(P226) PHYSICIAN AND PATIENT CHARACTERISTICS ASSOCIATED WITH PRESCRIPTION OF DIABETES MEDICATION WITH CARDIORENAL BENEFIT: A POPULATION-BASED ANALYSIS FROM ALBERTA, CANADA

Background: Randomized control trials have demonstrated reductions in cardiovascular and renal complications among patients with type 2 diabetes treated with newer classes of antihyperglycemic medications such as sodium glucose-cotransporter 2 (SGLT2i) inhibitors and glucagon-like peptide receptor agonists (GLP1RA). However, uptake of these drugs within clinical practice remains relatively low. Our study objective was to assess the physician and patient characteristics associated with prescriptions of antihyperglycemic medications with cardiorenal benefit.

METHODS AND RESULTS: Using linked administrative data from Alberta, Canada (2008-2021), adults (≥18 years) with type 2 diabetes and a new prescription of any antihyperglycemic medication were included. Prescriptions were categorized into those with cardiorenal benefit (SGLT2i or GLP1RA) and those without cardiorenal benefit. Patient-level and physician-level predictors of dispensation of prescriptions for agents with cardiorenal benefit were identified using multilevel logistic regression models, accounting for clustering of physician prescribing practices.

Of the 384,078 patients with new antihyperglycemic prescriptions (mean age 56.6 years, 46.4% female, 23.8% with >3 comorbidities), 5.2% (n=19,915) of prescriptions were for agents with cardiorenal benefit (SGLT2i and GLP1RA). These 19,915 prescriptions were most common provided by family physicians (76.4%, n=15,228) and physicians reimbursed via fee-for-service (69.5%, n=13,489). Patient characteristics associated with an increased odds of SGLT2i and GLP1RA prescription included male sex (odds ratio (OR) 2.96, 95%CI 2.84-3.08 compared to female sex) and no comorbidities (OR 1.82, 95%CI 1.11-1.25 when compared to 4+ comorbidities). Patients with higher quintiles of household income also had a higher odds of SGLT2i and GLP1RA prescriptions (OR 2.34, 95%CI 2.12-2.49 highest quintile versus lowest quintile). Physician factors associated with an increased odds of prescription included a higher clinical workload (OR 2.39 95%CI 1.98-2.89 when comparing highest tertile versus lowest tertile of annual clinic volume). Additionally, cardiology and endocrinology specialities were more likely to prescribe cardiorenal protective antihyperglycemics, despite accounting for the minority of overall prescriptions.

Conclusion: Patient and physician factors are associated with prescription of antihyperglycemic agents which confer cardiorenal benefits. Interestingly, patients with a higher burden of comorbidities were less likely to receive an SGLT2i or GLP1RA, which may suggest a disparity in prescription rates among those who may benefit from these medications most. Furthermore, while family physicians accounted for the majority of total SGLT2i and GLP1RA prescriptions, they were less likely to prescribe SGLT2is or GLP1RAs compared to endocrinologists or cardiologists. These findings can aid policymakers in targeting specific groups of physicians to improve the incongruence between guideline recommendations and diffusion rates into clinical practice.