People are fat because they eat too much, eat the wrong stuff, and don’t move their body. That’s the equation. Eat less and move more. There’s no clever way around it, there’s no math trick, and there’s no medication that writes you out of it. What the GLP-1 drugs actually do, and I’ll say it up front so you don’t have to read three thousand words to find the point, is they make sticking to the equation doable. They quiet the appetite signal so the choices feel possible. They make the results show up faster, which is the part that encourages more effort, which is the part that makes the whole thing stick. The patient is still the one driving. The meds just stop the steering wheel from fighting back.
The patient is still the one driving. The meds just stop the steering wheel from fighting back.
Full disclosure up front, I prescribe these. I think they are one of the more interesting tools we’ve gotten in a long time, and I also think a lot of what’s been written about them, in both directions, is nonsense. The “miracle drug, you’ll never have to think about food again” headlines are a damn liar. The “this is cheating, real men just have discipline” takes are a different kind of damn liar. The honest version is in the middle, and it’s the version most of my patients land on a few months in once the novelty has worn off and the real work is the only thing left.
What these drugs actually are
GLP-1 stands for glucagon-like peptide-1, which is a hormone your gut already makes after a meal. Its job is to tell your pancreas to release insulin, tell your stomach to empty more slowly, and tell your brain you’ve had enough. The first GLP-1 drug, exenatide, was a synthetic version of a peptide isolated from Gila monster spit (genuinely, the venom of a lizard that eats once every few months and needed a long-acting satiety signal to survive between meals, which is a hell of an origin story). The newer drugs are engineered versions of human GLP-1 that resist breakdown long enough to be dosed once a week instead of needing constant infusion.
The semaglutide drugs (Ozempic, Wegovy, Rybelsus) hit one receptor, the GLP-1 receptor. Tirzepatide (Mounjaro, Zepbound) hits two, GLP-1 and GIP, GIP being a second gut peptide that adds its own metabolic effects on top. The dual-agonist mechanism is most of why the tirzepatide trials posted bigger weight loss numbers than the semaglutide trials. More receptors, more downstream effects on satiety and insulin sensitivity, more weight off. Whether you need the dual mechanism or the single one depends on how much room you have to lose, what your insurance will cover, and what your gut tolerates.
The mechanism story matters because it tells you what the drugs can and cannot do. They quiet the appetite signal. They slow gastric emptying, so the meal you ate at noon is still sitting in your stomach at three and you don’t want a snack. They reduce reward signaling around food, which is the part patients describe as “food noise getting quieter,” meaning the constant background chatter about what’s in the pantry just turns down. Appetite signaling is real biology, and willpower has been losing the fight with it for a long time. The drugs level the playing field. That’s the entire thing.
The specific drugs and what to call them
This part confuses everybody, including a fair number of prescribers, so worth being clear. Ozempic and Wegovy are the same molecule, semaglutide, sold under two labels by Novo Nordisk. Ozempic is FDA-approved for type 2 diabetes, dosing up to 2 mg weekly. Wegovy is the same drug at higher doses (up to 2.4 mg weekly) approved for weight management in adults with obesity or overweight with a comorbidity. Same molecule. Different label, different price, different insurance pathway. The reason your friend’s Ozempic costs eight hundred dollars and your Wegovy costs sixteen hundred is not a chemistry difference, it’s a market segmentation move.
Mounjaro and Zepbound are the same molecule, tirzepatide, sold under two labels by Eli Lilly. Mounjaro is the diabetes label. Zepbound is the obesity label. Same drug, same pen, different box, different insurance pathway. The dance most patients end up doing is figuring out which label their insurance will pay for, which is often the diabetes label if they have any glucose dysregulation, and the prescriber writes accordingly. Nothing about that is shady, the molecule is the molecule and the prescriber is doing the patient a favor by routing around the dumb insurance friction.
There are two other drugs in the family worth naming alongside, starting with Saxenda which is liraglutide, an older GLP-1 that dosed once daily instead of weekly, with weaker weight loss numbers, mostly displaced by semaglutide and tirzepatide for the obesity indication. Rybelsus is oral semaglutide, a pill version, dosed daily, which gets around the needle question but at a tradeoff in absorption (you have to take it on an empty stomach with a sip of water, then wait thirty minutes before anything else, which is more lifestyle friction than people expect). Useful for the right patient. Not the default.
Compounded semaglutide and tirzepatide, the honest version
For a long stretch in 2023 and 2024 the brand-name drugs were on the FDA shortage list, which legally opens the door for compounding pharmacies to make their own version of the active ingredient and sell it directly. That’s how the entire gray-market telehealth wave got off the ground. Once the shortages resolved, the FDA started enforcing the cutoff, and the legal status got messier. Some compounding pharmacies have stopped. Some are still selling under various legal theories. Some are selling research-grade semaglutide that was never meant for human use, which is a category I’d avoid.
The honest version. Compounded GLP-1s from a 503A compounding pharmacy with a real prescription, real labs, and real follow-up are not categorically worse than the brand-name drug, especially if the price difference is what stands between a patient and being on the medication at all. The active ingredient is the same molecule. The risk goes up when the supply chain gets cloudy, when the pharmacy can’t tell you where the active pharmaceutical ingredient came from, when there’s no provider doing follow-up, when the dosing instructions are coming from a chat window instead of a clinician. I’m not going to moralize about the gray market. I’m going to point out that the part that makes it dangerous isn’t the compounding, it’s the absence of medical oversight around it. Get the medication from somebody who’s also going to draw your labs.
What the trial data actually says
The SURMOUNT-1 trial of tirzepatide in adults with obesity, no diabetes, posted an average body weight reduction of around 20.9% at the 15 mg dose over 72 weeks. That number is, by historical standards, ridiculous. Bariatric surgery numbers. The STEP-1 trial of semaglutide in a similar population posted around 14.9% over 68 weeks at the 2.4 mg dose. Both numbers blow past anything the old weight loss drug class (phentermine, orlistat, the Contrave / Qsymia generation) was able to do, which is most of why the field reorganized around GLP-1s in the span of about three years.
What the numbers don’t tell you is the distribution. Average weight loss of 20% means a lot of patients lost more, some lost less, and a chunk of non-responders (probably ten to fifteen percent of patients) lost almost nothing. Nobody knows exactly why the non-responders don’t respond. Could be genetic variation in receptor density, could be eating patterns that the medication can’t fully address, could be other hormonal stuff going on. If you’re at six months and the scale hasn’t moved, that’s a real signal, and the right move is to either escalate the dose, switch molecules (semaglutide to tirzepatide is a common move), or have a real conversation about whether this is the right path for this body.
Why “just try harder” hasn’t worked, and what that means
If you’re a guy in your forties reading this, you’ve probably tried Atkins, you’ve probably done keto, you’ve probably done some version of intermittent fasting, you’ve probably bought a gym membership, you’ve probably told yourself last January that this was the year. Some version of that. Most guys have a binder full of diet attempts and a body that has slowly added weight anyway. The cultural script says that’s a willpower problem. The biology says willpower is a finite resource and you’ve been spending it against an appetite signal that gets louder the longer you restrict. The math hasn’t been fair.
None of that means the equation is wrong. Eat less and move more is still the equation. What’s been wrong is the framing that the equation should be effortless once you “decide.” The GLP-1s don’t change the equation, they make executing the equation possible by quieting the part of your physiology that’s been making it feel impossible. People who lose weight on these drugs also eat less and start moving. They have to. The drug isn’t doing the work, it’s making the work doable. The minute the patient stops eating less and stops moving, the drug stops working too, which is the part most of the headlines leave out.
The math is the math. GLP-1s do not write you out of the math, they make sticking to the math doable.
Side effects, the direct version
Most of what people experience is gut. Nausea, sometimes vomiting, constipation that swings into diarrhea, the famous sulfur burps that taste like rotten eggs (real, common, mostly dose-related, mostly resolves with time or with stepping back a dose), reflux, early fullness so dramatic that patients describe putting a fork down at half a meal because the body says no more. The mechanism is the same one that makes the drugs work. The stomach is emptying slower, the satiety signal is firing earlier, and at higher doses or faster titration the body protests. Most of it is manageable with slower titration, eating smaller meals more often, hydrating like you actually like it, and giving the body two to four weeks to settle at each new dose.
The more serious stuff worth knowing about. Gastroparesis (where the stomach essentially stops emptying) is a recognized complication, mostly in patients who already had marginal motility. Gallstones happen more often with rapid weight loss, GLP-1 or not, because of how the gallbladder concentrates bile when fat intake drops. Pancreatitis is on the label as a rare event, and the signal in the trial data is small but real. Anyone with a history of pancreatitis needs a real conversation before starting. There’s also been an investigation by the FDA into a suicidality signal that surfaced in early reporting, and the most recent reviews have not borne it out as a true association, but if you have an active mood picture going on you and the prescriber should be paying attention to it anyway.
The one I push patients hardest on is muscle loss. Rapid weight loss of any kind takes muscle along with fat, and the GLP-1 trials show somewhere between a quarter and a third of the lost weight being lean mass if the patient does nothing about it. That’s not fine. That’s a metabolism that’s now lower than where you started, a body that’s weaker than where you started, and a setup for regain that’s worse than where you started. The fix isn’t mysterious. Eat protein, somewhere in the range of 0.7 to 1 gram per pound of goal body weight, and lift weights, two or three days a week, real loads, full-body movements. If you’re going to do this medication, you have to do the weight room piece. That part isn’t optional, and it isn’t set in stone what your body composition looks like coming out the other side. The choices you make on the medication decide that.
How the ramp-up actually works
Both semaglutide and tirzepatide titrate slowly, and most of the side effect grief comes from skipping steps. Semaglutide starts at 0.25 mg weekly for four weeks, then 0.5 mg for four weeks, then 1 mg, then 1.7 mg, then 2.4 mg at the top end for the weight management label. Tirzepatide starts at 2.5 mg for four weeks, then 5 mg, then 7.5, 10, 12.5, 15. The titration is partly about giving the gut time to adapt and partly about finding the lowest dose that does the work, since the side effect profile gets uglier as the dose climbs and there’s no prize for being on the maximum dose if 7.5 mg of tirzepatide is doing the job.
Most patients will land somewhere in the middle of the ladder. Pushing to the ceiling is fine if you need the additional weight loss and you’re tolerating it, but it’s worth checking, before you climb, whether the rate of loss has been adequate at the current dose. If you’re losing one to two percent of body weight a month at a mid-ladder dose, that’s a healthy rate, and climbing further is buying you nausea for marginal additional results. Slower loss preserves more lean mass anyway.
Who these fit
The FDA labels read: BMI of 30 or higher, or BMI of 27 or higher with at least one weight-related comorbidity (hypertension, dyslipidemia, type 2 diabetes, obstructive sleep apnea, cardiovascular disease). For diabetes specifically, the Ozempic and Mounjaro labels don’t require a BMI threshold. The clinical reality is broader than the label, and a lot of prescribing happens in the area where the patient is metabolically dysregulated, the labs are starting to slide, the weight is creeping, and the patient has done several rounds of “trying harder” without lasting results. That’s the patient who tends to do best on these drugs, because they bring the work ethic already and the medication adds the leverage their body has been missing.
The teen and young adult population is a slightly different conversation. Wegovy is FDA-approved down to age 12, Zepbound has a pediatric study running. I think about teens carefully. A fifteen-year-old who’s been heavy since elementary school, has fatty liver on imaging, has a glucose that’s starting to drift, and has tried every reasonable behavior intervention is a different patient than a fifteen-year-old who’s annoyed about looking different than his friends. Both are real conversations. They don’t get the same answer.
Who these don’t fit
The boxed warning is medullary thyroid carcinoma. Animal studies in rats showed thyroid C-cell tumors at high doses. The human signal is unclear and probably absent, but the warning stands. Personal history of MTC or family history of MTC means these drugs are off the table. Same with Multiple Endocrine Neoplasia syndrome type 2 (MEN2), which is a rare genetic condition that predisposes to MTC. If you’re not sure whether your family history qualifies, that’s a question for the prescriber, not a question for the internet.
Pregnancy is a contraindication. Animal data shows fetal harm and there’s no good human data, which means anybody who could become pregnant needs reliable contraception while on the drug and a wash-out period before trying to conceive (two months for semaglutide given the long half-life, similar for tirzepatide). Active gastroparesis is a contraindication for obvious reasons (you don’t want to further slow a stomach that’s already not emptying). Active eating disorders, particularly the restrictive ones, are a hard no. Suppressing appetite in someone whose disorder is built around restriction is throwing gasoline on a fire.
What happens when you stop
The honest version, which the marketing usually skips. The STEP-4 trial took patients who’d lost weight on semaglutide for twenty weeks, then randomized half of them to continue and half to switch to placebo. The continued group kept losing. The placebo group regained, on average, about two-thirds of what they’d lost over the next year. The conclusion that got attached to that trial, “GLP-1s are a chronic medication, you stay on forever,” is partly right and partly wrong. It’s right that if the patient stops the medication and goes back to the pre-medication eating pattern, the weight comes back. It’s also right that not everybody has to stop, and not everybody who does stop regains. The variable is whether the behavior change stuck.
This is the part where the choice-versus-biology framing gets clarifying. The drug was leverage on a set of choices. If, while on the drug, the patient built a habit of eating protein at breakfast instead of grabbing a pastry, of lifting three times a week, of taking a walk after dinner instead of sitting on the couch, of getting seven hours of sleep instead of five, then a lot of that infrastructure stays after the drug taper. Some patients can taper off and hold most of their loss with the new behaviors carrying the weight. Some patients do better on a maintenance dose, which is half or a quarter of what they were on at peak loss, just enough to keep the appetite signal quieter than baseline while the lifestyle changes do the rest. Some patients need to stay on the full dose. That’s a conversation, not a default.
The long-term picture, still being written
The cardiovascular data from the SELECT trial is the big one. Semaglutide reduced major cardiovascular events by about 20% in adults with established cardiovascular disease and obesity, no diabetes, over about three and a half years of follow-up. That’s another tick in the GLP-1-is-better column, because the historical weight loss drugs either had no cardiovascular signal or had a bad one (the fen-phen disaster being the cautionary tale that haunted the whole field for a generation). The FLOW trial showed slowing of kidney disease progression in diabetic patients on semaglutide, which is real, important, and pushes the calculus toward continuing the medication in the right patient.
The more speculative stuff. There’s a craving signal that keeps showing up, anecdotal reports of patients on GLP-1s losing interest in alcohol, gambling, nicotine, the whole class of behaviors that ride on the same dopamine reward circuits as food. The trials are starting to follow up on this. If it pans out, it’s going to reshape a chunk of addiction medicine. There’s an Alzheimer’s signal that’s even earlier, with some evidence that GLP-1 receptor activity in the brain might be protective against neurodegeneration. That’s a trial we’re going to be reading for the next decade. None of this is mature enough to prescribe on. All of it is interesting enough to watch.
How a real prescribing visit should go
If you’re getting these drugs from a website that has never asked you to draw labs, has never asked about your family history, has never had a video visit with you, has never followed up on anything past the first prescription, you’re not in a medical relationship, you’re in a vending machine relationship. The drugs are powerful enough that the vending machine model is genuinely a bad idea, not because of the molecule itself but because of everything that needs to happen around it. Real prescribing means a baseline metabolic panel, a thyroid panel, a lipid panel, and an HbA1c. It means a family history conversation that goes past the form. It means follow-up at the four to six week mark to check on tolerance, then quarterly while titrating, then twice a year on maintenance. It means somebody to call when the nausea won’t quit and you’re not sure whether to step the dose down or push through.
I’m not going to pretend the in-person model is the only ethical way to do this. Telehealth is fine, and in a lot of cases it’s the only way the patient was going to engage with care at all. What matters is that there’s a clinician on the other end of the screen who knows your labs, knows your case, and is going to call you back when you have a real question. Anything past that is preference.
Where this all lands
If we’re being honest, the GLP-1 drugs are the biggest swing the field has taken at the weight problem in fifty years, and they work better than anything before them. They also don’t change the underlying equation. Eat less, eat better, move more, lift heavy things, sleep enough, drink water like you actually like it. The boring stuff still wins. The drugs make the boring stuff easier to stick to, and they make the early results show up fast enough that you stay motivated to keep doing the boring stuff, which is the part that compounds.
The risk if you take the drug and don’t change behavior is real and it’s worth naming. You lose weight, but a third of it is muscle. You stop the drug because you got to a number on the scale that you liked, and the weight regains because nothing else changed. You spent a year and ten thousand dollars and a fair amount of nausea to end up roughly where you started, with less muscle than you started with. That is a wasted opportunity, and it’s the version that happens when the medication is doing all the work and the patient is doing none. Don’t be that patient.
The math is the math. The choices are yours. The meds make sticking to the choices easier, and they make the results show up faster, which encourages more effort, which is how the whole thing actually works. That’s the version I tell patients, that’s the version the data supports, and that’s the version that holds up two years out when the novelty has worn off and the only thing left is whether the patient kept lifting, kept eating protein, kept walking, kept sleeping. The medication is leverage. The work is yours.
Sources
FDA prescribing information for Ozempic (semaglutide), Wegovy (semaglutide), Mounjaro (tirzepatide), Zepbound (tirzepatide), Saxenda (liraglutide), and Rybelsus (oral semaglutide). STEP program (Wilding JPH et al., NEJM 2021; Davies M et al., Lancet 2021) on semaglutide 2.4 mg for weight management. SURMOUNT-1 (Jastreboff AM et al., NEJM 2022) on tirzepatide for obesity without diabetes. SELECT trial (Lincoff AM et al., NEJM 2023) on semaglutide and major adverse cardiovascular events. FLOW trial (Perkovic V et al., NEJM 2024) on semaglutide and kidney outcomes in type 2 diabetes. American Gastroenterological Association clinical practice guideline on pharmacological interventions for adults with obesity (2022). Cochrane systematic reviews of GLP-1 agents for adults living with obesity (Bracchiglione J et al., semaglutide, Cochrane Database Syst Rev 2025, CD015092; Franco JV et al., tirzepatide, Cochrane Database Syst Rev 2025, CD016018). STEP-4 (Rubino D et al., JAMA 2021) on weight regain after semaglutide discontinuation.