Guides
How large was the cardiovascular difference in SELECT?
Short answer
In SELECT, 569 of 8,803 people assigned semaglutide and 701 of 8,801 assigned placebo experienced at least one of the specified cardiovascular events. That was 6.5% and 8.0% over an average follow-up of 39.8 months. The rounded proportions differ by about 15 people per 1,000.
Which event did the researchers count?
The primary outcome combined three events. The researchers tracked the time to a first nonfatal heart attack, nonfatal stroke or death from a cardiovascular cause. Someone who had a heart attack and later a stroke entered this comparison when the first event occurred. The often-quoted hazard ratio of 0.80 comes from analysing the timing of these events. To answer “How many people experienced one?”, look at the group counts above. The difference of 15 per 1,000 describes the combined outcome. Those 15 cannot all be attributed to any one of the three events.
Who does this comparison apply to?
Participants were at least 45, had a BMI of 27 or above and already had cardiovascular disease, such as a previous heart attack. They had no prior diagnosis of diabetes. Both groups received standard care. SELECT therefore addresses prevention of further events. Its findings do not establish how many first events could be prevented in people with no known cardiovascular disease.
Primary sources
- Lincoff et al., SELECT, primary publication and results, 2023
- SELECT full paper, Table 2 and Figure 1A on the composite outcome
- SELECT trial registry, eligibility and primary outcome
Cite this page
Fastzuleicht · Editorial team. How large was the cardiovascular difference in SELECT? (18 November 2026). https://fastzuleicht.de/guides/select-herzinfarkt-schlaganfall/