Semaglutide tied to 15 percent fewer fractures in type 2 diabetes study

Semaglutide was associated with 15 percent fewer bone fractures than other weight-loss and diabetes medications in adults with type 2 diabetes, according to a Stanford University-led analysis presented June 14, 2026 at ENDO 2026, the Endocrine Society's annual meeting in Chicago ^[1][2]. The semaglutide group recorded 794 fractures, compared with 1,045 in the comparison group ^[1][2].

The work was a retrospective cohort analysis, meaning researchers looked back at medical records rather than assigning people to treatments ^[1][2]. It drew on the Atropos Health Eos electronic health record dataset, which represents 161 million patients seen at U.S. community hospitals and academic medical centers, with data from January 2016 through December 2023 ^[1]. Participants were adults 18 and older diagnosed with type 2 diabetes who had no history of fractures or use of osteoporosis medications ^[1].

Of the 59,879 people in the analysis, 26,324 took semaglutide and 33,555 took dulaglutide, phentermine/topiramate, or bupropion/naltrexone ^[1][2]. The researchers then matched 17,506 semaglutide users with 17,506 patients from the comparison group using a 1:1 high-dimensional propensity score method intended to balance differences between the groups ^[2]. In the matched groups, 55.4 percent and 53.9 percent were women, mean ages were 54.1 and 54.5 years, and mean body mass index was 38.6 and 37.8 ^[2].

Over a mean follow-up of 1,327 days — about three and a half years — semaglutide was linked to a lower risk of fracture, with a hazard ratio of 0.85 (95 percent confidence interval, 0.77 to 0.93; P < .001) ^[2]. Semaglutide users also lost more weight. In a matched body mass index analysis of 4,191 pairs, their BMI fell by an average of 1.9 points versus 1.2 points in the comparison group, and the difference in BMI at one year was -0.72 (95 percent confidence interval, -0.87 to -0.57; P < .001) ^[2].

"Bone fractures are painful, expensive and can seriously affect quality of life — especially as people get older," said Jairo Noreña, M.D., a former endocrinology fellow at Stanford University Medical Center and an author of the study. "We hope this study encourages monitoring of bone health in weight-loss programs" ^[1]. Noreña also called the work "an important early step toward understanding the impact of semaglutide-induced weight loss on bone health in patients with type 2 diabetes" ^[1].

The authors recommend that prospective studies be done to confirm the findings ^[1][2]. The results were presented as a conference abstract (ORF34-04) and have not been reported as published in a peer-reviewed journal ^[2]. One study author disclosed affiliations with biotech, pharmaceutical or device companies ^[2]. Several details were not reported, including which bones were fractured, what doses patients took, and whether bone density was measured. The analysis did not include tirzepatide as a comparison treatment, and it covered only people with type 2 diabetes, so it is not yet known whether the pattern would appear in other groups.

Why it matters for patients

Rapid weight loss can thin bone, and previous research has raised the question of whether GLP-1 medicines add to fracture risk as people lose weight ^[1]. This analysis points in the opposite direction for semaglutide, showing fewer fractures alongside greater weight loss than the comparison treatments ^[1][2]. That is a notable signal because fractures in older adults can bring pain, expense and lasting effects on daily life, as Noreña noted ^[1].

At the same time, the study design cannot show that semaglutide prevented fractures. People were not randomly assigned to treatments, so unmeasured differences between the groups could explain part of the result ^[1][2]. The average participant was in their mid-50s, not the age range where fractures are most common, and everyone in the study had type 2 diabetes and no prior fractures or osteoporosis drugs ^[1][2]. The evidence for any bone effect in other patients, or for other GLP-1 medicines, has not been reported from this dataset.

What happens next

The findings were presented during ENDO 2026, which runs June 13 to 16, 2026 in Chicago ^[2]. The authors' stated next step is prospective research — studies that follow patients forward in time — to confirm whether the link holds up ^[1][2]. No such study has been announced in these results, and no timeline for publication in a peer-reviewed journal has been disclosed.