Medications for Alcohol Use Disorder: Rates and Predictors of Prescription Order and Fill in Outpatient Settings

Dominic Hodgkin; Alisa B Busch; Alene Kennedy-Hendricks; Hocine Azeni; Constance M Horgan; Lori Uscher-Pines; Haiden A Huskamp

doi:10.1007/s11606-024-09002-3

Alcohol use disorders (AUD) are prevalent and responsible for substantial morbidity and mortality; yet efficacious treatments are underused. Previous studies have identified demographic and clinical predictors of medication fills, yet these studies typically do not include patients who were prescribed a medication but did not fill it.BACKGROUNDAlcohol use disorders (AUD) are prevalent and responsible for substantial morbidity and mortality; yet efficacious treatments are underused. Previous studies have identified demographic and clinical predictors of medication fills, yet these studies typically do not include patients who were prescribed a medication but did not fill it.To examine rates of and factors associated with prescription order and prescription fill for medications for AUD (MAUD) among individuals diagnosed with AUD in outpatient settings.OBJECTIVESTo examine rates of and factors associated with prescription order and prescription fill for medications for AUD (MAUD) among individuals diagnosed with AUD in outpatient settings.In a cross-sectional analysis, we used multivariate logistic regression to identify factors associated with prescription order and fill.DESIGNIn a cross-sectional analysis, we used multivariate logistic regression to identify factors associated with prescription order and fill.We used data from the Optum Labs Data Warehouse that linked 2016-2021 de-identified claims and electronic health record (EHR) data, allowing us to observe prescription orders and whether they were filled. We identified 14,674 patients aged ≥ 18 who had an index outpatient encounter with an AUD diagnosis in the EHR.PATIENTSWe used data from the Optum Labs Data Warehouse that linked 2016-2021 de-identified claims and electronic health record (EHR) data, allowing us to observe prescription orders and whether they were filled. We identified 14,674 patients aged ≥ 18 who had an index outpatient encounter with an AUD diagnosis in the EHR.We computed the proportion for whom a MAUD prescription was ordered within 1 year of index visit, and for whom one was filled within 30 days of the order.KEY MEASURESWe computed the proportion for whom a MAUD prescription was ordered within 1 year of index visit, and for whom one was filled within 30 days of the order.5.8% of the sample had a MAUD prescription order within 1 year of their index visit. Among those with an order, 87% filled their MAUD prescription within 30 days of receipt (i.e., 5.1% of full sample). After multivariable adjustment, receipt of a MAUD prescription order was more likely for patients who were female (adjusted odds ratio (aOR) [95%CI] = 1.44 [1.24-1.67]), or had moderate or severe AUD (1.74 [1.50-2.01]). Patients receiving an order were more likely to fill it if they had a comorbid mental disorder (1.64 [1.09-2.49]).KEY RESULTS5.8% of the sample had a MAUD prescription order within 1 year of their index visit. Among those with an order, 87% filled their MAUD prescription within 30 days of receipt (i.e., 5.1% of full sample). After multivariable adjustment, receipt of a MAUD prescription order was more likely for patients who were female (adjusted odds ratio (aOR) [95%CI] = 1.44 [1.24-1.67]), or had moderate or severe AUD (1.74 [1.50-2.01]). Patients receiving an order were more likely to fill it if they had a comorbid mental disorder (1.64 [1.09-2.49]).The low rate of prescription orders was notable. Low use of MAUD appears to result chiefly from prescription order decisions, rather than from prescription fill decisions made by patients.CONCLUSIONSThe low rate of prescription orders was notable. Low use of MAUD appears to result chiefly from prescription order decisions, rather than from prescription fill decisions made by patients.