Effects of antidiabetic drugs on mortality risks in individuals with type 2 diabetes: A prospective cohort study of UK Biobank participants

Abstract

Objective:To investigate the mortality risk linked to prescription of different anti-diabetic medication classes.Design:Prospective population-based study.Setting:UK Biobank.Participants:410 389 of the 502 536 participants in UK Biobank with covariate data, clinical and prescription records were included in the analyses, 43 610 of which had been diagnosed type 2 diabetes (T2D). A nearest neighbour covariate matching (NNCM) algorithm based on covariates with relevant effects on survival was applied to match cohorts of anti-diabetic medication class users to minimally differing control cohorts, either with a T2D diagnosis or without. Kaplan Meier estimates and Cox proportional models were used to evaluate survival differences and hazard ratio between drug classes and controls.Main outcome measures:All-cause mortality and causes of death.Results:13667 (3.3%) individuals died during a median of 12.2 years of follow-up. After applying NNCM, participants with T2D on metformin (average hazard ratio 0.39, 95% confidence interval 0.31 to 0.49) or SGLT2I (average hazard ratio 0.58, 95% confidence interval 0.36 to 0.93) have an increased survival probability compared to matched individuals with T2D. When compared to matched individuals without T2D, the survival probability of individuals with T2D increases only if prescribed SGLT2I (average hazard ratio 0.31, 95% confidence interval 0.19 to 0.51). NNCM based analysis of matched individuals with T2D on both SGLT2I and metformin versus metformin only reveals increased survival in the presence of SGLT2I (average hazard ratio 0.29, 95% confidence interval 0.09 to 0.91), also when compared to matched identical individuals without T2D (average hazard ratio 0.05, 95% confidence interval 0.01 to 0.19). All the other anti-diabetic drugs analyzed are either detrimental in prolonging lifespan (insulin, thiazolidinediones, and sulfonylureas), or have no effect (DPP4 inhibitors and GLP1 receptor agonists).Conclusion:The use of the current first-line anti-diabetic treatment, metformin, or sodium glucose cotransporter 2 inhibitors (SGLT2I) increases the survival probability compared to matched individuals with diabetes using other anti-diabetic drugs. Only individuals on SGLT2I experience increased survival when compared to individuals without T2D.