Start · Catalogue · Profile · Table
Healthcare BODY HANDBOOK
Healthcare · §625
GLP-1s Beyond Weight Loss
The drug class everyone is talking about for weight loss does five other things that probably matter more. It cuts heart attacks and strokes by about a fifth in adults with established heart disease — and most of that benefit isn't explained by the weight loss itself. It slows kidney decline in diabetics with chronic kidney disease. It roughly halves nightly breathing pauses in moderate-to-severe sleep apnea. It blunts the urge to drink. Blood pressure, blood sugar, and inflammation all move in the right direction in the same person at the same time. The catch: prescription-only, a weekly injection at $200 to $1,000 a month, and most of what you gain comes back if you stop.
Decide · Weekly Evidence Strong Chapter Healthcare

Five separate large trials in the last two years pushed this past the weight-loss headline. Cardiology, nephrology, and sleep medicine are rewriting guidelines around what these molecules do to the heart, kidneys, and airway. The evidence is about as solid as drug evidence gets. The price of entry is a clinician, a weekly injection, and an ongoing bill that doesn't end.

The story most people know — these drugs make you less hungry — is real but partial. GLP-1 receptor agonists mimic a gut hormone your body releases after eating. The receptors they hit are not only in the gut and pancreas. They sit on satiety neurons in the hypothalamus (the reason appetite drops), on dopamine cells in the brain's reward circuit (the reason cravings drop), on kidney filter cells called podocytes, on the lining of blood vessels, and on heart tissue. Same molecule, eight or nine different jobs.

For the heart, the simplest theory was always weight loss does it. The cleanest test of that theory said no — only about a third of the heart-attack reduction in the largest cardiovascular trial was statistically explained by the weight or waist-size change Ryan 2025. The rest looks like direct effects on blood vessel inflammation, blood pressure, the cholesterol mix, and the inflammatory marker hsCRP — which the drug roughly halves.

For the kidney, the drugs blunt the over-filtration that wrecks the organ in diabetes and reduce the stress on podocytes that leads to protein leaking into urine. For the brain's reward system, activating the same kind of receptor in the ventral tegmental area and nucleus accumbens flattens the dopamine spike that food, alcohol, nicotine, and opioids normally produce — a single mechanism that, on paper, explains why one molecule shows up in trials for food, drink, and substance use at once. For sleep apnea, the dominant lever is fat loss in the neck and around the airway, but reduced inflammation and lower fluid retention contribute. None of these effects operate in isolation; the drug is doing all of them, in one body, every week, for as long as you keep taking it.

What's actually been tested

The case for these drugs as something more than weight-loss tools rests on five large randomised trials, all of them in the New England Journal of Medicine, all of them tested against placebo on top of standard care. Each one targeted a different organ system. Each one hit.

Two more trials covered heart failure with preserved ejection fraction — the version of heart failure where the pumping looks normal on echo but the patient still can't walk up a flight of stairs. STEP-HFpEF showed semaglutide improving the standard symptom score by nearly 8 points more than placebo and adding ~22 metres to 6-minute walk distance Kosiborod 2023. SUMMIT, with tirzepatide, went further: it cut the composite of cardiovascular death or worsening heart failure event by 38% Packer 2025. HFpEF had been almost untreatable for decades. Two GLP-class trials in eighteen months have changed that.

The class signal in diabetics predates the obesity wave: LEADER (liraglutide, 9,340 patients) cut major heart events by 13% and cardiovascular death by 22% in 2016 Marso 2016a; SUSTAIN-6 (semaglutide, 3,297 patients) cut the same composite by 26%, driven mostly by stroke Marso 2016b. So the cardiovascular signal does not depend on weight loss being large or being the only thing happening.

The addiction signal is the newest and the thinnest. Hendershot 2025 randomised 48 non-treatment-seeking adults with alcohol use disorder to nine weeks of low-dose semaglutide or placebo. In a controlled lab session at the end, the drug group drank less alcohol and reached lower breath alcohol concentrations. Weekly cravings dropped. A handful of smokers in the sample cut cigarettes per day Hendershot 2025. One small trial doesn't settle anything, but it lines up with consistent pharmacoepidemiologic patterns: diabetics put on GLP-1 drugs go on to develop fewer new substance use disorders — alcohol, cannabis, nicotine, cocaine, opioids — than diabetics put on other drug classes. The mesolimbic dopamine mechanism makes the pattern biologically plausible. Treat the addiction line as promising hypothesis, not as a prescription you'd write today.

What happens if you keep ignoring this

The reader this section is for is the 55-year-old with a prior heart attack, a statin, an ACE inhibitor, a blood pressure that's pretty good on most readings, a BMI of 32, an HbA1c creeping toward 7, and a partner who's stopped asking about the snoring because the answer hasn't changed in five years. The medicine cabinet looks adequate. The next event is still in the post.

The baseline forecast for that person without this drug class — assuming standard care holds — is roughly the SELECT placebo arm: an 8% chance of dying from cardiovascular causes, having another heart attack, or having a stroke over the next three years Lincoff 2023. Year by year, that's the slow accumulation of small dramas. The afternoon angina that the spouse notices. The cardiology follow-up that goes from "stable" to "we should add another agent." The kidney filtering rate that drops a few millilitres a year — and the year your nephrologist starts using the word predialysis. The CPAP that you wore for six months in 2022 and now lives in the closet, and the daughter who notices that you nod off at family dinners.

None of those are dramatic. None of them are a moment. They are the version of the next decade where the medicine cabinet looks the same and the body keeps doing what bodies with high cardiovascular and renal load do. The 20% MACE reduction, the 24% kidney composite reduction, the half-cut in apnea events — those numbers are the difference between that decade and a different one. Not a transformed body. A body that doesn't have the heart attack it was on track to have, doesn't end up on dialysis on the schedule the labs were predicting, doesn't fall asleep in the chair after dinner.

What it looks like in practice

All current versions are subcutaneous injections, the pen looks like the insulin pens diabetics have used for decades, and you give it to yourself once a week. Semaglutide (sold as Wegovy or Ozempic depending on the indication) and tirzepatide (Zepbound or Mounjaro) are the two molecules with the trial data that matters here. Liraglutide is the daily older cousin and largely superseded. An oral non-injectable GLP-1 that doesn't require refrigeration is in late-stage trials but not yet on shelves.

Every regimen starts on a low dose and steps up monthly over four or five months — a slow ramp specifically designed so the gut adapts and the nausea is bearable. Skipping the ramp is the classic mistake. The target maintenance dose depends on what you're treating, and the doses that produced the trial results are not arbitrary:

  • Heart and HFpEF — semaglutide 2.4 mg per week, the same dose used for weight loss.
  • Kidney in diabetic CKD — semaglutide 1.0 mg per week, the diabetes-glycaemic dose.
  • Sleep apnea and HFpEF — tirzepatide titrated to 10 or 15 mg per week.
  • Alcohol use disorder (research-grade) — semaglutide 0.25 to 0.5 mg per week, well below the obesity doses.

When not to take it

The signal that didn't hold up: a spontaneous-report flag from European regulators in 2023 raised concerns about suicidal thoughts, and the FDA opened a review. Three years of pharmacoepidemiologic data and trial-arm comparisons later, the FDA in 2025 asked manufacturers to remove the suicide warning from labelling, finding no causal signal FDA 2025 Wang 2024. The depression-risk narrative in lay press outran the data.

If you're on a sulfonylurea or insulin for diabetes, the dose of those will need to come down — combined hypoglycaemia is the main avoidable bad day. Gallbladder problems and gallstones are more common with rapid weight loss, an effect that isn't unique to this drug class but is amplified by the weight-loss magnitude.

Where this goes wrong

You stop and it mostly comes back. The standard trial design for withdrawal told a consistent story: in STEP 4, patients who switched from semaglutide to placebo after 20 weeks regained roughly 7% of body weight in the next year, while those who stayed on the drug kept losing Rubino 2021. The longer extension of the original obesity trial showed two-thirds of the weight back within a year of stopping, with blood pressure, cholesterol and HbA1c drifting back toward pre-treatment baselines Wilding 2022. The cardiovascular and kidney trials studied continuous use; no one has yet shown that a year on, then off, leaves you with the heart-attack reduction. The implicit recommendation, like statins, is long-term.

You lose muscle along with fat. Roughly a quarter to two-fifths of the weight lost on these drugs is lean tissue, mostly skeletal muscle, when nobody trains and nobody eats protein. That ratio isn't unique to the drug — most weight loss looks like this — but the absolute muscle loss is larger because the absolute weight loss is larger. In an older adult already losing muscle to age, that's the kind of side effect that shows up as a fall on a kitchen tile two years later. Two strength sessions a week and a kilogram of protein per kilogram of body weight, or a bit more, blunts the loss substantially.

The gut wins. A small fraction of people get nausea so bad they quit. Slowing the titration almost always rescues this; quitting on month one and writing off the class is the avoidable error.

You buy the wrong version. Compounded semaglutide and tirzepatide flourished during the 2023–24 shortage; that's over now. Anything labelled "research grade" or sold by a telehealth site that bypasses the legitimate manufacturers is unregulated and increasingly illegal.

What most coverage gets wrong

"It's just a weight-loss drug." The cleanest counter is the SELECT mediation result: in 17,604 patients with established heart disease, only about a third of the cardiovascular benefit tracked with how much waist size came off Ryan 2025. The benefit shows up in diabetics on glycaemic doses where weight loss is modest. The kidney protection in FLOW is barely waist-mediated. Weight loss is one of the things this class does. It is not most of what this class does.

"It causes depression." Spontaneous reports said maybe; rigorous review said no. The FDA's 2025 evaluation found no causal signal in trial data or in real-world cohorts and asked manufacturers to drop the warning FDA 2025.

"It only works if you're obese." The diabetes cardiovascular trials were not selected for obesity. The alcohol trial used a dose smaller than the weight-loss dose, in adults who didn't have to be obese. The CV benefit in SELECT was consistent across baseline weight categories.

"You can microdose to get the brain effects without the gut effects." Possibly, eventually — the alcohol trial used roughly a fifth of the obesity dose and reported real signal. But that's one trial of 48 people. Anyone selling a microdose protocol for cravings today is ahead of the data.

What changes if you start

The honest answer is that most of the wins are wins you never see — the heart attack that didn't happen, the kidney that's still filtering, the stroke your kids didn't get the call about. Long-game pharmacology is forecast through statistics, not felt through a sensation.

That said, the short-term shifts are real and trackable.

  • Weeks one through eight. Mostly nausea. The version of the day where lunch feels like an obligation rather than a craving — that's the drug working, and also that's why a quarter of people quit. The first labs at month three are when the story starts: blood pressure usually down a notch, fasting glucose noticeably lower, the inflammation marker hsCRP cut in half on its way to baseline.
  • Months three through six. If you have sleep apnea: the partner stops elbowing you, the morning headache fades, the afternoon meeting you used to fight to stay awake in goes differently. If you have HFpEF symptoms: the flight of stairs that took two pauses now takes one, then none. In the trial, treated patients added 21.5 metres to their 6-minute walk distance over a year Kosiborod 2023 — about a quarter of a track lap, in a population that previously couldn't add any.
  • Year one through three. The statistical window of the SELECT and FLOW trials. In a cohort of 67 people like you with established heart disease, one fewer of them has a major cardiovascular event over three years than would have without the drug Lincoff 2023. You don't feel an NNT. You sit at your nephrologist's office and watch the eGFR slope flatten out from what was a three-year line on a graph headed somewhere bad.
  • The decade ahead. The drug class is too new for any single person to have a decade on it; the diabetes molecules have the longest track record at around ten years now. The mortality reductions in FLOW (20% lower all-cause death) and the absolute event reductions across SELECT, SUMMIT, and the diabetes CVOTs add up, when projected forward, to a meaningful difference in how the next ten years end. Cardiovascular events the body wasn't on track to dodge. A kidney that wasn't on track to fail.

Adjacent territory this entry doesn't cover:

  • The weight-loss case itself. Treated as its own entry. The dosing protocol, side-effect curve, and trial programme overlap, but the indication and the social context don't.
  • Glycaemic management in type 2 diabetes. The original indication, with decades of specialist context; sits in standard diabetes care rather than the "beyond weight loss" frame.
  • SGLT2 inhibitors — the other modern cardiometabolic class with overlapping kidney and heart-failure benefit. Often combined with GLP-1s and worth knowing about in their own right.
  • Fatty liver. Another organ these drugs act on; the benefit overlaps with resmetirom, the drug approved specifically for the inflamed form (MASH).
  • PCOS. The metabolic cousin where these drugs work on the insulin resistance underneath — the weight, the cycles, and the long-term diabetes risk.
  • Bariatric surgery. The historical reference point for both weight loss and cardiometabolic remission; trial-to-trial comparison is incomplete.
  • ApoB and lipoprotein(a) — the cardiovascular risk markers that statins and GLP-1s both move, but through different pathways.
  • The oral non-peptide GLP-1 agonists in late-stage trials — same target, different molecule, pill not injection.
Related in the handbook
Helps
  • — Beyond weight, GLP-1s help fatty liver directly — useful when lifestyle change isn't reversing it fast enough.
  • — In PCOS these drugs target the insulin resistance underneath, helping weight, cycles, and the long-term diabetes risk.
  • — In diabetics with kidney disease these drugs slow the decline — one of the benefits beyond weight loss.
  • — Fatty liver is usually first spotted as a quiet ALT bump on a routine panel — learn to read it before deciding on a drug.
  • — Halving sleep apnea is one of the under-sold wins of these drugs, alongside the heart and kidney effects.
  • — These drugs quietly blunt the urge to drink — an unexpected effect now being studied for alcohol use.
Alternatives
  • — On top of weight loss, GLP-1s improve fatty liver — overlapping territory with the dedicated MASH drug resmetirom.
  • — Berberine gets sold as a natural stand-in for these — it nudges metabolism, but nowhere near as much.
Related
  • — Newly diagnosed type 2 diabetes is a prime case for a GLP-1, which does far more than drop blood sugar.
  • — Rapid GLP-1 weight loss can hollow the face ('Ozempic face') — the body-fat-and-face trade-off explains why and how to soften it.
  • — The big upside of GLP-1s comes with a catch: a third of the weight lost is muscle and bone unless you defend it.
  • — Sleep apnea is one more condition these incretin drugs help, on top of heart, kidney, and metabolic gains.
·
625