Abstract 18569: Patient and Provider Factors Associated With the Initiation of Sodium-Glucose Cotransporter-2 Inhibitors (SGLT2i) in the Year Post FDA Approval for HFpEF and HFmrEF

Abstract

Introduction: Despite robust evidence for, and Food and Drug Administration (FDA) approval of sodium-glucose cotransporter-2 inhibitors (SGLT2i) for heart failure with preserved and mildly reduced ejection fraction (HFpEF/HFmrEF), use of this therapy is suboptimal. Hypothesis: We aimed to determine 1) prescribing context (inpatient/outpatient) and prescriber type of SGLT2i; 2) patient and provider level predictors of SGLT2i prescription in the year following FDA approval. Methods: All patients in a diverse, multi-state (Ohio and Florida), multi practice/hospital integrated system who had least 1 ambulatory encounter for HF and documented EF > 40% were included. Data come from the electronic health record. We compared patients who were prescribed SGLT2i in the year post FDA approval 2/24/22-2/24/23 against those who were not using multivariable logistic regression to assess independent predictors of SGLT2i prescription. Results: 20,255 patients with HFpEF/HFmrEF were included in the study. Mean age 73±9 yrs, female (49.8%), Black (17.4%), DM2 (34%), AFib (43%), CKD (35%), baseline medications: BB (52.4%), ACEi/ARB (41%), MRA (19.7%), ARNI (6.5%), SGLT2i (3.2%). There were 1118 (5.6%) patients were newly prescribed SGLT2i in the study period. SGLT2i were most commonly initiated by cardiology 54.2%, general medicine 25.3% and endocrinology 8.5%. Of prescriptions, 82.9% were outpatient, 17.1% inpatient. Independent patient level predictors of SGLT2i prescription were DM2 2.82 (2.09-3.82), p<0.001; AFib 1.42 (1.07-1.92), p=0.018; higher eGFR 1.16 (1.09-1.25), p<0.001 and increasing EF 0.88 (0.81-0.97), p<0.001. Insurance type/median income were not significant. Provider level predictors were total HF visits with each specialty: HF cardiology, 1.21 (1.13-1.29), p<0.001; general cardiology, 1.1 (1.04-1.15), p<0.001; and primary care, 1.1 (1.08-1.17), p<0.001. After excluding patients with DM2, predictors were total visits with HF specialists, and general cardiologists. Conclusions: In the year post FDA approval, there has been significant underutilization of SGLT2i. Diabetes status and provider factors were associated with SGLT2i use, the majority of which were initiated in the outpatient setting by cardiology providers.