Randomized trials found fewer deaths during follow-up among people assigned to GLP-1 drugs. The SELECT trial’s cardiovascular results apply to adults with overweight or obesity and established cardiovascular disease but without diabetes. A separate analysis of the same trial examined deaths from any cause.[2][3]
What SELECT found about deaths
Our review of deaths after respiratory infection covers a narrower outcome, including a secondary SELECT analysis. A result for deaths after infection does not establish whether the drug prevents infection.
Over a median of 3.3 years, 375 of 8,803 participants assigned semaglutide (4.3%) and 458 of 8,801 assigned placebo (5.2%) died. The hazard ratio was 0.81 (95% CI 0.71–0.93). The trial’s ordered testing plan stopped at an earlier secondary outcome, so all-cause mortality was not formally tested for superiority under that plan.[3][4]
What the pooled trials found
The study of GLP-1 use after age 80 examined health records in adults with type 2 diabetes. Check whom a study enrolled before applying its mortality findings to another group.
A 2025 analysis of 99,599 participants across randomized trials reported an all-cause mortality incidence rate ratio of 0.88 (95% CI 0.84–0.92). The authors estimated that about 121 people would need treatment over the pooled follow-up to prevent one additional death, and about 66 to prevent one major cardiovascular event. These numbers needed to treat depend on whom the trials enrolled and how long they were observed.[2]