One nagging concern about GLP-1 weight-loss drugs is that losing weight fast can also mean losing bone and muscle — potentially raising fracture risk. A large new analysis suggests the opposite may be true for semaglutide.

Presented at ENDO 2026, the Endocrine Society’s annual meeting, the study found that in people with type 2 diabetes, semaglutide was associated with about a 15% reduction in bone fractures compared with other anti-obesity medications — alongside greater weight loss.

Why the concern existed

The worry was grounded in established physiology rather than speculation.

Bone adapts to the load placed on it, so weight loss reduces the mechanical stimulus maintaining bone density — which is why bone loss accompanies weight loss by any means, including diet and bariatric surgery.

Rapid weight loss also risks inadequate protein and calcium intake, and any loss includes some lean mass alongside fat. Muscle both loads bone directly and contributes to the balance and strength that prevent falls, so losing it raises fracture risk through a second route.

Fracture risk depends on falling as well as on bone strength, and that division matters for interpreting the result.

How big the study was

The population-based study drew on the Atropos Health Eos electronic-health-record dataset representing roughly 161 million patients across US hospitals and academic centres (2016–2023). It focused on adults with type 2 diabetes who had no prior fractures and were not taking osteoporosis medications.

Those exclusions are appropriate. Prior fracture is the strongest predictor of future fracture, and osteoporosis treatment identifies people already recognised as high-risk — including either group would confound the comparison substantially.

Why diabetes complicates bone

The population studied has a specific and counterintuitive bone problem worth stating.

People with type 2 diabetes have higher fracture risk despite typically having normal or above-normal bone mineral density. The bone is present in adequate quantity and is qualitatively poorer — changes in collagen structure associated with chronic high glucose make it more brittle.

Standard density scanning therefore underestimates their risk, which is a recognised clinical problem. Diabetes complications contribute independently too: neuropathy impairs balance and sensation, and retinopathy impairs vision, both raising fall risk.

How the drug might protect

The finding decouples the traditional worry and points to a possible bone-protective effect.

Several routes are plausible without any being established. Improved glucose control would reduce the glycation damage that degrades bone quality in diabetes. GLP-1 receptors are present on bone cells, raising the possibility of direct effects on bone turnover.

Reduced falling is the explanation that requires least assumption. Less weight means less load in a fall, better mobility and less mechanical strain on joints — and fewer or less forceful falls reduce fractures without any change in bone at all.

The study measures fractures, not bone density, so it cannot distinguish stronger bones from fewer falls — and the practical benefit is the same either way.

What the comparison actually was

Semaglutide was compared with other anti-obesity medications rather than with no treatment, which shapes the claim.

It establishes relative advantage within a treated population, not that semaglutide reduces fracture risk against baseline. Comparators would have included agents with different metabolic effects, and some may increase risk rather than semaglutide reducing it.

The caveats

Two matter: this was presented at a conference and is not yet peer-reviewed, and it is observational.

Confounding by indication is the standing concern with drug comparisons in records data. Patients prescribed one drug rather than another differ in ways clinicians perceived and databases do not capture — and if semaglutide was preferentially given to healthier or more mobile patients, that alone would produce fewer fractures.

What is being done about muscle loss

The bone question sits alongside a related concern that has attracted more development activity: lean mass lost during rapid weight reduction.

Any substantial weight loss includes muscle as well as fat, typically a quarter or more of the total. That proportion is not unique to these drugs — it occurs with dieting and with bariatric surgery too — but the magnitude and speed of loss on GLP-1 therapy makes the absolute amount larger.

The concern is greatest in older patients, who have less muscle to begin with and for whom sarcopenia predicts falls, disability and mortality independently of weight.

Several companies are developing agents intended to preserve lean mass alongside GLP-1 therapy, targeting pathways such as myostatin and activin that normally restrain muscle growth. The premise is that combining them would let weight loss come disproportionately from fat.

Whether that is necessary for most patients is unsettled. Losing muscle in proportion to a smaller body is partly appropriate, and resistance exercise with adequate protein addresses much of it without an additional drug — which is the comparison those programmes will need to beat.

The authors recommend prospective studies to confirm it. General information, not medical advice.