not well-represented in our cohort. However, these findings suggest that hospitalizations for AWS can be opportunities for antiracism in medicine.

*

References:
1. Karaye, I. M., Maleki, N., Hassan, N., & Yunusa, I. (2023). Trends in Alcohol-Related Deaths by Sex in the US, 1999-2020. JAMA network open, 6(7), e2326346. https://doi.org/10.1001/jamanetworkopen.2023.26346

2. Ayares, G., Díaz, L. A., Fuentes-López, E., Idalsoaga, F., Cotter, T. G., Dunn, W., Simonetto, D., Shah, V. H., Kamath, P. S., Lazarus, J. V., Bataller, R., Arrese, M., Wong, R. J., Singal, A. K., & Arab, J. P. (2024). Racial and ethnic disparities in the natural history of alcohol-associated liver disease in the United States. Liver international: official journal of the International Association for the Study of the Liver, 44(10), 2822–2833. https://doi.org/10.1111/liv.16023

3. Acevedo, A., Adams, R. S., Cook, B. L., Feltus, S. R., Panas, L., & Stewart, M. T. (2025). Disparities in Alcohol Treatment Use at the Intersection of Race, Ethnicity, Gender, and Insurance. Substance use & addiction journal, 46(1), 78–89. https://doi.org/10.1177/29767342241278871

4. Lambert, E., Regan, S., & Wakeman, S. E. (2025). The impact of addiction consultation and medication for opioid or alcohol use disorder on hospital readmission. Journal of General Internal Medicine. https://doi.org/10.1007/s11606-024-09301-9

5. Bernstein, E. Y., Baggett, T. P., Trivedi, S., Herzig, S. J., & Anderson, T. S. (2024). Outcomes After Initiation of Medications for Alcohol Use Disorder at Hospital Discharge. JAMA network open, 7(3), e243387. https://doi.org/10.1001/jamanetworkopen.2024.3387