Almost every conversation about losing weight covers fat and, if it is a good conversation, muscle. Bone is the tissue that gets left out, which is unfortunate, because it changes too and it is harder to get back.
The short answer
Losing a significant amount of weight reduces bone mineral density, and the amount you lose tends to track with the amount of weight you lose. This is not a side effect specific to any one medication. It is what a skeleton does when the job it is doing gets lighter.
Why bone responds to weight at all
Bone is not inert scaffolding. It is living tissue under constant remodelling, with cells breaking old bone down and other cells building new bone up, and the balance between those two is heavily influenced by the mechanical load running through the skeleton.
Carrying more body weight means more load through the hips, spine and legs with every step, and bone maintains itself accordingly. Take a large amount of weight off and the load falls. The skeleton does what it does with any reduced demand: it adapts down.
Your bones are built for the body you have been carrying. Change the body substantially and they will rebuild for the new one, which means losing some of what they had.
Two other things pile on. An extended stretch of negative energy balance is itself unfavourable for bone formation, and losing muscle removes some of the pull that muscle exerts on bone, which is a separate loading signal from body weight alone.
What the trials measured
The clearest recent measurement comes from a 52-week trial of once-weekly semaglutide, which found hip bone mineral density down 2.6% and lumbar spine density down 2.1% relative to placebo. Bone resorption markers rose without a compensating rise in bone formation markers, which is the pattern that explains the density change rather than merely restating it.
Work published in the Journal of Clinical Endocrinology and Metabolism on semaglutide and tirzepatide in people at increased fracture risk found significant declines at the total hip and femoral neck, and importantly found similar declines in the comparison groups. That similarity is the useful part: it points at weight loss as the driver rather than at the drug class specifically.
Within the treated groups, the amount of weight lost was directly associated with the amount of bone lost at the hip and femoral neck. Same relationship, showing up again.
Who loses more, and why it varies
The picture is not uniform, and the differences are informative. Bone loss was greater in people without diabetes than in people with it, where losses were comparable to controls. A separate study in people with obesity and type 1 diabetes found bone density maintained over twelve months on GLP-1 or dual agonist therapy.
Read across those results and the same theme keeps appearing: the magnitude of weight change, the starting bone status, and what else the person was doing matter more than which agent was involved. The honest summary is that this is a weight-loss effect that varies by person, not a fixed property of a medication.
What actually protects bone
The countermeasure follows directly from the mechanism. If bone is adapting to reduced load, then the intervention is to keep loading it.
Resistance training is the main lever, because it applies mechanical stress to bone through muscle pull and impact regardless of what you weigh. It is also the same intervention that protects muscle during weight loss, which makes it unusually efficient: one habit, covered in does strength training preserve muscle during weight loss, addressing both tissues at once.
Adequate protein is the second lever and it is easy to underrate. Protein is a structural component of the bone matrix, not just of muscle, and after appetite suppression it is the input most likely to quietly fall short. Preventing muscle loss on a GLP-1 covers the floor that also serves this purpose.
Calcium and vitamin D matter as prerequisites. Being deficient makes everything worse, and correcting a deficiency will not on its own replace the loading signal.
The rate of loss appears to matter as well, in the same direction it does for muscle. Slower tends to be gentler on both tissues.
How OffRamp helps
We chose not to put a bone health score in OffRamp, for the same reason we left out a body fat percentage. Bone density is measured by DXA. It cannot be inferred from weigh-ins and logged meals, and a number that looked like a bone assessment but was actually a guess would be worse than no number, because someone might use it to decide they were fine.
What the app does is run the two habits the evidence actually supports. A daily protein floor you log against, so the input most likely to slip quietly is visible. Short strength sessions, which are there for muscle and turn out to serve bone through the same mechanical route. Regain Radar keeps the weight trend readable, which matters here because rate of loss is one of the few things that shows up in both the muscle and bone literature.
If you want the frequency question answered, how often to strength train covers it, and the answer is a smaller commitment than most people brace for. What we will not do is tell you anything about your medication. If you have had a bone density result or you are concerned about fracture risk, that conversation belongs with your prescriber.
