For the first year I prescribed semaglutide, I never once ordered a body composition scan. I checked weight. I checked A1c. I told patients the number was moving the right direction, and I meant it. I didn’t check what kind of weight was actually leaving the body.
That gap isn’t some secret I stumbled onto on my own. Ask any endocrinologist and you’ll hear close to the same thing. The body composition sub-studies from the STEP trials found that close to a third of the weight people lost on semaglutide was lean mass, not fat. Nobody built a marketing campaign around that number. “You may be losing muscle along with the fat” doesn’t sell the way “lose fifteen percent of your body weight in a year” sells.
The wellness side took GLP-1s and turned them into a shortcut. Something you inject once a week while your life stays exactly the same around it. Mainstream medicine mostly watched the scale move and called the visit a win. Neither side was asking what was actually happening inside the specific person taking the drug. That question is the only one that has ever mattered to me in an exam room.
I was wrong about these medications for a while, and I’ll say that plainly. When semaglutide first showed up in my clinic, I treated it like the next diet drug in a long line of diet drugs. I’d watched phentermine come and go. I’d watched the metabolism boosters that emptied a wallet and did nothing else. So I assumed this was more of the same, dressed up in a better delivery system. I was wrong about the mechanism, and I underestimated what it could do for the right patient. I still think I was right to be careful with it.
A patient of mine, a composite of several I see most weeks, came in eight months into a semaglutide prescription. A telehealth company had started it over a video call, no exam room involved. Her weight was down thirty-two pounds. Her primary care doctor was thrilled. She was exhausted in a way she couldn’t explain, cold all the time, and her hair was coming out in the shower drain. Good labs. Bad life. There’s a reason, and it usually isn’t the one anyone went looking for first.
We ran a DEXA scan. She’d lost almost as much lean mass as fat. Her resting metabolic rate had dropped further than her calorie intake could explain. What was the drug actually doing to the rest of her body while that number on the scale kept dropping? Nobody had asked her about protein. Nobody had asked her to lift anything heavier than a grocery bag from the HEB parking lot. The drug had done exactly what it was built to do. It suppressed her appetite. It didn’t know the difference between the muscle that keeps a person walking at eighty and the fat she wanted gone by summer.
We slowed the dose titration at Prime Vitality Care and added resistance training twice a week, which she hated for the first month and then stopped hating. We got her protein intake up to something her body could actually use. Six months later her weight was still down, her energy was back, and her hair had stopped coming out in the shower. The drug never changed. What changed was the plan built around it.
I think about the check-engine light a lot of my patients drive around with for months before they bring the car in. GLP-1s work the same way for some people. The light goes off, the number drops, and everyone assumes the problem underneath got solved. Sometimes it did. Sometimes the light just wasn’t wired to that particular problem, and something else was quietly wearing down while the dashboard looked fine.
This gets more complicated with compounded semaglutide and tirzepatide, which a lot of patients are on without knowing the dose, the purity, or even the pharmacy behind it. That’s off-label territory in the truest sense, and I tell every patient who brings it up exactly that. I’d rather prescribe the branded medication through a real pharmacy with a documented dose curve. When cost makes that impossible, I still want to know what’s actually in the vial.
I’m not against these medications. I prescribe them most weeks, and for a lot of my patients they’ve done real good. But I am against treating a falling number on a scale as the same thing as a person getting healthier. Those two things can travel together for a while and then split apart. Most people don’t find out until the scale stalls, the fatigue sets in, and nobody can explain why. That’s the same question I keep circling back to in the book I’m writing, Healing the Split — the labs said one thing, and the life said another. I write through more of it every week on Medium.
The muscle loss question isn’t fully settled yet. Long-term bone density data on these drugs is still young. Anyone who claims to know exactly what ten years of GLP-1 use does to a fifty-year-old body is overselling the evidence. I don’t know everything about where this goes. I do know what I check now that I didn’t check that first year. I wrote more about the principles behind that shift in Functional Medicine, Without the Hype.
My patient still texts me photos from the HEB checkout line sometimes, holding up a protein bar and asking if the sugar content is acceptable. It usually isn’t. I tell her so, and she buys it anyway about half the time. That’s fine. She’s still here, still lifting, still asking the question underneath the number. That was never going to fit on a prescription pad.
Dr Shiv Kumar Goel is an internal medicine and functional medicine physician in San Antonio. Across 25 years in hospitals, intensive care, and functional medicine, he has studied the gap between what medicine can measure and what patients actually live through. He writes Healing the Split and is the author of the forthcoming book Healing the Split: When Your Biology Is Fighting Your Biography.