What are GLP-1 agonists, really?
GLP-1 stands for glucagon-like peptide-1, which is a terrible name for a very elegant thing. It is a hormone your small intestine releases within minutes of food arriving — a chemical announcement that says, more or less, dinner has landed, everybody get ready. GLP-1 receptor agonists are laboratory-built copies of that announcement. They plug into the same receptors, say the same thing, and — crucially — refuse to shut up for days at a time.
The family includes semaglutide (Ozempic for type 2 diabetes, Wegovy for obesity), liraglutide (Victoza, Saxenda), dulaglutide (Trulicity), exenatide (Byetta, Bydureon BCise), and, sitting slightly apart from the rest, tirzepatide (Mounjaro, Zepbound). Some are daily, some weekly, most injected, one swallowed. Cousins, not clones — and their differences matter more than the shared label suggests.
You already know why you have heard of them. Between 2023 and 2025 this class stopped being a diabetes story and became a cultural one — red-carpet gossip, before-and-after reels, shortages that left people with type 2 diabetes hunting pharmacies for their own prescriptions. The hype is not entirely unearned; the trial data is genuinely striking. But it also flattened the pharmacology into a single sentence — "the skinny shot" — and that sentence is wrong in at least four different ways. Let's take the mechanism apart instead.
How they work — the simple version
Here is a fact that sounds like a mistake: if you take glucose by mouth, your pancreas releases substantially more insulin than if the identical amount of glucose is dripped straight into your vein. Same sugar, same blood level, very different insulin response. The gap is called the incretin effect, and it exists because your gut does not passively wait for sugar to show up in the bloodstream. It reports ahead.
GLP-1 is one of the two main messengers doing that reporting (the other is GIP — remember that one, it comes back later). Think of GLP-1 as a memo the intestine writes the moment food arrives, addressed to the pancreas, the stomach and the brain simultaneously.
Natural GLP-1 is broken down quickly by the enzyme DPP-4. Medicines extend the signal in different ways. Some resist that enzyme, some bind to albumin, and some use a slow-release formulation. Their duration varies: there are daily and weekly medicines, with product-specific instructions.
That one memo, now stretched across a week, lands in four places:
- The pancreas, beta cells. GLP-1 tells them to release insulin — but only while glucose is high. This is the single most underappreciated feature of the class. Injected insulin is a light switch: someone flips it and it stays on regardless of what your blood sugar is doing, which is exactly why insulin can drive you into hypoglycaemia. GLP-1 is a thermostat. When glucose falls back toward normal, the signal quietly stops mattering. On its own, this class barely causes low blood sugar at all.
- The pancreas, alpha cells. These make glucagon, insulin's opposite number, the hormone that tells the liver to dump stored sugar into the blood. In type 2 diabetes glucagon is often inappropriately loud after meals. GLP-1 turns it down — again, in a glucose-dependent way.
- The stomach. GLP-1 slows gastric emptying: food leaves the stomach into the intestine more gradually. Two consequences. The post-meal glucose spike gets flattened, because sugar arrives as a trickle instead of a flood. And you feel full sooner and stay full longer, because a stomach that stays occupied keeps saying so.
- The brain. Receptors in the hypothalamus and hindbrain read GLP-1 as a satiety signal. Patients describe the effect in oddly consistent language — the constant background hum about food, sometimes called food noise, gets turned down. Not willpower. Signal.
Which brings us to tirzepatide, the asterisk in this article. It is a dual agonist: it hits the GLP-1 receptor and the GIP receptor, that second incretin from a moment ago. Two mailboxes, not one. In the placebo-controlled SURMOUNT-1 obesity trial it produced substantial average weight loss (Jastreboff et al., NEJM, 2022). That trial did not directly compare it with semaglutide. Calling it "a GLP-1" is convenient shorthand and mechanistically incorrect — a distinction that matters when you are reading side-effect data or comparing trials.
What else they do to your body, beyond lowering blood sugar
Work backward from the mechanism and the side-effect list stops looking like fine print and starts looking inevitable.
The gut. Nausea, vomiting, diarrhoea, constipation, burping, a heavy full feeling an hour after two bites of lunch. This is the most common reason people quit the class, and it is not an unfortunate coincidence — it is the slowed stomach and the direct gut receptor activity that the drug was designed to produce, experienced from the inside. It is usually worst in the first weeks and after each step up in strength, and it usually settles. This is precisely why every one of these drugs is started low and increased slowly: the titration schedule exists to let your gut catch up. Skipping ahead does not speed up results, it just makes you miserable.
Muscle, not just fat. Rapid weight loss of any cause — surgical, dietary, pharmacological — costs lean mass along with fat, and body-composition data from the semaglutide and tirzepatide obesity trials show a meaningful share of the loss is lean tissue. Whether that share is worse than with other methods is genuinely debated. The countermeasure is not: resistance training and adequate protein while losing weight. If you take nothing else practical from this article, take that. Muscle is not a cosmetic detail; it is metabolic infrastructure and, later in life, the difference between a stumble and a broken hip.
The pancreas, in a different sense. Cases of acute pancreatitis have been reported since the first drugs in this class reached the market, and it appears in FDA labelling. A causal link at the population level remains debated — people with type 2 diabetes and obesity already have a higher baseline risk of pancreatitis and gallstones, which muddies every dataset. The practical version: severe persistent abdominal pain on these drugs is never something to wait out.
Thyroid C-cells. US labels for semaglutide, liraglutide, tirzepatide and several other products carry a boxed warning based on thyroid C-cell tumours in rodents; relevance to humans remains uncertain. These products are contraindicated with a personal or family history of medullary thyroid carcinoma or MEN2. Warnings differ between products, so check the exact label with the prescriber.
Gallbladder. Fast weight loss of any kind raises gallstone risk, and these drugs deliver it.
Blood sugar, when other drugs are in the room. By itself the class is a thermostat and hypoglycaemia is rare. Add a sulfonylurea or insulin — drugs that are light switches — and the floor can drop. This is why doses of those medications are often reduced when a GLP-1 agonist is added.
And the upside side effects. The cardiovascular findings were not what anyone was originally testing for. LEADER showed liraglutide reduced major cardiovascular events in people with type 2 diabetes at high cardiovascular risk (Marso et al., NEJM, 2016), and SUSTAIN-6 showed the same direction for semaglutide (Marso et al., NEJM, 2016). Then SELECT went further: in people with obesity and established cardiovascular disease but no diabetes, semaglutide cut cardiovascular events compared with placebo (Lincoff et al., NEJM, 2023). A drug class born to lower glucose turned out to protect hearts, including hearts belonging to people whose glucose was fine.
What people usually take with them, and why
For type 2 diabetes, the starting point has not changed: metformin is still the workhorse. What changed is the logic of what comes next. The American Diabetes Association's Standards of Care in Diabetes—2024 moved the field away from "add whatever lowers HbA1c most" and toward choosing the second agent based on what else the person is carrying. If there is established atherosclerotic cardiovascular disease or high risk of it, a GLP-1 receptor agonist with proven cardiovascular benefit is recommended — independently of how far the HbA1c is from target (ADA, Diabetes Care, 2024, Section 9).
The other modern partner is the SGLT-2 inhibitor family — empagliflozin, dapagliflozin and relatives, which make the kidneys spill glucose into the urine. The two classes work through completely unrelated doors, which is exactly why they combine well: additive glucose lowering, and complementary evidence in heart failure and chronic kidney disease. ADA 2024 explicitly supports combining them in people who need both benefits.
Around this core sits the rest of the standard cardiometabolic package — a statin for lipids, an ACE inhibitor or ARB for blood pressure and kidney protection. GLP-1 agonists were added to that scaffolding in the trials, not tested instead of it.
For obesity without diabetes, the pairing is not another drug. STEP 1 tested weekly semaglutide against placebo — with both arms receiving lifestyle intervention, counselling and activity guidance (Wilding et al., NEJM, 2021). SURMOUNT-1 did the same with tirzepatide (Jastreboff et al., NEJM, 2022). Nobody has ever run the trial where the drug replaces the behaviour change, because that trial was never the point. ADA's obesity guidance and the trial protocols agree: these medications are an addition to nutrition and activity, not a substitute for them.
A few interaction notes worth carrying. Because these drugs slow the stomach, oral medications can be absorbed differently — usually irrelevant, but it matters for drugs with narrow safety margins, and oral semaglutide has strict instructions about timing relative to food and other pills. Sulfonylureas and insulin, as mentioned, usually need adjusting downward. And there is a well-documented anaesthesia issue: a stomach that empties slowly may still hold food before a procedure, which surgical teams need told in advance.
Red flags — when to call a doctor
Ordinary nausea in week two is expected. The following are not, and none of them belong in the "I'll mention it at my next appointment" pile.
- Severe, persistent abdominal pain, often in the upper abdomen and sometimes boring through to the back, especially with vomiting that will not stop. Possible pancreatitis. This is an emergency-department symptom, not a wait-and-see one.
- A lump in the neck, hoarseness that does not clear, trouble swallowing, or persistent shortness of breath. Rare, but these are the specific thyroid symptoms the boxed warning exists for.
- Vomiting or diarrhoea severe enough to stop you keeping fluids down. Dehydration on top of these drugs can injure the kidneys, particularly in anyone also taking diuretics or blood-pressure medication that acts on the kidneys.
- Sudden changes in vision in anyone with diabetes, especially with existing retinopathy. Rapid improvement in long-standing high blood sugar can transiently worsen diabetic eye disease — a known signal from SUSTAIN-6 — and needs an eye assessment, not a wait.
- Shakiness, sweating, confusion, racing heart — classic low blood sugar, which should prompt a review of every other glucose-lowering drug on the list.
- Chest pain, sudden severe shortness of breath, one-sided weakness or slurred speech — the usual cardiovascular emergencies, and this class is prescribed to a lot of people who already carry that risk.
- Pain in the upper right abdomen with fever or yellowing of the skin or eyes — possible gallbladder trouble.
And one that is less dramatic but more common than people admit: if the drug has left you eating so little that you feel weak, dizzy, cold, or your hair is thinning, that is not the medication working extra well. That is undernutrition, and it is a conversation to have with a clinician promptly.
What people get wrong
"Ozempic is a weight-loss drug." Ozempic is semaglutide approved for type 2 diabetes. Wegovy is the same molecule, at higher strengths, approved for weight management. Same chemistry, different label, different indication, different studies behind each. The reason this matters is not pedantry — it is that during the shortage years, people conflating the two contributed to a supply squeeze that left diabetic patients without their treatment.
"It just makes you stop eating." Appetite is one of four effects, and on its own it would not explain the glucose data. Strip out the incretin insulin response, the glucagon suppression and the slowed gastric emptying and you have a very different, much weaker drug. The satiety part is simply the effect people can feel, so it is the part that got the headlines.
"They cause hypoglycaemia like insulin does." The risk is generally low when these medicines are used alone, but it is not zero. It rises with insulin or a sulfonylurea, so the prescriber may need to adjust the accompanying treatment.
"Tirzepatide is a GLP-1 agonist." It is a dual GIP and GLP-1 receptor agonist. Two incretin pathways, not one. It is grouped with GLP-1 agonists for convenience and it genuinely shares most of the class behaviour, but treating trial results from one as interchangeable with the other is sloppy reading.
"Once you've lost the weight, you can stop." This is the myth with the best evidence against it. The STEP 1 trial extension followed participants after semaglutide was withdrawn: on average they regained roughly two-thirds of the weight they had lost within a year, and the cardiometabolic improvements drifted back toward baseline alongside it (Wilding et al., Diabetes, Obesity and Metabolism, 2022). This is not a moral failure or a willpower problem. It is what happens when you remove a signal the body was responding to — the same way blood pressure returns when an antihypertensive stops. Obesity behaves like a chronic condition, and these drugs treat it like one.
"Anyone can take them to lose a few pounds." Eligibility and contraindications depend on the product and the medical indication. Pregnancy, a history of pancreatitis, severe gastrointestinal disease and the thyroid conditions described above need a prescriber-led assessment. Unregulated products bought online are not a substitute for a prescribed, authorised medicine.
"Nausea proves the medicine is working." It does not. Gastrointestinal symptoms can occur during treatment, but they are neither necessary for benefit nor a measure of effectiveness. Persistent vomiting, dehydration or difficulty eating needs medical advice; do not simply endure symptoms or adjust the dose yourself.
Tirzepatide and contraception. Its US label advises people using oral hormonal contraception to use a non-oral method or add a barrier method for four weeks after starting and for four weeks after each dose increase. Discuss the plan with the prescriber; this warning is specific to the product.