Latino-White Disparities in Screening and Receipt of Medications for Unhealthy Alcohol Use in the OCHIN Network

Context: Unhealthy alcohol use and the sequelae of alcohol-related diseases continue to be an important health concern; there is evidence that the COVID-19 pandemic was associated with increased alcohol use and related mortality. The US Preventative Services Task Force in 2018 updated its recommendation to screen for unhealthy alcohol use in primary care settings among patients >18 years of age. Given existing cardiovascular health disparities in Latino populations, screening for unhealthy alcohol use and alcohol use disorders (AUD) is important in community health settings which serve a high proportion of Latino patients.

Objective: We evaluated the odds of alcohol screening and receipt of medications for AUD in adults by ethnicity and language preference.

Dataset: This was an observational cohort study using OCHIN data, a multi-state electronic health record (EHR) network of community-based health clinics.

Population: Adults seen in the OCHIN network between 2012-2020.

Study Design and Analysis and Outcomes: Logistic regression using general linear modeling equations estimated the covariate-adjusted relative odds of receipt of alcohol screening and among those with AUD, receipt of medications for AUD. Our predictor of interest was ethnic-language preference groups (Non-Hispanic White [NHW], Latino-Spanish language preferred, Latino-English language preferred).

Results: There were 1,781,206 patients in the sample across 26 states with average age of 41.5 years (SD=15.3). Over half were female (57%); 22% were never insured, and 51% reported public insurance. There were 57% NHW, 27% Latino-Spanish language preferred, and 16% Latino-English language preferred patients. Overall, 41% received alcohol screening during this period. Both Spanish preferring (aOR=2.14, 95% CI: 2.11, 2.17) and English preferring (aOR=1.41, 95% CI 1.39, 1.42) Latinos had increased odds of alcohol screening compared to NHW. However, among those with an AUD, both Latino groups, had lower odds of receipt of medications for AUD versus NHW (Latino-Spanish preferred aOR=0.74, 95% CI: 0.64, 0.84; Latino-English preferred aOR=0.77, 95% CI 0.70, 0.85)

Conclusions: In a multi-state cohort of patients seen in primary care, we found that regardless of language preference, Latinos had higher rates of alcohol screening compared to NHW, but among those with AUD, had lower odds of receipt of medications for AUD, a mismatch in evidence-based care for Latinos with addictions.