Study finds faster hip bone loss on semaglutide and tirzepatide in people without diabetes

A single-center study of people at increased risk of fractures found that those taking semaglutide or tirzepatide lost bone density at the hip faster than similar people who were not taking the drugs, but only among participants who did not have diabetes ^[1]. A separate, much larger study of older adults with type 2 diabetes found the opposite direction on hard outcomes: fewer falls and fewer hip fractures among people on the same two drugs ^[2].

Semaglutide is sold as Ozempic, Wegovy and Rybelsus. Tirzepatide is sold as Mounjaro and Zepbound.

The bone density findings

The first study, published in the Journal of Clinical Endocrinology & Metabolism, was a single-center retrospective analysis of patients with increased risk of fractures. It matched 255 people taking semaglutide or tirzepatide with 255 non-users and compared annualized change in total hip bone mineral density ^[1].

Among users without diabetes, annualized hip bone loss averaged 1 percent, compared with 0.6 percent in non-users. Among participants with diabetes, the two groups did not differ ^[1]. The amount of bone lost tracked the amount of weight lost ^[1].

Bone mineral density is one input into fracture risk, not the same thing as a fracture. Whether the difference seen in this study leads to more broken bones has not been established.

The fracture findings

The second study, in Osteoporosis International, drew on the TriNetX electronic health record database covering 2018 to 2025. It included adults 65 and older with type 2 diabetes and a body mass index of at least 25, and compared patients prescribed semaglutide or tirzepatide with patients prescribed DPP-4 inhibitors, a different class of diabetes drug. Patients were matched 1:1 on their characteristics ^[2].

After matching, the semaglutide comparison included 27,896 patients and the tirzepatide comparison 12,808. Femoral fractures occurred in 0.3 percent of semaglutide patients versus 0.5 percent of DPP-4 patients (hazard ratio 0.488; 95% confidence interval 0.367–0.649), and in 0.2 percent versus 0.4 percent for tirzepatide (HR 0.452; 95% CI 0.280–0.729). Falls occurred in 3.6 percent versus 5.4 percent for semaglutide (HR 0.663; 95% CI 0.612–0.718) and 3.6 percent versus 5.7 percent for tirzepatide (HR 0.664; 95% CI 0.591–0.747). The authors reported that the fracture difference was more pronounced in patients with a BMI of 30 or higher ^[2].

The two studies measured different things in different groups. One tracked bone density scans in people at elevated fracture risk; the other counted falls and fractures in the medical records of older adults with type 2 diabetes and overweight or obesity. Both are observational, so neither can show that the drugs caused the outcomes. The Osteoporosis International authors noted that weight loss from these drugs can reduce mechanical loading on bone and come with loss of lean mass, which could increase frailty and fall risk in older adults ^[2].

Why it matters for patients

Hip fractures and falls are among the most serious events in older adults. The Osteoporosis International paper notes that these events are linked to higher one-year mortality, long-term disability, loss of independent living and more institutionalized care ^[2].

The bone density difference in the first study was modest in absolute terms — four-tenths of a percentage point per year — and appeared only in people without diabetes ^[1]. Whether that gap widens over years of treatment, whether it reverses after stopping a drug, and whether it matters for people who already have osteoporosis are questions the study does not answer.

On the other side, the fracture and fall reductions in the larger study were seen in people with diabetes, where the bone density study found no difference between users and non-users ^[1][2]. That study used DPP-4 inhibitors as the comparison rather than placebo, and it followed patients for one year ^[2]. Whether the lower fracture rate holds in people without diabetes has not been reported.

Neither study measured muscle mass, strength or frailty in the groups it analyzed, and neither reported on calcium, vitamin D or exercise, so how those factors interact with these findings is not yet known.

What happens next

No follow-up analyses or trial timelines were disclosed with either report. The Osteoporosis International authors wrote that evidence on skeletal safety in adults 65 and older with obesity remains limited because clinical trials have largely excluded frail or elderly participants ^[2]. No randomized trial results focused on bone density or fracture outcomes in this population have been reported in these two papers.