Can people with type 2 diabetes live longer than those without? A comparison of mortality in people initiated with metformin or sulphonylurea monotherapy and matched, non-diabetic controls.
Level 3 - non-randomized controlled study
Retrospective matched cohort study using electronic health record data.
PubMed 25041462 · doi:10.1111/dom.12354
What was done
A retrospective matched cohort study analyzed electronic health record data from the UK Clinical Practice Research Datalink (CPRD) from the year 2000 onward. Researchers identified patients with type 2 diabetes initiated on first-line monotherapy with either metformin (n = 78,241) or a sulphonylurea (n = 12,222) and matched them to individuals without diabetes (n = 90,463). All-cause mortality was compared across groups over 503,384 person-years of follow-up using parametric survival models adjusted for relevant covariates.
What was found
Across 7,498 total deaths, unadjusted mortality rates per 1,000 person-years were 14.4 for metformin monotherapy versus 15.2 for their matched controls, and 50.9 for sulphonylurea monotherapy versus 28.7 for their matched controls. Relative to the metformin group (survival time ratio [STR] = 1.0), adjusted median survival time was 15% lower in matched non-diabetic controls (STR = 0.85, 95% CI 0.81-0.90) and 38% lower in diabetic patients treated with sulphonylurea monotherapy (STR = 0.62, 95% CI 0.58-0.66).
Why it matters
This observational study showed that patients prescribed metformin lived longer than matched individuals without diabetes, generating hypotheses about potential survival benefits of metformin beyond glycemic control and highlighting risks associated with sulphonylureas.
Limits
The retrospective observational design cannot establish causality and is vulnerable to unmeasured confounding and confounding by indication. Reasons for selecting sulphonylureas over metformin (such as renal impairment or greater baseline frailty) were not fully controllable from registry data, and matching non-diabetic individuals via primary care records may introduce surveillance or selection biases.
Cited by
- supports Observational research observed that diabetic patients taking metformin had longer survival than non-diabetic controls.