BMJ fact-checks the UK headlines linking GLP-1 drugs to hundreds of deaths

The BMJ has pushed back on a wave of UK newspaper headlines that tied GLP-1 weight-loss and diabetes drugs to hundreds of deaths and thousands of adverse events. The review, prompted by an August 5 story in The Times, found that the underlying data come from the UK's Yellow Card scheme, which collects suspected-reaction reports rather than confirmed cases of drug-caused harm ^[1].

According to the BMJ's analysis, semaglutide accounted for 3,454 serious reports in women and 1,216 in men logged through the Yellow Card system ^[1]. Fatal outcomes were far smaller in number and split almost evenly by sex, with 28 deaths reported in women and 29 in men ^[1]. Most of these deaths clustered in people aged 60 to 69 ^[1].

The BMJ's point is that a Yellow Card report does not mean a drug caused a death or injury. The scheme exists to flag possible safety signals so regulators can investigate, not to serve as a confirmed tally of drug-caused harm. People who take a medicine and later have a heart attack, an infection, or another medical event can generate a report even if the drug played no role. That distinction, the BMJ said, got lost in some of the press coverage that added up raw report totals and presented them as proof of a death toll ^[1].

Separately, the piece confirmed that Genomics England and the UK's Medicines and Healthcare products Regulatory Agency have closed recruitment for a Yellow Card Biobank pilot study. The project is testing whether genetic differences can help predict who is more likely to have side effects from GLP-1 drugs. Analysis of the collected samples is now under way, though results were not yet available at the time of the BMJ's report ^[1].

Why it matters for patients

For people in the US taking or considering semaglutide (Ozempic, Wegovy, Rybelsus) or other GLP-1 drugs, this UK episode is a reminder that raw counts of adverse-event reports, whether from the UK's Yellow Card system or the US FDA's similar FAERS database, are not the same as confirmed rates of harm. These systems are built to catch early warning signs, and a report can be filed for any health event that happens after starting a drug, regardless of whether the drug caused it. Large numbers of reports can look alarming in a headline but do not by themselves establish that a medicine is dangerous or unsafe.

The BMJ's breakdown also shows that fatal reports made up a small fraction of the total serious reports for semaglutide, and that deaths were reported at similar rates in men and women, clustering in an older age group ^[1]. None of this proves the drug did or did not contribute to any individual death; it only describes what was reported and to whom.

The genetics research now underway through the Yellow Card Biobank could eventually help identify which patients face higher risk of side effects from GLP-1 medicines, but that work is still in progress and no findings have been released ^[1].

What happens next

Analysis of samples collected through the Yellow Card Biobank pilot is ongoing, according to the BMJ's report, though no date has been given for when results might be published ^[1]. The BMJ did not indicate whether UK regulators plan any further public statement responding directly to the press coverage it examined ^[1].