Abstract
SGLT2 inhibitors show promise in reducing hospitalization for heart failure in diabetics, but their long-term effects and time-dependency remain unclear. We conducted a retrospective nested case-control study within a large type 2 diabetic cohort (n = 11,209) using electronic health records. Cases (heart failure hospitalization, n = 352) were matched to controls (n = 1372) based on age, sex, cohort entry date, and diabetes duration. Matched-set conditional logistic regression was used to estimate hazard ratios (HRs) for antidiabetic drug class and heart failure hospitalization risk. SGLT2 inhibitors were associated with a significant reduction in heart failure hospitalization risk (adjusted HR 0.56, 95% CI 0.38-0.82, p = 0.028). This protective effect appeared more pronounced with a longer duration of treatment, suggesting a potential cumulative benefit. Time-varying analysis within propensity score-matched cohorts revealed a progressive decrease in hospitalization risk with continued SGLT2 inhibitor use, indicating a strengthening effect over time (greedy nearest neighbor: HR 0.52, CI 0.31-0.87, p = 0.015; optimal matching: HR 0.54, CI 0.34-0.85, p = 0.008). While promising, further investigation with larger datasets is warranted to definitively confirm these findings.