The Daily Inference
Science & Space · News

Weight-Loss Users Saw No Extra Risk, but Diabetes Users Had Higher Premature Birth Odds

The study covered more than 750,000 pregnancies, but just 529 involved possible exposure to semaglutide or liraglutide. Higher odds of premature birth appeared only among women prescribed the medicines for diabetes.

A Danish study found no statistically significant rise in the pregnancy complications examined after early exposure to semaglutide or liraglutide for weight management, offering reassurance about an increasingly pressing medical question. [1]

The higher odds of premature birth appeared only among women prescribed the medicines for pre-existing diabetes. Researchers said diabetes itself, or related differences between patients, could explain that association. [1]

Semaglutide, sold as Ozempic and Wegovy, and liraglutide, sold as Saxenda, belong to a group of medicines called GLP-1 receptor agonists used to treat type 2 diabetes and manage weight. Exposure can happen before someone knows they are pregnant, leaving patients and clinicians to make decisions with limited evidence. [1]

The findings make the reason for taking the medicine central to that discussion. Combining diabetes treatment and weight management into a single category would obscure the study's most useful result: the association with premature birth differed between those groups. [1]

Henriette Svarre Nielsen, who led the research, said the results were not enough to change current advice to stop these medicines eight weeks before a planned pregnancy. More research is needed before new counseling guidance can be established, she said. [1]

A large study with a small exposed group

Published in Human Reproduction Open, the study examined 756,636 pregnancies among 480,231 women in Denmark, with deliveries between October 1, 2009, and December 31, 2023. All were singleton pregnancies, meaning the woman was carrying one baby rather than twins or other multiples. [1]

Within that population, researchers identified 529 pregnancies with possible exposure to either medicine around the start of pregnancy. That is fewer than one in 1,000 pregnancies in the study, a useful measure of how much smaller the group of interest was than the headline total. [1]

The researchers counted a pregnancy as exposed if the woman redeemed a prescription during the eight weeks before or after the first day of her last menstrual period. This created a 16-week window encompassing time before pregnancy and its early stages. [1]

They examined premature birth, pre-eclampsia, gestational diabetes, babies large for their gestational age, stillbirth and placental problems. Pre-eclampsia is a potentially serious pregnancy condition involving high blood pressure, while gestational diabetes develops during pregnancy. [1]

To make the comparisons more informative, the researchers adjusted for differences that could affect pregnancy outcomes, including age, body mass index, smoking, education, region and pre-existing diabetes. After those adjustments, the only remaining statistically significant increase reported was in premature birth among women prescribed the drugs for diabetes. [1]

The study was observational, so it cannot establish that the medicines caused an outcome. Prescription redemption does not confirm that someone took a medicine, and the findings do not prove safety throughout pregnancy or address every possible outcome. The number of exposed pregnancies also leaves a need for larger studies, particularly for semaglutide, which was prescribed for weight loss mainly late in the study period. [1]

Why the diabetes finding matters

Among women prescribed liraglutide for diabetes, the adjusted odds of premature birth were 70% higher than among women not exposed to GLP-1 medicines. For those prescribed semaglutide for diabetes, the adjusted odds were 84% higher. [1]

Those figures describe relative differences in odds. They do not mean that 70% or 84% of the exposed pregnancies ended prematurely, or that the chance increased by that many percentage points.

For weight management, the researchers found no statistically significant association with premature birth. After adjustment, they also found no remaining statistically significant association with the other complications examined. [1]

Kathrine Hviid, the study's first author, said the premature-birth association appeared when the medicines were used to treat diabetes, rather than weight management. Diabetes itself might be driving the finding, she said. [1]

That is the central problem in interpreting the numbers. People receive a medicine because they have a condition that needs treatment. When that condition can also affect pregnancy, an association between the prescription and an outcome may reflect the illness, the treatment or other differences between patients.

Separating patients by the reason for their prescription helps investigate that problem. Here, it points researchers toward underlying metabolic health as an explanation worth testing, rather than treating every early exposure as equivalent. [1]

Reassurance has a practical use

For someone who discovers a pregnancy after taking a weight-management medicine, evidence about inadvertent early exposure answers a different question from whether to continue treatment deliberately. The Danish study is most useful for the first conversation.

In an invited commentary, researchers Yeyi Zhu and Monique Hedderson said the findings support more individualized discussions after inadvertent exposure. They also emphasized metabolic health and glucose management for patients with diabetes. [1]

That approach gives the study a practical role without asking it to settle the whole pregnancy-safety question. A patient's reason for treatment belongs near the start of the discussion, alongside the timing of possible exposure.

Nielsen said the existing recommendation to stop treatment eight weeks before planning pregnancy rests on early-stage studies in model organisms rather than real-world evidence. The Danish registry adds human pregnancy data to that evidence base, while leaving the recommendation in place. [1]

The next research task is to build larger exposed groups and separate patients by why they take the medicines. Larger studies with longer follow-up, particularly for semaglutide, could help establish whether the diabetes-associated signal persists and what explains it.

Topics: Medical research · Ozempic

Every edition in brief, three times a day, on our Telegram channel, on Bluesky and on Threads.

Sources
  1. Increased risk of premature births is linked to use of weight loss drugs during early pregnancy, but only when they are used to treat diabetes - Focus on Reproduction - ESHRE focusonreproduction.eu