Adults over 50 who took a GLP-1 drug to lose weight were recorded with fewer broken bones than people who took a different weight-loss medication. In a database of 18,062 matched patients, the GLP-1 group had about an 85 percent lower rate of major osteoporotic fractures, the hip, spine, wrist, and shoulder breaks that put older people in the hospital.
That is the headline number from a retrospective study published online June 1 in Obesity ↗, and it runs straight into one of the loudest worries about this drug class. Rapid weight loss strips muscle and bone along with fat. Smaller frames break more easily. The fear is that a 50-year-old who drops 15 percent of their body weight on semaglutide is trading their waistline for a fragile skeleton a decade early.
This dataset says the opposite, with a catch worth holding onto.
What the study actually compared
The team, led by researchers at Taipei Medical University, pulled deidentified records from TriNetX, a network that aggregates electronic health records across many hospitals. They selected adults 50 and older with obesity and no diabetes, then split them by whether they were prescribed a GLP-1 receptor agonist, drugs like semaglutide ↗ (sold as Ozempic and Wegovy) and tirzepatide ↗ (Mounjaro and Zepbound), which work through the GLP-1 receptor ↗. The comparison group took other obesity medications. Then they used propensity matching, a statistical method that pairs people with similar age, sex, weight, and lab values so the two groups look alike on paper, and compared what happened to their bones and joints.
The GLP-1 group came out ahead on every outcome measured. Incident osteoporosis, the thinning of bone that precedes most fragility fractures, was 52 percent lower (relative risk 0.48). Major osteoporotic fractures were 85 percent lower (RR 0.15). Disc degeneration in the neck and lower back, and osteoarthritis, the wear-and-tear joint disease that drives most knee and hip replacements, were all lower too. Osteoarthritis carried a number needed to treat of 9, meaning that for every 9 people on a GLP-1 drug instead of another weight-loss agent, one fewer was diagnosed with it. The protective pattern held across age, sex, weight, and which specific GLP-1 drug people took.
The catch
Read the comparison again. This is not GLP-1 versus nothing. It is GLP-1 versus other weight-loss medications. So the finding is relative. It says that among people losing weight with drugs, the GLP-1 group fared better on the skeleton than people on the alternatives, not that GLP-1 drugs build bone in an absolute sense.
And it is observational, built from records of care that already happened, not a trial that randomly assigned people to one drug or the other. That matters because the people a doctor puts on a GLP-1 drug may differ from the people who end up on something else in ways no database fully captures. Healthier, more mobile, better-resourced patients tend to land on the newer, pricier drug, and those same patients break fewer bones for reasons that have nothing to do with the prescription. An 85 percent reduction in fractures is a very large effect, large enough that some of it is probably the patients, not the pharmacology.
What pushes back on pure confounding is the consistency. The benefit showed up across every subgroup and survived the authors' sensitivity checks, and it points in the same direction as the mechanistic story now accumulating around these drugs, where weight loss offloads arthritic joints and GLP-1 signaling appears to have direct effects on bone turnover.
It also sits in tension with a different recent finding. A separate study had semaglutide stripping muscle along with fat ↗, the lean-mass loss that worries clinicians most. Muscle and bone are not the same tissue, and a drug can spare one while thinning the other. The honest read of both papers together is that the body-composition story for GLP-1 drugs is still being written, and the skeleton may be getting a better deal than the muscle.
For now, the practical takeaway is narrow and useful. The bone-fragility fear that follows GLP-1 drugs around does not show up in this large a sample. The number that would settle it, a randomized trial tracking fractures over years, does not exist yet.