DOI: 10.1161/circ.148.suppl_1.15684 ISSN: 0009-7322

Abstract 15684: Adherence and Persistence to SGLT2 Inhibitors in Patients With Heart Failure

Shogo Kono, Jungyeon Moon, Alice Chong, Jiming Fang, Dennis T Ko, Peter Austin, Clare Atzema, David Naimark, Jacob A Udell, Therese Stukel, Karen Tu, Gillian Booth, Cynthia Jackevicius
  • Physiology (medical)
  • Cardiology and Cardiovascular Medicine

Introduction: While originally used solely as antihyperglycemic agents, sodium-glucose cotransporter 2 inhibitors (SGLT2i) are now recommended for use in heart failure (HF) to reduce hospitalization and cardiovascular mortality. Unfortunately, benefits of treatment can only be fully realized if a patient is adherent to therapy. While SGLT2i adherence has been found to be suboptimal in diabetes populations, SGLT2i adherence has not been well-studied in patients with HF. Since medication non-adherence in HF tends to be high in general, and can lead to worse clinical outcomes, this study aimed to estimate SGLT2i adherence/persistence rates in a population-based cohort of HF patients.

Methods: This retrospective cohort study included patients ≤65 years discharged alive from a hospital in Ontario, Canada between 4/2016 and 3/2021, who had a primary diagnosis of HF and were dispensed a SGLT2i prescription. Prescription claims were used to identify and calculate 6-month and 1-year SGLT2i adherence and 1-year persistence, from the first date of fill for an SGLT2i. Adherence was calculated as the proportion of days covered (PDC), with PDC≥80% defined as “good” adherence. Persistence was calculated with a “permissible” gap of 30 days allowed before being considered non-persistent. A cumulative incidence function was utilized to account for the competing risk of death in persistence calculations.

Results: There were 5,257 eligible patients (59% male, mean age 76.94±7 years, 73.8% with diabetes), for whom empagliflozin (64.9%), dapagliflozin (25.5%) and canagliflozin (9.5%) were used. The overall mean PDC was 83.9% at 6-months and 80.5% at 1-year, with the proportion of patients with “good” adherence (PDC≥80%) at 76.2% and 72.5%, respectively. Persistence to all prescribed SGLT2i was 71.3% at 1-year, with highest persistence with dapagliflozin at 74.4% and lowest persistence with canagliflozin at 68.1% [p=0.01].

Conclusion: In this HF cohort on SGLT2i, nearly 3 out of 4 patients were highly adherent and persistent at one year. SGLT2i adherence/persistence rates are higher than seen in studies in diabetes populations. Further studies are needed to identify those at risk of non-adherence/non-persistence and to examine association with clinical outcomes.

More from our Archive