JAMA letter links weight-loss use of semaglutide and liraglutide to three gastrointestinal risks
The findings come from insurance claims rather than a randomized trial, so they show an association rather than proof of cause, and the analysis did not include tirzepatide.
A research letter published in JAMA reports that adults prescribed semaglutide (sold as Ozempic, Wegovy and Rybelsus) or liraglutide for weight loss were diagnosed with three gastrointestinal conditions more often than adults taking bupropion-naltrexone, a weight-loss drug from an unrelated class [1].
The analysis drew on a random sample of 16 million patients from the PharMetrics Plus for Academics database (IQVIA), a US claims database that the authors say captures 93% of outpatient prescriptions and physician diagnoses in the country, covering 2006 through 2020 [1]. The researchers compared new users of semaglutide (613 people) and liraglutide (4,144 people) with new users of bupropion-naltrexone (654 people). Everyone in the three groups had an obesity diagnosis code, and people with a diabetes code or a diabetes drug code were excluded [1].
Participants were followed until the first diagnosis of biliary disease, pancreatitis, bowel obstruction or gastroparesis, until they switched drugs, or until the study period ended in June 2020 [1]. After adjustment for age, sex, alcohol use and other factors, the hazard ratios compared with bupropion-naltrexone were 3.67 for gastroparesis (95% CI, 1.15-11.90), 9.09 for pancreatitis (95% CI, 1.25-66.00) and 4.22 for bowel obstruction (95% CI, 1.02-17.40). For biliary disease the adjusted hazard ratio was 1.50 (95% CI, 0.89-2.53), which the letter treats as not significant because the confidence interval crosses 1 [1].
The event counts behind those ratios were small. The letter reports 4 gastroparesis diagnoses among 613 semaglutide users, 66 among 4,144 liraglutide users and 3 among 654 bupropion-naltrexone users. Pancreatitis was recorded for 2 semaglutide users, 71 liraglutide users and 1 bupropion-naltrexone user; bowel obstruction for 0, 73 and 2; and biliary disease for 5, 162 and 16 [1].
Because those counts were low, the confidence intervals are wide, especially for pancreatitis, where the range runs from 1.25 to 66.00 [1]. The authors also reported E-values, a measure of how strong an unmeasured risk factor would have to be to explain away the finding, of 6.80 for gastroparesis, 17.67 for pancreatitis and 7.91 for bowel obstruction [1]. A sensitivity analysis using a less restrictive definition of obesity produced lower estimates: 2.35 (95% CI, 1.20-4.58) for gastroparesis, 5.94 (1.90-18.60) for pancreatitis and 2.44 (1.00-5.95) for bowel obstruction [1].
The groups also differed at the start. Semaglutide users had a median follow-up of 0.6 years, compared with 1.7 years for both liraglutide and bupropion-naltrexone users. Their mean age was 53.5 years, versus 51.3 for liraglutide users and 45.2 for bupropion-naltrexone users, and 55.8% of semaglutide users were male, versus 82.4% of bupropion-naltrexone users [1]. The letter notes that semaglutide was not marketed for weight loss until 2021, after the study period, so its semaglutide users were taking the drug off label for that purpose [1]. JAMA has also published comment letters about the findings and a reply from the authors [1].
Why it matters for patients
The results come from insurance claims, not a randomized trial, so they can show an association but cannot prove that the drugs caused these diagnoses [1]. The absolute numbers are small: 4 cases of gastroparesis among 613 semaglutide users, and 2 cases of pancreatitis [1]. The wide confidence intervals mean the true size of any increase is uncertain, and the letter does not report an individual patient's risk or a risk difference per person [1].
The findings cover only semaglutide and liraglutide. Tirzepatide, sold as Mounjaro and Zepbound, was not part of the analysis [1]. The typical follow-up in the study was about seven months for semaglutide users, so the data do not describe risks over many years of use [1].
What is not yet known
Whether these drugs cause gastroparesis, pancreatitis or bowel obstruction has not been established by this analysis. Whether the risks change with dose or duration of use, whether they persist beyond the follow-up periods in this study, and whether the same pattern applies to other GLP-1 medications have not been reported [1].
About 16 million people die each year from cardiovascular disease, according to the World Health Organization, making it the leading cause of death globally.
Sources
Semaglutides.org provides information only, not medical advice. Talk to a licensed provider about your own care before making changes. Some telehealth links are affiliate links and are labeled where they appear.