(056) Investigating ethnic differences in the comparative effectiveness of classes of second-line antidiabetic medications at preventing cardiovascular outcomes.

Background: There are stark ethnic inequalities in type 2 diabetes (T2D) outcomes in the United Kingdom (UK). One contributing factor may be that therapeutic guidelines were derived using data from predominantly white European populations, potentially leading to suboptimal care for other ethnic groups. However, little evidence is available regarding ethnic differences in the comparative effectiveness of commonly prescribed antidiabetic medications.

Objectives: 1. To compare the effectiveness of sulfonylureas (SUs), dipeptidyl peptidase 4 inhibitors (DPP4is) and sodium-glucose cotransporter 2 inhibitors (SGLT2is) when used in combination with metformin as second-line treatment at preventing cardiovascular (CV) outcomes;
2. To determine if the comparative treatment effects of SUs, DPP4is and SGLT2is for prevention of CV outcomes differ by ethnicity.

Methods: Using data from CPRD Aurum, we conducted a retrospective cohort study of adults (18+) with T2D in the UK prescribed second-line treatment of an SU, DDP4i or SGT2i in combination with metformin between January 2012 to April 2022. The study only included new users, excluding individuals who had been prescribed any diabetes medication except metformin three years before index. The primary outcome was major adverse cardiovascular events (MACE, a composite endpoint of myocardial infarction, ischaemic stroke, and cardiovascular death). The secondary outcomes are individual components of MACE. The effectiveness of DDP4is vs SUs and SGT2is vs SUs were estimated in the overall population and by ethnicity (white, south Asian and black) using Cox proportional hazards models. The preliminary analyses used an intention to treat approach, with adjustment for age and sex, to broadly assess of ethnic disparities. Subsequent analyses use more robust causal methods to understand these discrepancies.

Results: Among 203,794 people initiated on second-line treatment, the distribution was 43%, 39% and 18 for SUs, DPP4is and SGLT2is, respectively. In ethnic subgroups (n=150,965 whites; n=26,127 south Asians; n=10,045 blacks), respective distributions were consistent. After age and sex adjustment, the hazards ratio (HRs) for MACE comparing DDP4is vs SUs were 0.96 (95% confidence interval (CI) 0.93-1.00) overall, 0.97 (95% CI 0.94-1.01) in the white population; 0.88 (95% CI 0.78-1.00) in the south Asian population; and 1.14 (95% CI 0.92-1.43) in the black population.

Conclusions: There was no significant evidence of differences in the effectiveness of SUs vs DPP4is in the overall population and ethnic subgroups in the preliminary analyses.