Identifying predictors of sodium-glucose cotransporter 2 inhibitor and glucagon-like peptide 1 receptor agonist use in hospital among adults with diabetes

AIMS: To identify factors associated with use of novel diabetes medications among patients hospitalized under general internal medicine.

METHODS: We conducted a cohort study of patients with type 2 diabetes mellitus (T2DM) hospitalized in Ontario, Canada between 2015 and 2020. We evaluated the patient- and physician-level factors associated with sodium-glucose cotransporter 2 inhibitor (SGLT2) and glucagon-like peptide 1 receptor agonist (GLP1R) use using a multivariable logistic regression model.

RESULTS: There were 253,152 hospitalizations and 68,126 involved patients who had T2DM. Prior to discharge, 3.7 % (N = 2490) of patients with T2DM received an SGLT2 and 0.2 % (N = 121) received a GLP1R. The strongest predictors for receiving a novel diabetes medication were hemoglobin A1C > 9.0 % (Odds Ratio (OR) = 1.81, 95 % Confidence Interval (CI) 1.28, 2.60) and patients aged 40-60 compared with patients <40 years old (OR = 1.81, 95 % CI 1.33, 2.68). The strongest predictors for not receiving a novel diabetes medication were dementia (OR = 0.47, 95 % CI 0.39, 0.56) and creatinine ≥200 μmol/L (OR = 0.11, 95 % CI 0.08, 0.15). Overall, 46.8 % of patients hospitalized with T2DM not receiving a novel diabetes medication would potentially benefit from an SGLT2 inhibitor.

CONCLUSIONS: Novel diabetes medications were rarely continued or initiated during hospitalization despite a high prevalence of cardiovascular disease, raising the concern for systematic under-utilization after discharge.