What the randomized SELECT trial found
SELECT randomized 17,604 adults aged 45 or older to weekly semaglutide or placebo. They had established cardiovascular disease and overweight or obesity, without diabetes. The primary question was about major cardiovascular events. In a secondary analysis, COVID-19 occurred at similar rates in the two groups, but fewer participants in the semaglutide group died of COVID-19 (43 versus 65; hazard ratio 0.66, 95% CI 0.44 to 0.96) [1].
What can records tell us about flu and COVID-19?
A separate medical records study compared people who were already taking semaglutide after a documented infection with matched metformin users who had no prior GLP-1 treatment. There were 7,092 people per treatment group for COVID-19 and 1,920 per group for flu. Thirty-day all-cause mortality was 0.3% versus 0.8% after COVID-19 (risk ratio 0.39) and 0.2% versus 0.7% after flu (risk ratio 0.29, 95% CI 0.09 to 0.87) [2].
In each comparison, the difference was about five deaths per 1,000 people with a documented infection, based on the rounded rates. The study matched people on factors including vaccination history and antiviral use. Differences the researchers could not measure, such as frailty or how closely people followed treatment, could still affect the results.
What the studies leave open
Studies of GLP-1 drugs and markers of inflammation examine a possible explanation for better outcomes. The COVID-19 and influenza studies do not establish which biological changes, if any, account for the observed differences.
Both studies looked at people assigned or treated before infection, so they do not show whether starting semaglutide after getting sick helps.