Association Between Metformin Adherence and All-Cause Mortality Among New Users of Metformin: A Nested Case-Control Study

Background: Metformin presents better survival rates than other oral antidiabetics in the treatment of type 2 diabetes. However, these benefits may be dampened by inadequate treatment adherence. Objective: We aimed to investigate the relationship between adherence level to metformin therapy and all-cause mortality over 10 years in incident metformin users. Methods: A nested case-control study was conducted using a large cohort of beneficiaries of the Quebec public drug insurance plan, aged 45 to 85 years, who initiated metformin between 2000 and 2009. Each case of all-cause death during follow-up was matched with up to 10 controls. Adherence to metformin was measured using the medication possession ratio (MPR). Conditional logistic regression models were used to estimate rate ratios (RRs) for mortality between adherent (MPR 80%) and nonadherent patients (MPR < 80%). Subgroup analyses were conducted according to age (45-64 and 65-85 years) and comedication use (antihypertensive/cardiovascular drugs and statins). Results: The cohort included 82 720 incident metformin users, followed up for 2.4 [0.8-4.4] years (median [interquartile range]) and 4747 cases of all-cause deaths. Analyses revealed decreased mortality risks after long-term adherence to metformin. Specifically, RRs were 0.84 (95% CI = [0.71-0.98]) and 0.69 [0.57-0.85] after 4 to 6 and 6 years of adherence to metformin, respectively. Survival benefits of long-term adherence (4 years) were also observed across most subgroups and particularly in patients using neither antihypertensive/cardiovascular drugs nor statins (0.57 [0.41-0.77]). Conclusions: Long-term adherence to metformin is associated with decreased risks of all-cause mortality in incident metformin users. Further research should investigate whether survival benefits vary according to the comorbidity burden of patients.