Twelve short chapters written in plain English, from how weight is regulated to eating, muscle, stopping treatment and costs. Tap any underlined term for a quick definition.
1. What obesity is, and what BMI can and cannot tell you
4 minute read.
A medical condition with a lot of biology behind it
Many medical organizations now describe obesity as a chronic, relapsing condition rather than a failure of willpower. The next chapter explains the biology behind that view.
That framing matters because it changes the question. Instead of asking why someone cannot stick to a plan, it asks what is happening in the body that makes weight hard to lose and hard to keep off.
What measures
BMI is a ratio of weight to height. Doctors use it as a quick screening number because it is easy to calculate. It does not measure body fat directly, and it cannot tell muscle from fat.
A high BMI is a reason to look further, not a diagnosis. A normal BMI does not rule out health problems either. You can calculate yours with the BMI calculator.
Key points
BMI is a screening number, not a diagnosis.
Weight regulation is biological, which is why this guide avoids blame language.
2. Why weight is hard to keep off
4 minute read.
Your body defends against weight loss
After weight loss, the body adjusts the hormones that control hunger and fullness. In a small study of people who lost weight on a very-low-energy diet, several of these hormones were still changed a year later, in directions that favor regaining weight.
This does not mean weight loss is impossible. It means the pull to regain is a measurable biological effect, not a personal flaw.
Why this matters for the medicines in this atlas
Many of the newer medicines work on the same hormone systems. The rest of this guide builds on that idea: the medicines act on appetite signals, they do not replace the need for a plan, and what happens after stopping is an open question in the evidence.
Key points
Hormone changes after weight loss can persist and favor regain.
This is evidence from a small study, so treat it as an explanation, not a promise about any one person.
After you eat, your gut releases hormones called . The best known is . It helps signal fullness to the brain, slows how fast the stomach empties and raises insulin when blood sugar is high.
Medicines such as semaglutide are . They switch on the GLP-1 receptor, but they are built to last about a week in the body instead of a few minutes.
More than one target
Tirzepatide also switches on the receptor. Retatrutide is being studied with a third target, . Amylin-based medicines, such as the cagrilintide part of CagriSema, add a different fullness signal with .
Researchers are still working out how much each added target contributes. The medication pages say where that is unknown.
Not every new medicine is a peptide
Orforglipron also switches on the GLP-1 receptor, but it is a , which is why it can be a pill.
Key points
These medicines mimic or boost natural gut-hormone signals.
New ones add targets or change the delivery method.
Approval status is checked against official labels where possible.
Under review or in trials
Medicines still in development, including retatrutide and CagriSema, are listed on the pipeline page. An investigational medicine is not available as an approved product.
Key points
Different classes act on different systems.
Pipeline medicines are not approved unless this atlas says so.
In STEP 1, adults taking semaglutide 2.4 mg weekly lost 14.9% of their body weight on average over 68 weeks. The group lost 2.4%. Both groups also got lifestyle support, so the comparison is fairer than looking at one group alone.
An average is not a promise
In the same trial, 86.4% of people on semaglutide lost at least 5% of their weight, compared with 31.5% on placebo. That means some people on the drug lost less than 5%. Group averages cannot tell you what one person will see.
Check the
Some analyses count everyone as randomized, even people who stopped treatment. Others count results only while people were taking the drug. The second kind usually looks larger. Compare trials only when they use the same approach.
Compare numbers across trials with caution. The people enrolled, the length of the trial and the design all differ. See the research page for the figures this atlas has checked.
Key points
Read the comparison group first.
An average hides the spread between people.
Do not compare numbers across trials as if they were head-to-head.
In trials of GLP-1 medicines the most common side effects involve the digestive system, such as nausea and diarrhea. They tend to appear when the dose is raised. In STEP 1, 4.5% of people on semaglutide stopped treatment because of digestive side effects, compared with 0.8% on placebo.
Where the official safety details live
The official label is the controlling source for warnings. Some medicines in this class carry a boxed warning about thyroid C-cell tumors seen in rodents. Each medication page summarizes what the label says and links to its sources.
This guide cannot tell you whether a medicine is safe for you. That depends on your health history, other medicines and pregnancy plans. Your prescriber and pharmacist can.
Key points
Digestive side effects are the most common in trials.
The label, not this guide, is the final word on warnings.
7. Getting care: questions to ask, cost and access
5 minute read.
Questions are more useful than answers here
The Atlas does not pick a medicine for you. It can help you walk into an appointment prepared. The checklist below is a set of questions to bring.
Approved versus compounded
Approved products have been reviewed by the FDA. A is made or altered by a pharmacy for an individual patient, and the FDA does not review compounded drugs for safety, effectiveness or quality before they are sold. The Atlas keeps the two clearly separate and does not rank pharmacies.
Insurance
Some insurers require before they will pay for a medicine. Your prescriber usually submits it. Chapter 12 goes deeper on costs and coverage.
Key points
Bring questions, not conclusions.
Approved and compounded products are not the same thing.
Questions to bring to an appointment
8. Eating well on medication
7 minute read.
What changes when appetite drops
These medicines turn the volume of hunger down. A 2025 advisory from four U.S. nutrition and obesity societies puts the observed drop in calorie intake at 16% to 39%. That is the point of the medicine, and it also creates a practical problem. When you eat much less, each meal has to carry more of the vitamins, minerals and protein your body needs.
The advisory warns that the risk of falling short rises when intake goes below about 1,200 calories a day for women or 1,800 for men. Signs it lists include fatigue beyond what you would expect, a lot of hair loss, itchy or flaky skin, muscle weakness, slow wound healing and unusual bruising. Any of those is worth mentioning to your prescriber.
A caution about this chapter: the advisory is expert consensus, not a trial that tested diets on people taking these medicines. Its authors say direct evidence is limited. The weight-loss trials themselves (STEP, SURMOUNT) all paired the medicine with regular counseling, and that included the placebo groups.
When nausea gets in the way
Digestive side effects are most likely when you start and each time the dose goes up. The advisory describes nausea as often worst in the morning or after a long gap without eating. Its practical suggestions are small, frequent meals, and avoiding fatty or high-fiber foods for the first few days of a new dose. Vomiting is more likely after large meals.
Some people fall into a loop where nausea stops them eating, and an empty stomach makes the nausea worse. For that situation the advisory suggests eating breakfast, then a small meal every 3 to 4 hours with plenty of fluids. It also mentions ginger or peppermint tea and acupressure bands as things some people find helpful. Prescription anti-nausea medicine is another option to raise with your prescriber.
In the clinical trials, people were allowed to stay on one dose for up to 8 weeks before stepping up, to let side effects settle. If nausea is hard to live with, ask whether the schedule can slow down. Do not change the dose on your own.
Protein: the figures experts have proposed
Protein matters more when you are eating less, because it helps protect muscle (more on that in the next chapter). The advisory lays out several reference points. They are proposals for clinicians to work with, not a prescription for you.
The standard U.S. recommended allowance for adults is 0.8 grams per kilogram of body weight per day. For active weight loss, higher targets of 1.2 to 1.6 g/kg per day have been proposed, and for people lifting weights, 1.2 to 1.7 g/kg. The advisory says it is unclear whether those numbers should use your actual weight or an adjusted weight, because actual weight can overstate the need in people with obesity. As a simpler alternative it mentions a fixed daily target of 80 to 120 grams, which may be easier to stick with.
It also marks the edges. Intake should not fall below about 0.4 to 0.5 g/kg per day, and staying above 2 g/kg for long stretches is something to avoid. A registered dietitian can turn these into a number that fits your weight, kidneys and appetite. The protein page sets out what has and has not been tested.
Getting enough is the hard part when food holds little appeal. The advisory suggests eating the protein food first at a meal, and choosing compact options such as fish, eggs, Greek yogurt, cottage cheese, nuts and nut butters. Smoothies, soups and protein drinks work for people who find solid food unappealing. It prefers beans, lentils, dairy, seafood, eggs and lean poultry as everyday sources, with red and processed meats in moderation.
Fluids and your kidneys
Dehydration is the digestive side effect that can turn serious. The current Wegovy and Zepbound labels both include a warning about acute kidney injury. The Zepbound label notes that postmarketing cases were mostly preceded by digestive reactions that led to dehydration, and some needed dialysis. Both labels tell prescribers to monitor kidney function in people whose side effects could cause fluid loss.
In plain terms, if you are vomiting, have diarrhea or cannot keep fluids down, call your prescriber rather than waiting it out. The advisory also lists heart palpitations as a sign of dehydration.
Constipation, diarrhea and alcohol
Constipation is common during weight loss, and the advisory says to deal with it early. It starts with fluids and food fiber. Because protein, fat and fiber slow stomach emptying, it notes that people sometimes need to ease off those foods for a while. When food changes are not enough, it lists daily magnesium (the citrate form is described as well tolerated), fiber supplements, polyethylene glycol 3350 and stool softeners. Which of those suits you depends on your health and other medicines, so ask first. The constipation page has the trial numbers.
Diarrhea is less common and shows up more often with tirzepatide, according to the advisory. Avoiding large or high-fat meals helps, and fiber powders or short-term anti-diarrheal medicine are options for a clinician to weigh.
The advisory also says alcohol can worsen nausea and reflux on these medicines and recommends keeping it low. Its authors also say they see cravings for alcohol, sugar and high-fat food fall in many of their patients, which is a clinical observation rather than a trial result.
Popular eating styles
Many people on these medicines also try low-carbohydrate eating or fasting windows. The advisory comments on both. A very-low-carbohydrate diet can work for some people and be hard to keep up for others. For anyone with type 2 diabetes who takes insulin or a sulfonylurea, it warns that combining that diet with a GLP-1 medicine could raise the risk of low blood sugar and, in some cases, diabetic ketoacidosis, so doses of those other drugs need medical supervision.
On fasting, it points out that many people on these medicines end up fasting by accident because they are simply not hungry. It advises regular meal times, enough protein and a varied diet either way. The eating and exercise section covers these eating styles, with experience kept separate from trial evidence.
Key points
Eating much less means the food you do eat needs to be nutrient-dense.
Small, frequent meals and slower dose changes are the usual first answers to nausea.
Published protein figures are proposals to discuss with a dietitian, not personal targets.
Vomiting, diarrhea or not drinking enough can harm the kidneys, so call early.
Whenever people lose a lot of weight quickly, part of what they lose is not fat. That holds for diets, for surgery and for these medicines. includes muscle, but also bone, organs and water, so losing lean mass does not mean losing the same amount of muscle.
Body-composition measurements from the big trials give a rough picture. The 2025 advisory reports that in STEP 1, of an average 13.6 kg lost, about 8.3 kg was fat and 5.3 kg (38%) was lean mass. Because muscle is roughly half of lean mass, it estimates that about a fifth of the weight lost was muscle.
In the SURMOUNT-1 body-composition substudy, 160 people had a before and after 72 weeks. Those on tirzepatide lost 21.3% of body weight, including 33.9% of their fat mass and 10.9% of their lean mass. About three quarters of the weight lost was fat and one quarter lean mass. The placebo group had almost the same split, losing 5.3% of body weight with 2.6% of lean mass gone. This was a small substudy of the larger trial, and the manufacturer funded it.
What this does and does not tell us
The takeaway from the trials is not that these medicines are unusually hard on muscle. The proportion of fat to lean mass lost looked similar to what happens with other methods that cause large weight loss. What we do not have is good long-term evidence on whether the lean mass lost matters for strength, falls or fractures in everyday life. Older adults and people who start with low muscle have the most reason to pay attention.
The advisory notes that muscle loss varies by sex and activity. It cites modeling that puts muscle at about 10% to 15% of weight lost in women and 20% to 25% in men when no strength training is done.
What helps: strength training
The advisory is firm on one point. Eating more protein alone is probably not enough to protect muscle without structured strength training. , or a mix of resistance and aerobic exercise, is well established as a way to keep lean mass during weight loss. Aerobic exercise on its own has a smaller effect.
The advisory also cites a randomized trial in which one year of a GLP-1 medicine plus exercise training preserved bone density, while the medicine alone reduced it.
The advisory does not give one exercise plan for everyone and neither does the Atlas. A physical therapist or qualified trainer can build a program around your joints, fitness and any health conditions. The exercise and strength training page sorts out what trials, experts and the community say in more detail.
How clinicians check
The advisory lists a few ways to keep an eye on muscle. Bioelectrical impedance scales are cheap and quick but are rough estimates. DXA is more precise. Neither measures how well your muscles work, so strength and function tests such as the sit-to-stand test are useful too. If your clinic offers any of these, ask whether a baseline reading makes sense before you start.
Key points
Some lean mass is lost along with fat during large weight loss, whatever the method.
In the SURMOUNT-1 substudy about three quarters of weight lost was fat and one quarter lean mass.
Experts say protein alone is probably not enough, and strength training is the best-supported tool.
Three sources give the clearest picture. They agree.
In the STEP 1 extension, 327 people were followed for a year after stopping semaglutide 2.4 mg. At the end of treatment they had lost an average of 17.3% of body weight. A year later they had regained 11.6 percentage points of it, about two thirds, and were left with a net loss of 5.6%. Improvements in blood pressure, cholesterol and blood sugar markers drifted back toward where they started. Placebo participants regained 1.9 points of a much smaller loss.
In SURMOUNT-4, everyone took tirzepatide for 36 weeks and lost 20.9% on average. Then half were randomly switched to placebo. Over the next 52 weeks the placebo group regained 14.0% of their body weight, while the group that continued lost a further 5.5%. At week 88, 89.5% of those who continued had kept at least 80% of their loss, compared with 16.6% of those switched to placebo.
A 2026 BMJ review pooled 37 studies and 9,341 people across weight-management medicines. On average people regained about 0.4 kg a month after stopping, and heart and metabolic markers were projected to return to baseline within about 1.4 years. Regain was faster than after stopping a behavioral weight-loss program, by about 0.3 kg a month.
How to read that
This is why many experts describe obesity treatment as long-term, in the same way blood pressure treatment is. The STEP 1 extension authors say their findings support ongoing treatment to keep the gains. The BMJ authors take a more cautious tone and suggest these drugs are best not used for short stretches without a broader plan for weight management.
It also helps to be clear about what the data cannot say. These sources followed people for a year or two at most. They do not tell you how a slower taper, a lower dose or intensive lifestyle changes would compare, because those were not what was tested. Lifestyle support in the STEP 1 extension also ended when treatment did, so that study cannot separate the two.
There are many valid reasons people stop. Side effects, cost, insurance changes, pregnancy and surgery all come up. Stopping is not a failure, but it works better with a plan.
Building a plan before you need one
The 2025 advisory says clinicians should help people build food, activity and sleep habits before and during treatment so they can carry them forward if the medicine stops. It lists the best nutrition approaches for maintaining weight after stopping as an area that still needs study.
Ask your prescriber about two things in advance. One is what the plan would be if you had to stop suddenly, for example because of cost or a shortage. The other is whether a lower maintenance dose is an option for you. The Atlas does not recommend either, and the trials above did not test lower maintenance doses.
Key points
In every large source, weight and health markers moved back toward baseline after the medicine stopped.
SURMOUNT-4: 89.5% of those who continued kept at least 80% of their loss, versus 16.6% who switched to placebo.
Trials did not test slow tapers or lower maintenance doses, so plan with a prescriber.
Weight-loss results in people with type 2 diabetes tend to be smaller than in people without it. In SURMOUNT-2, 938 adults with obesity and type 2 diabetes took tirzepatide or placebo for 72 weeks. Average weight loss was 12.8% on 10 mg and 14.7% on 15 mg, compared with 3.2% on placebo. In SURMOUNT-1, which excluded people with diabetes, the 10 mg and 15 mg groups lost 19.5% and 20.9%.
The bigger practical issue is low blood sugar. The Wegovy and Zepbound labels both say the risk of hypoglycemia goes up when the medicine is combined with insulin or a sulfonylurea, and both suggest considering a lower dose of those other drugs. Never adjust diabetes medicine yourself. The Wegovy label also says people with a history of diabetic retinopathy should be monitored for it getting worse.
Adolescents
The Wegovy injection label covers adolescents aged 12 and older with obesity. The Wegovy tablets are labeled for adults only. The Zepbound label says safety and effectiveness have not been established in pediatric patients.
The evidence for adolescents comes from STEP TEENS. It enrolled 201 young people aged 12 to under 18 with obesity, or with overweight plus a related condition, for 68 weeks of semaglutide 2.4 mg or placebo plus lifestyle support. BMI fell by 16.1% on semaglutide and rose 0.6% on placebo. Digestive side effects were more common on semaglutide (62% versus 42%), and 5 participants (4%) developed gallstones versus none on placebo. This is a decision for a pediatric specialist and the family together, and it is not covered further here.
Pregnancy and breastfeeding
The labels are direct about pregnancy. The Wegovy label says to stop the medicine when pregnancy is recognized, and to stop at least 2 months before a planned pregnancy because semaglutide stays in the body for a long time. The Zepbound label says weight loss offers no benefit to a pregnant patient and may cause fetal harm, and also says to stop when pregnancy is recognized.
Zepbound also has a contraception warning. People using oral birth control are told to switch to a non-oral method or add a barrier method, for 4 weeks after starting and 4 weeks after each dose increase. The Wegovy label text the Atlas read did not include a similar statement, and the two medicines are not interchangeable here, so ask your prescriber and read the label for your own medicine.
On breastfeeding, the Wegovy tablet label says it is not recommended during treatment. The Zepbound label says tirzepatide was undetectable or low in breast milk, and that no data exist on effects on a breastfed infant or on milk production.
Surgery, sedation and other conditions
Both labels carry a note on anesthesia. They report rare cases of food or liquid entering the lungs during general anesthesia or deep sedation in people taking these medicines, even after fasting. The labels say the data are not enough to recommend a specific fix, and tell patients to inform their care team before any planned procedure. Tell the surgeon, anesthesiologist or endoscopy team that you take the medicine, and ask what they want you to do.
The labels also list warnings for gallbladder problems and pancreatitis, so severe stomach pain needs prompt medical attention. Kidney, thyroid and mental-health history all belong in the conversation with your prescriber before you start. The Atlas pages for each medicine summarize its warnings.
Key points
Weight loss is smaller on average with type 2 diabetes, and low blood sugar risk rises with insulin or sulfonylureas.
Only some products are labeled for adolescents.
Labels say to stop in pregnancy, and Zepbound adds a birth-control warning.
Tell every care team about the medicine before sedation or surgery.
Prices and rules for these medicines change often, so the Atlas does not list dollar prices. What stays steady is the way coverage works.
Coverage depends on why the medicine is prescribed, not only which one it is. The same product can be covered for type 2 diabetes and not for weight loss. Many plans require , which means the plan must approve before it pays. Your prescriber usually files it, and plans often set conditions such as a minimum BMI or a weight-related health problem. Ask the prescriber's office what your plan requires before the first prescription is written, and ask what to do if it is denied. Denials can often be appealed.
Because these medicines need to be continued, as the previous chapter explained, it is worth asking what the cost will be a year from now, not only this month.
Medicare's temporary $50 program
Medicare Part D has traditionally not covered medicines used only for weight loss. A temporary bridge program changed that for some people. According to NPR's May 2026 report, it started July 1, 2026 and runs through December 31, 2027. People enrolled in a Part D plan can get Wegovy (pill or injection), the Zepbound KwikPen or Foundayo for a flat $50 a month if they meet the criteria, which are a BMI of 35 or higher, or a BMI of 27 or higher with a condition such as heart disease or prediabetes. Prior authorization is required.
The report adds three limits. The $50 does not count toward the Part D deductible or the annual out-of-pocket cap. Extra Help cannot be used for these drugs. And it applies only to weight-loss use, so people taking them for diabetes stay on regular Part D pricing. What happens after 2027 is uncertain. This is a news report, so check Medicare.gov and your plan for the current rules.
Manufacturer programs and paying yourself
Both Novo Nordisk and Eli Lilly run their own programs, including direct sales and savings offers, and the terms change. Eligibility often depends on whether you have commercial insurance, government insurance or none. The reliable move is to read the current terms on the manufacturer's own site or ask your pharmacist, rather than rely on a number you saw in an article, including this one.
Approved versus compounded
During the shortages, pharmacies were allowed to compound copies of semaglutide and tirzepatide. According to a TechTarget report from April 2025, the FDA decided both shortages had ended and set deadlines in spring 2025 for compounders to wind down those allowances, while keeping the power to act against unsafe or mislabeled products. In March 2026 the FDA announced 30 warning letters to telehealth firms, according to Pharmaceutical Commerce. The letters targeted marketing that suggested compounded versions were the same as approved products, and ads that did not make clear the firm was not the compounder. Rules and court cases have continued to move, so treat both reports as snapshots.
The core point from Chapter 7 still applies. A is not FDA-approved and the FDA does not review it for safety, effectiveness or quality before sale. Doses, ingredients and added substances can differ from the approved product. If a lower price is the reason you are considering one, ask the pharmacy who makes it, what exactly it contains and how it is tested, and ask your prescriber what they recommend. Be especially cautious with products sold online without a prescription.
Key points
Coverage depends on the reason for the prescription as well as the medicine.
Medicare's $50 bridge program runs July 1, 2026 to December 31, 2027, with limits.
Compounded products are not FDA-approved, and the rules for them have been changing.
Check prices and terms on official sites, because they change quickly.
FDA winds down GLP-1 compounding policies TechTarget, April 1, 2025. News report on FDA actions in early 2025; rules may have changed since. Trade press, secondary source
A medicine that switches on a receptor, the docking point a hormone normally uses. A GLP-1 receptor agonist switches on the same receptor GLP-1 does.
amylin
A hormone made in the pancreas along with insulin. It helps slow stomach emptying and adds to the feeling of fullness after a meal.
BMI
Body mass index, a ratio of weight to height. It is a screening number, not a diagnosis, and it does not measure body fat directly.
compounded drug
A medicine made or altered by a pharmacy for an individual patient. The FDA does not review compounded drugs for safety, effectiveness or quality before they are sold.
DXA scan
A low-dose X-ray scan that estimates how much of the body is fat, lean tissue and bone. Short for dual-energy X-ray absorptiometry.
estimand
The exact question a trial analysis answers. One common version counts everyone as randomized, even if they stopped treatment. Another counts results only while people were taking the drug. The two can give different averages.
GIP
Glucose-dependent insulinotropic polypeptide, another gut hormone released after eating. It also helps the pancreas release insulin, and researchers are still working out its role in appetite and body weight.
GLP-1
Glucagon-like peptide-1, a hormone released by the gut after eating. It helps signal fullness, slows how fast the stomach empties and raises insulin when blood sugar is high.
glucagon
A hormone made in the pancreas that tells the liver to release stored sugar. Some investigational medicines also activate its receptor, which may raise energy use.
incretin
A gut hormone released after eating that helps the body handle the meal. GLP-1 and GIP are the two best known.
lean mass
Everything in the body that is not fat, including muscle, bone and organs. Weight loss can reduce lean mass as well as fat.
placebo
An inactive treatment given to the comparison group so researchers can tell the drug's effect from the effect of taking part in a study.
prior authorization
A step some insurers require before they will pay for a medicine. Your prescriber usually submits it.
resistance training
Exercise that makes muscles work against a load, such as weights, bands or body weight. Also called strength training.
small molecule
A medicine built from a simple chemical structure rather than a chain of amino acids. Many pills are small molecules, which is part of how some can be taken by mouth.