Study wrapper · #766
GLP-1 Receptor Agonists for Weight Loss and Risk of Major Safety Outcomes: A Multicentre Cohort Study.
Editor's note
This multicentre cohort (13 South Korean hospitals, 2018-2025, OMOP-mapped) compared weight-management semaglutide or liraglutide initiators against propensity-matched non-initiators. Researchers reported that semaglutide initiation was associated with higher risks across several outcomes: psychiatric disorders overall (HR 2.02), anxiety (2.39), depressive disorder (3.42), gastrointestinal dysmotility or obstruction (3.91) and vision impairment (1.58). This is observational, not a trial: propensity matching cannot rule out residual confounding, and differences between people who start these agents and those who do not can inflate associations. Some signals (semaglutide-linked vision impairment) were not consistent across sensitivity analyses. Read this as a monitoring and patient-selection signal that adds to the safety conversation around GLP-1 receptor agonists, not as evidence of causation. Randomised trial data remain the stronger reference for weighing benefit against these harms.
Plain-language abstract
Researchers used electronic health records from 13 hospitals in South Korea to study people who started semaglutide or liraglutide for weight management. Each person who started the drug was matched to similar people who did not, so the groups could be compared more fairly. Over follow-up, people who started semaglutide had higher recorded rates of several problems, including anxiety, depression, gut-movement problems or blockage, and vision issues. Liraglutide was linked to higher rates of psychiatric, liver, gallbladder or pancreas, and vision problems. Because this is a real-world observational study rather than a controlled trial, it can show that these events happened more often in people on the drugs, but it cannot prove the drugs caused them; other differences between the groups may play a role. Some links weakened in extra analyses. The authors stress careful patient selection and monitoring.