Obesity and Metabolic Syndrome

TL;DR. Obesity is excess body fat in an amount that damages health. The arithmetic is energy in versus energy out, and that is where most explanations stop and go wrong, because body weight is not a bank balance, it is a regulated system. The brain defends a weight the way it defends body temperature, using hormones from fat, gut, and pancreas, and it defends the highest weight reached far more vigorously than it resists gaining. That biology explains almost everything that confuses people about obesity: why deliberate weight loss usually stalls and reverses, why willpower is a poor predictor of outcome, and why drugs that act on the brain's appetite circuits have suddenly achieved what decades of advice could not.

Key takeaways

• About 1 in 8 people worldwide live with obesity, and roughly 2.5 billion adults are overweight or obese. Prevalence has more than doubled since 1990.
• BMI is a screening tool for populations, not a diagnosis for individuals. It cannot distinguish muscle from fat and it misclassifies athletes, older adults, and several ancestries.
• Where fat sits matters more than how much there is. Visceral fat around the organs and fat inside the liver and muscle drive metabolic damage; fat under the skin on hips and thighs is comparatively benign.
• After weight loss, the body reduces energy expenditure below what its new size predicts and raises hunger hormones. This metabolic adaptation persists for years and is the main mechanical reason diets fail.
• Obesity raises the risk of type 2 diabetes, cardiovascular disease, at least 13 cancers, fatty liver disease, sleep apnoea, and osteoarthritis. Roughly 5 million deaths a year are attributed to high BMI.
• GLP-1 based drugs changed the field. Semaglutide achieves about 15 percent average weight loss and tirzepatide about 20 percent, versus 5 to 8 percent for intensive lifestyle programmes and 25 to 30 percent for surgery.

What it is

In short: BMI is a population screening tool used as an individual diagnosis, and where the fat sits matters more than how much of it there is.

Obesity is excess adiposity sufficient to impair health. Measuring it is where the trouble starts.

Body mass index (BMI) is weight in kilograms divided by height in metres squared. It was devised in the 1830s by the Belgian statistician Adolphe Quetelet to describe populations, not to diagnose patients, and that is still what it is good for.

Its failures are systematic, not random. It counts muscle as excess weight, so athletes are misclassified. It misses fat gained as muscle is lost with age, so an older person with a "normal" BMI can carry a high fat percentage. And it needs different cut-offs by ancestry: people of South and East Asian descent develop metabolic disease at lower BMI, so many countries use 23 for overweight and 27.5 for obesity in these groups.

Better single measures for individual risk are waist circumference and waist-to-height ratio (keeping waist under half your height is a serviceable rule), because they capture where the fat is. A 2025 Lancet Commission proposed splitting the diagnosis into preclinical obesity (excess fat, organs still working normally) and clinical obesity (excess fat that is already causing organ dysfunction), precisely to stop treating a number as a disease.

Metabolic syndrome is the clustering of the damaging features, diagnosed when three of these five are present: large waist circumference, raised triglycerides, low HDL cholesterol, raised blood pressure, and raised fasting glucose. The cluster is not a coincidence. All five follow from insulin resistance driven by fat in the wrong places.

Don't be confused: obesity and being unfit are not the same thing, and neither is a moral category. Cardiorespiratory fitness independently predicts mortality, and a fit person with obesity has substantially lower risk than an unfit person of the same weight. Weight is one input to health, not a summary of it.

The history

In short: Human genetics did not change; the food supply, portion sizes, and the physical activity built into ordinary life did.

For nearly all of human existence, food scarcity was the design problem, and the metabolic system was tuned to store energy whenever it was available. Obesity was rare and, in many societies, a marker of wealth.

The shift is recent and fast. Global obesity prevalence has roughly doubled to quadrupled since 1975, depending on the age group, with childhood obesity rising fastest. Nothing about human genetics changed in that window. What changed was the environment: cheap refined carbohydrate and vegetable oil, industrially formulated foods engineered for palatability, portion sizes, sugary drinks, cars, desk work, screens, and the near-elimination of incidental physical activity.

Medicine's classification followed slowly. The American Medical Association formally recognised obesity as a disease in 2013, a contested decision: supporters argued it unlocked treatment and insurance coverage, critics argued it medicalised a body size and that many people with high BMI are metabolically healthy. Both points have merit, which is what the 2025 preclinical/clinical split is trying to resolve.

What actually goes wrong

In short: Body weight is defended by a hormonal control system that resists loss far more vigorously than it resists gain.

Weight is regulated, not chosen

Fat tissue secretes leptin in proportion to its mass. Leptin acts on the hypothalamus to reduce hunger and permit energy expenditure. The stomach secretes ghrelin, which rises before meals and drives hunger. The gut releases GLP-1, PYY, and CCK after eating, which signal fullness. The pancreas releases insulin and amylin. The hypothalamus integrates all of it and adjusts appetite and energy use to defend a weight.

The system is asymmetric. It defends strongly against loss and weakly against gain, which makes evolutionary sense in a world of periodic famine and is a catastrophe in a world of continuous abundance.

Leptin's discovery in 1994 created enormous excitement: mice lacking leptin were grossly obese, and giving them leptin normalised them. Giving leptin to humans with common obesity did almost nothing, because people with obesity already have high leptin levels and have become leptin resistant. The brain reads the signal as absent. That failure was the field's most instructive dead end and it reframed obesity as a disorder of signalling rather than of storage.

Metabolic adaptation, the reason diets fail

Lose weight and three things happen. Resting energy expenditure falls, by more than the smaller body accounts for. Ghrelin rises and satiety hormones fall, so hunger increases. And muscle becomes more efficient, burning fewer calories for the same movement.

The clearest demonstration came from following contestants of the television programme The Biggest Loser. Six years after extreme weight loss, most had regained much of the weight, and their resting metabolic rates remained suppressed by hundreds of calories a day relative to predictions for their body size. The defence did not fade with time.

This is why "eat less, move more" is technically correct and practically insufficient. The behaviour is being fought by a control system that treats the loss as a threat.

What the environment contributes

A tightly controlled inpatient randomised trial by Kevin Hall's group at the US National Institutes of Health fed participants either ultra-processed or minimally processed diets, matched for calories, sugar, fat, fibre, and macronutrients, with unlimited eating. On the ultra-processed diet, people ate about 500 more calories a day and gained weight; on the minimally processed diet they lost it. Same person, same nutrients on paper, opposite outcome. Something about the form of the food (energy density, texture, eating rate, and how little it satisfies) drives intake independently of its nutrient label.

Other contributors with real evidence: short sleep (raises ghrelin, lowers leptin), chronic stress and cortisol, several drug classes (antipsychotics, some antidepressants, corticosteroids, insulin, sulfonylureas), and endocrine conditions such as hypothyroidism and Cushing's syndrome, which are rare but treatable.

Genes

Obesity is 40 to 70 percent heritable by twin studies, which does not mean fated (Chapter 15 explains why heritability is not a personal percentage). Most of it is polygenic: hundreds of variants, each small, mostly acting on brain appetite circuits rather than on metabolism. The FTO variant, the best known, shifts average weight by only a couple of kilograms.

Rare single-gene forms exist and are informative: leptin deficiency, leptin receptor mutations, and MC4R mutations, the last being the commonest monogenic cause at roughly 1 to 5 percent of severe early-onset obesity. Children with these have relentless hunger from infancy. Setmelanotide, a drug that activates the MC4R pathway downstream of the defect, works dramatically in specific genetic subtypes and not at all in common obesity.

The honest summary: genes largely determine who gains weight in an obesogenic environment, and the environment determines how many people do.

What it does to the body

In short: The damage is metabolic, mechanical, and neoplastic, with at least 13 cancers linked to excess body fat.

Fat tissue is an active endocrine organ, not padding. Visceral fat drains directly to the liver and secretes inflammatory cytokines. The consequences fall into three categories.

Metabolic

• Insulin resistance and type 2 diabetes. Obesity is the dominant modifiable risk factor; the mechanism is in Chapter 18.
• Metabolic dysfunction-associated steatotic liver disease (MASLD), formerly called non-alcoholic fatty liver disease. Fat accumulates in liver cells, in some people progressing to inflammation (MASH), then fibrosis, cirrhosis, and liver cancer. It is now among the leading causes of liver transplantation.
• Dyslipidaemia: high triglycerides, low HDL, and small dense LDL particles.

Mechanical

• Obstructive sleep apnoea: fat in the neck and tongue collapses the airway during sleep, causing repeated oxygen drops, unrefreshing sleep, daytime sleepiness, and raised blood pressure. It is common and frequently undiagnosed.
• Osteoarthritis, mostly of knees and hips, from load plus inflammatory signalling.
• Gastro-oesophageal reflux, and hernias.

Vascular and neoplastic

• Hypertension, coronary disease, heart failure, atrial fibrillation, and stroke.
• Cancer. The International Agency for Research on Cancer identifies sufficient evidence linking excess body fat to at least 13 cancers, including oesophageal adenocarcinoma, colorectal, postmenopausal breast, endometrial, kidney, liver, pancreatic, gallbladder, ovarian, thyroid, gastric cardia, meningioma, and multiple myeloma. Mechanisms include insulin and IGF-1 signalling, oestrogen produced by fat tissue, and chronic inflammation.

Plus reduced fertility in both sexes, pregnancy complications, and higher rates of depression, with causation running in both directions and stigma contributing to the mental health burden independently of the physiology.

Is it deadly?

Yes, indirectly and at scale. High BMI is estimated to contribute to roughly 5 million deaths a year, overwhelmingly through cardiovascular disease, diabetes, and cancer. Severe obesity (BMI above 40) is associated with a reduction in life expectancy on the order of 8 to 10 years, comparable to lifelong smoking.

Two caveats keep this honest. The relationship with mortality is J-shaped, and the lowest-risk BMI in older adults is higher than in the young. And a subset of people with obesity show no metabolic abnormality for decades, though the risk of transitioning into metabolic disease over time remains higher than for people without obesity.

Is it contagious?

No. Obesity is not an infection and cannot be transmitted.

There is one genuinely interesting finding that is often misreported as contagion. In the Framingham social network analysis (Christakis and Fowler, 2007), obesity appeared to cluster through social ties, with a person's chance of becoming obese rising when a close friend did. The proposed mechanism is social: shared norms about portion size, activity, and acceptable body weight. The study has been criticised on methodological grounds (shared environments and friend selection are hard to separate from influence). It is a story about culture, not transmission.

The gut microbiome is likewise sometimes described as making obesity "transmissible" because transferring gut bacteria between mice transfers some weight phenotype. In humans the effect is far smaller and not established as causal.

Who gets it

In short: One person in eight worldwide, rising fastest in low- and middle-income countries, and socially patterned in opposite directions depending on national income.

• Scale. In 2022, about 2.5 billion adults were overweight, of whom about 890 million had obesity: 1 in 8 people alive. Over 390 million children and adolescents aged 5 to 19 were overweight, including 160 million with obesity.
• Fastest growth is now in low- and middle-income countries, particularly in urban areas, producing the double burden of undernutrition and obesity coexisting in the same country and sometimes the same household.
• Highest prevalence is in Pacific Island nations (exceeding 50 percent of adults in several), the Gulf states, and the United States (around 40 percent of adults).
• Income gradient. In high-income countries, obesity is more common among poorer people, tracking the cost of calories per dollar, food retail environments, working hours, and neighbourhood walkability. In low-income countries the gradient is often reversed, though it flips as countries get richer.
• Sex and age. Prevalence is somewhat higher in women globally, with wide regional variation. It rises through middle age and declines in the very old.

Treatment, and how it works

In short: Lifestyle programmes achieve 5 to 8 percent, GLP-1 based drugs 15 to 22 percent, and surgery 25 to 30 percent sustained over a decade.

Treatment intensity is matched to risk, not to appearance. The goal is health outcomes, and the amount of weight loss needed differs by outcome: about 3 to 5 percent improves glucose and triglycerides, 5 to 10 percent improves blood pressure and sleep apnoea, and 10 to 15 percent or more is where diabetes remission and MASH improvement appear.

Lifestyle programmes. Structured, supported programmes with regular contact produce average losses of 5 to 8 percent at a year, with substantial regain thereafter unless support continues. That average conceals wide variation: some people do far better. The key evidence point is that intensive programmes work considerably better than advice alone, and that maintenance contact is what preserves the result.

Medications

These drugs are covered in full in Chapter 66, including how they were engineered, all five of their mechanisms, and the complete safety picture: the vomiting and dehydration pathway, the deaths linked to compounded products, the anaesthesia risk, and why they are not a substitute for insulin.

The GLP-1 class matters beyond the numbers for two reasons. It works on the regulatory system itself, lowering the defended weight rather than fighting it, which is why people describe hunger and "food noise" simply receding. And the SELECT trial showed semaglutide reduced major cardiovascular events by about 20 percent in people with obesity and established cardiovascular disease but without diabetes, moving these drugs from cosmetic framing to outcome-modifying therapy.

Metabolic (bariatric) surgery. Sleeve gastrectomy and Roux-en-Y gastric bypass produce 25 to 30 percent weight loss sustained over a decade or more, with high rates of diabetes remission and, in long-term cohort studies, reduced mortality. The mechanism is only partly restriction; changed gut hormone signalling (GLP-1 and PYY rise sharply after bypass) and altered bile acid handling do much of the work, which is why glucose improves within days of surgery.

What treatment costs

• GLP-1 and dual agonists: nausea, vomiting, constipation, diarrhoea, all dose-related and usually improving. Gallstones with rapid loss. Rare pancreatitis. Loss of lean mass alongside fat, which matters most in older patients and is an active research target. Weight regain on stopping is substantial and rapid, because the drug was lowering the defended weight rather than curing anything, which makes these long-term therapies rather than courses. Cost and supply remain major access barriers.
• Orlistat: oily stools and urgency if fat intake is not reduced, plus reduced absorption of fat-soluble vitamins.
• Surgery: operative mortality is low in experienced centres (comparable to gallbladder surgery) but not zero. Long-term requirements include lifelong vitamin and mineral supplementation (B12, iron, calcium, vitamin D), risk of internal hernia and dumping syndrome, and a documented increase in alcohol use disorder after bypass.

What the person can do

In short: Protein, fibre, resistance training, sleep, and cutting liquid calories, undertaken with the expectation that the body will fight back.

Framed honestly, because this is a field saturated with confident bad advice.

• Protein and fibre first. Both increase satiety per calorie, and protein preserves muscle during weight loss, which protects resting metabolic rate.
• Cut liquid calories. Sugary drinks are the single most consistently implicated dietary item, because liquid calories produce little satiety.
• Reduce ultra-processed food where feasible. The Hall trial gives this a causal basis rather than a moralistic one.
• Resistance training, not only cardio. Exercise is a weak weight-loss tool on its own (it is easy to eat back the deficit) and a strong tool for preserving muscle, improving insulin sensitivity, and maintaining loss.
• Sleep 7 to 9 hours, and get snoring with daytime sleepiness assessed. Treating sleep apnoea improves energy, blood pressure, and often the ability to be active.
• Self-monitoring (weighing regularly, tracking intake) is among the most consistent predictors of maintenance in long-term registries.
• Expect defence, and plan for it. Weight loss plateaus are physiology, not failure. Maintenance is an active, ongoing task rather than a finish line.
• Ask about medication-driven gain. If weight rose sharply after starting a drug, alternatives often exist.

Living with it

Weight stigma is itself a health hazard. People with obesity report discrimination in healthcare, employment, and education, and experimental work shows clinicians spend less time with and give less information to patients with higher BMI. The measurable results are avoidance of medical care, delayed diagnoses, and worse mental health. Stigma does not produce weight loss; the evidence points the other way, toward increased eating and avoidance of exercise settings.

Weight cycling (repeated loss and regain) is common and demoralising, and the evidence on whether it independently harms health is mixed. What is clear is that repeated failure of a strategy that is fighting a regulatory system should update the strategy rather than the person's self-assessment.

What's next

• Oral GLP-1 drugs at doses matching injections, removing the needle barrier.
• Amylin analogues such as cagrilintide, alone and combined with semaglutide, in trials showing loss beyond current agents.
• Muscle-sparing combinations, pairing weight-loss drugs with myostatin or activin pathway agents to preserve lean mass.
• Better subtyping: identifying who has hunger-driven, satiety-driven, or emotional eating patterns, and matching the treatment.
• Policy: sugar-sweetened beverage taxes have measurably reduced purchases in several countries, and reformulation rules and marketing restrictions to children are where population-level change, as opposed to individual treatment, is likely to come from.

Sources and notes

Prevalence figures are WHO (2022 data, published 2024): 2.5 billion adults overweight, 890 million with obesity, 1 in 8 people, and the child and adolescent counts as given. Deaths attributable to high BMI are Global Burden of Disease estimates and vary by method. The Lancet Commission on Clinical Obesity (2025) proposed the preclinical/clinical distinction. Leptin: Zhang et al., Nature, 1994. Metabolic adaptation after major weight loss: Fothergill et al., Obesity, 2016 (Biggest Loser follow-up). Ultra-processed food trial: Hall et al., Cell Metabolism, 2019. Semaglutide weight loss: STEP trials (Wilding et al., NEJM, 2021). Tirzepatide: SURMOUNT-1 (Jastreboff et al., NEJM, 2022). Cardiovascular outcomes: SELECT (Lincoff et al., NEJM, 2023). Obesity-cancer links: IARC Handbooks of Cancer Prevention, volume 16. Social network clustering: Christakis and Fowler, NEJM, 2007, with subsequent methodological criticism. Asian BMI cut-offs: WHO expert consultation, 2004.

Open questions. Whether ultra-processed food harms through composition, texture, additives, or simply energy density is unresolved. Long-term (decade-plus) outcomes of GLP-1 therapy, including what happens to bone and muscle, are not yet known. Whether weight cycling independently harms health remains debated.

Next: the risk factor that kills more people than any other, and that almost nobody can feel. 👉