Identifying targets for increasing evidence-based use of novel cardiovascular medications in the Veterans Affairs System: Provider-level characterization

Abstract

Background: Novel therapies such as angiotensin receptor-neprilysin inhibitors (ARNI), sodium-glucose cotransporter-2 inhibitors (SGLT2i), and glucagon-like peptide-1 receptor agonists (GLP-1 RA) have well-established benefit for patients with heart failure (HF) and coronary artery disease (CAD), but real-world use remains low. We aimed to characterize provider-level variation in the use of these medications in the Veterans Health Administration (VA). Methods: Using VA health record and administrative data, two patient cohorts were retrospectively identified with hospitalization from 2017-2023 for either 1) HF or 2) CAD with concurrent Type 2 diabetes (CAD+T2D). Medication utilization and associated provider-level characteristics were assessed for ARNI and SGLT2i for HF and SGTL2i or GLP-1 RA for CAD+T2D. Providers were considered users if they had an outpatient visit with a patient who had a filled prescription for a given novel medication class within 6 months of discharge, regardless of when or by whom the prescription was initiated. Descriptive and summary statistics were used to characterize use, non-use, and new starts of medications by provider specialty. Histograms and median odds ratios were used to characterize provider-level variability, and logistic regression was used to estimate likelihood of medication use by specialty. Results: The HF and CAD+T2D cohorts included 83,849 and 71,678 unique patients, respectively, with 14,645 unique providers. There were 322,156 qualifying clinical encounters in the HF cohort and 251,895 encounters in the CAD+T2D cohort. There were no substantial differences in patterns of use over time for male vs female providers or by provider training path, but there were differences by specialty. Patients with an endocrinology visit had the highest likelihood of receiving SGLT2i or GLP-1 RA for CAD+T2D (RR 1.81, 95% Confidence Interval [CI] 1.74-1.88). Patients with a cardiology visit had the highest likelihood of both ARNI (RR 1.81 [95% CI 1.75-1.88] and SGLT2i treatment for HF (RR 1.44 [95% CI 1.40-1.49]). Patients with a nephrology visit had lower likelihood of use for all 3 medication classes. Cardiology providers had the highest rate (11.9%) and largest number (n=11,397) of new starts for ARNI. For SGLT2i, endocrinology had the highest rate (12.9%) but the smallest number (n=794) of new starts. For SGTL2i or GLP-1 RA for CAD+T2D, endocrinology had the highest rate (13.8%) and primary care had the highest number of new starts (n=7,426). In total, <10% of visits had a new start and >70% of patients were untreated with novel medication classes. Conclusion: Utilization of ARNI, SGLT2i, and GLP-1 RA has increased over time with substantial variability by provider specialties. Cardiologists and endocrinologists were mostly likely to use novel cardiovascular medications but accounted for different proportions of total HF and CAD care, and there was substantial provider-level variation across all prescribing subspecialities. Future work may focus on developing, testing, and refining implementation strategies for HF and CAD.