Most of the GLP-1 conversation is about the number on the scale. A study presented at ENDO 2026, the Endocrine Society’s annual meeting, looked at a number nobody markets: how much people actually moved after starting the medication. The answer was less, not more.
What the study did
A team led by Sajana Maharjan, M.D., of HSHS St. John’s Hospital in Springfield, Illinois, used the NIH’s All of Us Research Program to link electronic health records with Fitbit data. Out of 1,950 adults with obesity who started a GLP-1 receptor agonist, 753 had enough wearable-device data from before and after starting treatment to analyze — a retrospective pre-post cohort design, not a controlled trial.
The cohort skewed heavily female (78.6%) with a mean age of 52.7.
What it found
After starting a GLP-1, average daily steps fell from 5,047 to 4,487. Time spent in moderate-to-vigorous physical activity dropped from 28 minutes a day to 22. The declines were largest in men and in people who reported joint or muscle pain. Age, heart failure, and prior stroke didn’t change the pattern.
The finding researchers flagged as most notable: there was no evidence that losing weight on these medications led people to become more active. If anything, the relationship ran the other way.
What it doesn’t establish
This is a retrospective, observational analysis, not a trial designed to test why activity changed. A few limits matter before drawing conclusions:
It can’t isolate a cause. The data shows steps went down after a prescription started. It does not show whether that’s from feeling less hungry and less motivated to move, from joint pain easing enough to reduce compensatory movement patterns, from fatigue, from the original reason someone had low activity to begin with, or from something else entirely. The researchers didn’t test a mechanism — they measured an association.
The sample has a real skew. Only 753 of 1,950 people had usable wearable data, and the group that did was almost 80% female with an average age in the early 50s. Whether the pattern holds the same way in men, younger patients, or people who never wore a fitness tracker in the first place isn’t something this dataset can answer.
A step count isn’t a full picture of health. Fewer steps on a Fitbit doesn’t by itself mean someone is worse off — a person in real pain doing less walking might still be improving overall. But it’s also not evidence of the opposite, and the researchers were explicit that they found nothing to support the assumption that weight loss automatically translates into a more active lifestyle.
Why it’s worth taking seriously anyway
The concern researchers raised isn’t really about steps — it’s about muscle. GLP-1 medications don’t only burn fat; a meaningful share of the weight lost on these drugs is lean mass, and lean mass is exactly what physical activity, especially resistance work, helps preserve. A person who is eating less and moving less at the same time is set up to lose more muscle than someone doing the same weight loss while staying active. Dr. Maharjan’s framing was direct: “exercise cannot be optional for people taking these medications.”
That’s a different claim from a drug being unsafe, and we’re not making the second one. It’s a claim about what the medication does and doesn’t do on its own. A GLP-1 changes appetite and, for some people, changes how food and movement feel. It doesn’t install a walking habit or a strength-training routine — that part is still on the patient, and apparently isn’t happening automatically for a lot of people.
What this has to do with compounded medications specifically
This study looked at GLP-1 receptor agonists as a drug class, drawn from real-world prescriptions rather than a single manufacturer’s trial population, so the activity finding isn’t tied to one branded product. But it’s worth repeating something we say in nearly every post here: efficacy and side-effect data from clinical trials is generated using FDA-approved, manufactured medications. Compounded semaglutide and tirzepatide, prepared by state-licensed pharmacies for an individual patient, are not FDA-approved products, and no trial — including this one — was designed to tell you how a compounded preparation specifically performs.
What this study does apply to, regardless of which version of the drug someone is on, is the plain mechanical fact that appetite suppression by itself doesn’t build or preserve muscle. That part isn’t pharmacology-specific.
The practical takeaway
If you’re on a GLP-1, or considering starting one, this study is a reasonable prompt to check in on your own activity level rather than assume the medication is handling everything. Nobody needs a fitness tracker to notice whether they’re moving less than before — but if you have one, it’s worth actually looking at the trend instead of just the scale.
We don’t sell a mandatory coaching package alongside a prescription, and we’re not going to pretend this post is one either. But a licensed provider reviewing your case can talk through what a reasonable activity target looks like for your situation, and that conversation is worth having before six months go by and the number on the scale is the only number anyone checked.
This post is general information, not medical advice. It does not describe a benefit or guarantee of any NoBsRx program. Compounded medications are not FDA-approved. Treatment is never guaranteed — eligibility, diagnosis, and every treatment decision are made by independent licensed providers based on an individual clinical assessment, and availability varies by state. Talk to a licensed clinician before starting, stopping, or changing any treatment or exercise routine.