Your GLP-1 Patients Are Moving Less, Not More — Here's What the 2026 Data Means for Your Plan of Care
If you treat any adult orthopedic or general ortho caseload right now, you already have patients on semaglutide, tirzepatide, or liraglutide. Probably several. And there's a decent chance you've noticed something that doesn't quite match the marketing: the scale's moving in the right direction, but strength, endurance, or general "get-up-and-go" isn't following along the way you'd expect.
Turns out that's not just a clinic-floor impression. It showed up in actual data this year, and it's worth fifteen minutes of your evidence-based attention before your next eval on one of these patients.
The new finding that should change how you ask intake questions
At the Endocrine Society's ENDO 2026 meeting this June, researchers pulled Fitbit and electronic health record data on adults with obesity through the NIH's All of Us program — 753 people with enough wearable data to actually analyze, all of whom had started a GLP-1 medication. Average daily steps dropped from 5,047 to 4,487. Moderate-to-vigorous activity minutes dropped from 27.9 to 22.2 a day. Both changes were statistically significant.
Here's the part that actually matters for your plan of care: the decline wasn't even across the group. Men lost more activity than women. And patients with musculoskeletal pain at baseline lost dramatically more — 679 fewer steps a day, versus 22 fewer steps a day in people without MSK pain. That's not a rounding difference. That's your exact caseload showing up as the highest-risk subgroup in the data.
(One honest caveat: this was a conference presentation, not yet a peer-reviewed published paper, so treat the specific numbers as preliminary rather than settled science. But it lines up with what a lot of us are already seeing in clinic, and it's consistent with the next section.)
What's actually being lost when the weight comes off
The reason activity decline matters so much on these medications is that GLP-1-driven weight loss isn't fat-only. A 2025 systematic review and meta-analysis in Obesity Reviews (Anyiam et al.) looked specifically at muscle mass outcomes across GLP-1 receptor agonist trials and confirmed measurable reductions in muscle mass measures in people on these drugs, with and without type 2 diabetes.
A separate network meta-analysis — Karakasis et al., published in Metabolism in early 2025, pooling 22 RCTs and over 2,200 patients — put a number on it: roughly 25% of total weight lost on a GLP-1RA is lean mass (muscle, water, glycogen), not fat. Tirzepatide 15mg and semaglutide 2.4mg, the two highest-dose, most-prescribed regimens right now, produced the most fat loss but were also the least protective of lean mass.
The reassuring nuance, from a smaller but well-designed 2025 study (the SEMALEAN trial, Diabetes, Obesity and Metabolism): lean mass loss on semaglutide was front-loaded in the first 7 months and then stabilized, and by 12 months, handgrip strength had actually improved and sarcopenic obesity prevalence dropped from 49% to 33% of the study group. Translation: the lean-mass hit is real, but it's not a one-way ratchet — what happens after the initial loss looks like it depends heavily on what the patient is doing with their body in that window.
Okay, but what about bone?
This is where the headlines get ahead of the evidence, so let's be precise. AAOS's 2026 annual meeting press materials reported a five-year risk increase for osteoporosis, gout, and osteomalacia among GLP-1 users compared to non-users — osteomalacia showed the largest relative jump (0.2% vs. 0.1% incidence, a risk ratio of 2.55). That's a real, registry-level signal worth knowing about, but it's also a conference report, not yet a peer-reviewed paper, so hold it a little loosely.
A more detailed 2025 narrative review in Osteoporosis International adds context: most systematic reviews and meta-analyses to date have found no significant difference in fracture risk between GLP-1RA users and non-users, and one large meta-analysis (38 studies, ~39,800 patients) actually found a decreased fracture risk after 52+ weeks of treatment — likely tied to less mechanical loading-related stress over time as patients lose weight, plus possibly direct bone-protective signaling from the drugs themselves. So: bone density can drop on these meds, but that hasn't reliably translated into more fractures across the literature. Both things are true at once, which is exactly the kind of nuance worth bringing to a patient conversation instead of a scary headline.
The actionable finding — and it's a good one
Here's the study that should actually change what you tell patients. A 2024 secondary analysis of a randomized controlled trial, published in JAMA Network Open (Jensen et al.), compared bone mineral density outcomes in adults with obesity across three groups over 52 weeks: exercise alone, liraglutide alone, and exercise plus liraglutide combined.
Liraglutide alone reduced hip and spine BMD compared to exercise alone, despite similar weight loss. The combination group — exercise plus the medication — lost just as much weight as the medication-alone group, but fully preserved hip, spine, and forearm BMD. Exercise didn't just help; it appears to have completely offset the bone-density cost of the drug.
That's about as clean a clinical message as you're going to get out of this literature: GLP-1 medications plus structured exercise look meaningfully different — for bone, and likely for lean mass — than GLP-1 medications alone. Which is, conveniently, exactly the gap PTs and OTs are positioned to close.
What this actually means for your plan of care
A June 2025 Physical Therapy journal article ("Transforming Care: Implications of Glucagon-Like Peptide-1 Receptor Agonists on Physical Therapist Practice," Mulcahy, DeLaRosby, and Norwood) lays out the practice-level version of all of this, and it tracks closely with what the data above actually supports:
Ask about GLP-1 use at intake, every time. Not just for diabetes management — for weight management too, and increasingly for patients who don't think of themselves as "on a diet," because plenty of them aren't framing it that way to themselves.
Treat declining activity tolerance as a red flag, not a personality trait. If a patient on a GLP-1 is moving less and you assume it's motivation, you're missing the more likely explanation: appetite suppression plus rapid weight loss plus possible early sarcopenia can genuinely sap functional capacity, especially in the first several months.
Prioritize resistance training specifically, not just "more activity." The mechanism that's protective in the literature above is loading — bone and muscle both respond to mechanical stress, and walking more steps is good but isn't the same stimulus as progressive resistance work.
Loop in protein intake as part of your patient education.
The PTJ piece flags resistance training plus adequate protein intake as the combination that best preserves lean mass during a caloric deficit — worth a conversation, a referral to a dietitian, or both.
Don't assume the bone story is settled, and say so out loud. Telling a patient "this might increase your fracture risk" isn't supported by most of the literature; telling them "we want to mitigate any risks associated with your current medications and build in some impact loading" is the honest, current version
Bottom line
The patients on GLP-1s in your schedule right now are losing weight the medication is designed to take off — and, the data increasingly suggests, also losing some lean mass and bone density the medication isn't designed to protect. The good news, and it's genuinely good news: the one RCT that directly tested it found that structured exercise alongside the medication closes that gap almost completely. That's not a reason to be alarmist about GLP-1s. It's a reason PT and OT need to be part of the conversation every time one gets prescribed.