The letter arrives in the mail, or it lands in an app. You go to a neighbourhood clinic, fast overnight, give blood, step on a scale, and get your waist measured with a tape. Two weeks later the results come back, and somewhere on the second page sits a single word: 비만. Obese. You are the same size you were in London or Chicago or Sydney, where nobody ever called you that. Welcome to the Korea obesity paradox, a statistical trapdoor that catches thousands of foreign residents every year and almost none of them understand why.
Here is the short version. South Korea draws the line for obesity at a body mass index of 25. The United States, the United Kingdom, and most of the rest of the world draw it at 30. That five-point gap is not a rounding error. Instead, it is the difference between a country that looks lean by global standards and a country where roughly one adult in three carries an official diagnosis.
What follows is not a story about Korean bodies. Rather, it is a story about a line, who drew it, who wants to move it, and what happens to a whole population when a number shifts by two points.
Korea runs one of the most aggressive population screening programmes on earth. Every adult enrolled in the National Health Insurance Service gets a free general health checkup, usually once every two years. That includes foreign residents holding a residence card. Eligibility follows your birth year. The screening is fully covered, and it takes about an hour.
That checkup measures height, weight, waist circumference, and BMI, alongside blood pressure, blood sugar, cholesterol, and kidney function. Consequently, the results sheet does not just hand you raw numbers. It sorts you into a category.
Those categories come from the Korean Society for the Study of Obesity. Its 2022 clinical guidelines set out six bands:
| Band | BMI |
|---|---|
| Underweight | below 18.5 |
| Normal | 18.5 – 22.9 |
| Overweight | 23.0 – 24.9 |
| Class I obesity | 25.0 – 29.9 |
| Class II obesity | 30.0 – 34.9 |
| Class III obesity | 35 and above |
Notably, a separate abdominal-obesity threshold applies at a waist of 90 centimetres for men and 85 for women.
Read that table again with Western eyes. In most of the world, a BMI of 24 is comfortably normal and 26 is merely overweight. In Korea, 24 is already overweight and 26 is stage-one obesity. As a result, someone who has never once thought about their weight can leave a Korean clinic holding a diagnosis. This is where the Korea obesity paradox begins.
Seoulz has covered Korea’s healthcare model in detail elsewhere. The screening programme is one of its real strengths. The classification attached to it, however, is where the argument starts.
Now consider the same population measured two ways.
Under Korea’s BMI-25 rule, the Korea Herald reported that 34.4 percent of adults were obese in 2024. The source is the Korea Community Health Survey, which covers roughly 230,000 people. That figure was 26.3 percent in 2015 and 32.2 percent in 2021. Meanwhile, men reached 41.4 percent against 23.0 percent for women. Men in their thirties and forties posted 53.1 and 50.3 percent respectively.
Apply the global BMI-30 threshold to the same country, though, and the picture inverts. Korean research in Endocrinology and Metabolism put class II and class III obesity at just above 7 percent in 2021. That band covers everyone at BMI 30 or higher. Roughly one adult in fourteen. By that measure, the Korea obesity rate would rank among the lowest in the developed world.
Same bodies. Same year. Two answers that differ by a factor of five.
This is the heart of the Korea obesity paradox. It also explains a confusing comparison that circulates in Korean media. The KDCA notes that Korea sits far below the OECD average of 56.4 percent. Yet that OECD figure counts everyone from BMI 25 upward, which is exactly Korea’s obesity threshold. In other words, the Korean obesity standard for disease matches the OECD’s definition of overweight-or-worse. The comparison is technically valid and rhetorically useless.
The threshold was not invented in Seoul. In 2000, the World Health Organization’s Western Pacific office proposed lower cut-offs for Asian populations, and Korea adopted them. As the WHO’s own guidance makes clear, BMI is a population screening tool rather than a diagnosis of individual health.
The rationale was real, not arbitrary. East Asian populations tend to develop metabolic disease at lower body mass than European populations. That includes type 2 diabetes, hypertension, and dyslipidaemia. Body-fat distribution and visceral fat at a given BMI both differ. Therefore, the argument runs, the Western line would let many genuinely at-risk people slip through undiagnosed.
Japan reached the same conclusion independently and also uses 25. China went the other direction and uses 28. So the “correct” number varies across three neighbouring countries with broadly similar body types. That alone tells you how settled the science really is.
Before anyone argues about where to draw the line, there is a more basic problem. Korea does not have one obesity number. It has two, produced by two government surveys, and they do not match.
The 34.4 percent figure comes from the Korea Community Health Survey. It asks 230,000 people to report their own height and weight. The peer-reviewed analysis appears in the Public Health Weekly Report, which also maps regional variation. Jeonnam and Jeju sit at the top. Sejong records the lowest rate at 29.1 percent.
Then there is the Korea National Health and Nutrition Examination Survey, which actually measures people. Its 2024 results, reported by the Korea Herald, found 48.8 percent of men and 26.2 percent of women obese. Men in their forties hit 61.7 percent. That was an eleven-point jump in a single year, and sixteen points above 2015. Women in their twenties came in at 20.4 percent. Women over seventy topped the female chart at 38.3 percent.
So which is it — 41.4 percent of men, or 48.8 percent? Both are official. Both came from the same agency. The gap comes down to sample size, methodology, and one small universal habit. People shave a couple of kilograms off the number they write down.
Korea debates its obesity threshold to two decimal places. The underlying measurement, however, is looser than the argument suggests. Furthermore, an annual jump of 11.5 percentage points among men in their forties invites questions about sampling noise. The survey covered 10,000 people. Nobody in the threshold debate talks about this much.
In November 2024, the argument stopped being academic.
The Health Insurance Research Institute sits inside the NHIS. It presented a study that followed 8.47 million adults for twenty-one years. Participants had taken national health checkups in 2002 and 2003. Researchers then tracked mortality and cardiovascular disease against BMI. As reported by Korea Biomedical Review, the conclusion was blunt. The Korean obesity standard should be raised to at least 27.
The data behind that call is worth understanding. Mortality risk followed a U-shaped curve. Its lowest point sat at a BMI of 25 — precisely the number Korea uses to define disease. After excluding deaths within five years of baseline, risk ran 1.72 times higher at BMI 18.5 or below. At 35 or above it ran 1.64 times higher. Both were measured against 25.
Disease risk told a similar story with different timing. Hypertension, diabetes, and dyslipidaemia showed meaningful increases starting around BMI 27. Cardiovascular disease climbed from 29. Cerebrovascular disease climbed from 31.
Professor Oh Sang-woo of Dongguk University Ilsan Hospital framed the historical shift directly. Twenty years ago, he noted, mortality bottomed out at a BMI of 23. Korean body shapes, diets, and disease patterns have since drifted closer to Western norms. The low point has moved to 25. Lee Sun-mi, who directs the institute’s Health Management Research Center, added that China already applies 28.
Three days later, the Korean Society for the Study of Obesity issued a statement. It read less like a scientific comment than a territorial claim. Korea Biomedical Review covered the response. It opened by objecting that the proposal came without consulting the society or related organisations. That move, the KSSO said, would confuse the public and damage public health.
On substance, the KSSO made three arguments.
First, mortality is the wrong yardstick. Death rates vary by age, smoking status, baseline health, cause of death, and follow-up length. Comorbidity incidence, by contrast, rises steadily with BMI. Therefore, if the point of diagnosis is to prevent complications, comorbidity should anchor the threshold.
Second, the risk starts early. Compared with normal weight, type 2 diabetes risk rose 1.55 times at the pre-obesity stage. At class I obesity it rose 2.46 times. Additionally, the society flagged that obesity in South Korea is climbing fastest among people aged 20 to 50. That is the group where early intervention matters most.
Third, the rest of the world is moving the other way. The European Association for the Study of Obesity had just revised its guidelines. The new version captures patients at BMI 25 who also show a waist-to-height ratio above 0.5 or an existing comorbidity. That effectively lowered its practical threshold from 27. Raising Korea’s line, the society argued, would run against that current.
Both sides are reading real data. However, they are answering different questions. The NHIS asked when extra weight starts killing people. The KSSO asked when it starts making them sick. Those two curves do not peak in the same place, and no amount of additional data will make them.
Korea’s neighbours offer a useful reframe. One of them solved the problem without touching the number at all.
Japan uses the same BMI-25 threshold as the Korean obesity standard. Crucially, though, the Japan Society for the Study of Obesity separates measurement from diagnosis. As set out in JASSO’s guidelines, a BMI of 25 or more makes you obese as a descriptive category. Yet obesity disease — the thing doctors actually treat — requires more. You must also present with one of eleven specified health disorders, or a visceral fat area of 100 square centimetres.
That distinction matters enormously in practice. A Japanese man at BMI 26 with clean bloodwork is categorised, not diagnosed. Add hypertension and fatty liver, though, and he becomes a patient. Meanwhile, Korea’s system hands both of them the same word.
China took the simpler route and set its threshold at 28. That is roughly where the NHIS study found cardiovascular risk beginning to accelerate. Neither approach is obviously superior. Nevertheless, both show that “BMI 25 or nothing” is a choice rather than a scientific inevitability.
While Korea argued over 25 versus 27, a much larger shift was already underway. It has since reframed the entire Korea obesity paradox.
On 14 January 2025, a Lancet Diabetes & Endocrinology Commission published a new framework. Led by Professor Francesco Rubino, it demoted BMI to a screening step rather than a diagnosis. More than 75 medical organisations endorsed it. The framework splits obesity into two conditions.
Preclinical obesity means excess body fat without organ dysfunction. It is a risk factor, monitored but not treated as illness. Clinical obesity means excess body fat that is actively impairing organs or daily function. That version qualifies as a standalone disease. Diagnosis first confirms excess adiposity through waist circumference, waist-to-height ratio, or direct body-fat measurement. Clinicians then check against 18 specified criteria for adults and 13 for children.
Notice what that framework does to the Korean debate. It renders the entire 25-versus-27 argument secondary, because under the Lancet model no single BMI number diagnoses anyone. Similarly, it vindicates Japan’s structure, which had been drawing that line for years.
Korea has not ignored this. The KSSO’s 2024 clinical practice guidelines introduced a clinical-obesity concept. They also urged clinicians to look past anthropometry alone. Even so, the national screening programme still prints a category based on a single number. The society’s 2025 Obesity Fact Sheet still reports prevalence at BMI 25. Guidelines have moved faster than infrastructure.
There is an uncomfortable question hovering over all of this. A low threshold does not just describe a population. In addition, it creates a market.
Set the line at 25 and roughly 34 percent of Korean adults become potential customers. Set it at 27 and that number falls sharply. Korea’s weight-management economy is enormous. Seoulz has documented the $5.5 billion “lifelong homework” diet industry, built on the assumption that most people need fixing. Around it sit the zero-sugar and high-protein product boom, the gym membership sector, and a vast network of clinics.
To be clear, there is no evidence that the KSSO’s position is commercially motivated. Its scientific argument about comorbidity onset is legitimate and well documented. Still, a diagnostic threshold that defines the size of an industry deserves scrutiny. In Korea, it rarely gets any.
Then there is the strangest wrinkle in the entire system. Korea diagnoses obesity at BMI 25 — but it does not treat it there.
Wegovy and Mounjaro now reshape weight management worldwide. In Korea, both are prescribable at a BMI of 30 or above, or 27 to 30 with an obesity-related comorbidity. Neither is covered by national insurance for weight loss, so patients pay out of pocket. Seoulz examined the resulting market in its report on Korea’s GLP-1 boom. Prescriptions there have exploded past 800,000 in a single quarter.
Sit with the arithmetic. An adult at BMI 26 is officially obese and formally recorded as such on a government health file. That same adult is ineligible for the leading treatment. The country tells millions of people they have a disease. Then it tells them they do not have enough of it to treat.
None of this means Korea’s obesity trend is imaginary. On the contrary, the underlying movement is real and it is steep.
Male obesity has risen across every measurement system and every survey. Men in their forties, whichever dataset you trust, are now the heaviest cohort in Korean history. Furthermore, the drivers are well documented. They include binge drinking, chronic short sleep, long sedentary hours, and a delivery-food network without global equal.
Abdominal obesity has climbed steadily too, and it concentrates in exactly the same demographic. That measure is the least sensitive to threshold arguments, because a waistline is a waistline. The peer-reviewed 2025 fact sheet documents the chronic-disease burden that follows. Meanwhile, the socioeconomic cost of obesity in South Korea was put at roughly 9 trillion won in 2016. That was double the 2006 figure, and the trajectory since has not improved.
Meanwhile, the opposite problem runs alongside it. Roughly 15 percent of Korean women aged 19 to 29 are underweight. Among normal-weight women in their twenties, 28.3 percent perceive themselves as obese. Only 6.9 percent of men the same age do. More than half of normal-weight young women had attempted to lose weight.
Korea therefore runs two body crises at once, in opposite directions, in different demographics. A single BMI threshold speaks usefully to neither.
If you live here, some practical translation helps. The Korea obesity paradox is easier to handle once you know which lines on the sheet actually matter.
Your Korean BMI category is not your home country’s category. Add roughly five points before comparing. A Korean class I obesity result maps onto what most Western systems would call overweight. That does not make the reading meaningless, though it does make direct comparison misleading.
Take the waist number seriously. Waist circumference and waist-to-height ratio predict metabolic risk better than BMI, particularly for East Asian bodies. Both the Lancet framework and the EASO guidelines lean on them. A waist above 90 centimetres for men, or 85 for women, matters more than the BMI band printed beside it.
Read the bloodwork, not the label. Fasting glucose, HbA1c, triglycerides, HDL, blood pressure, and liver enzymes tell you whether excess weight is actually doing damage. That is precisely the distinction Japan formalised and the Lancet Commission adopted. Your Korean checkup already includes most of these markers.
Muscle mass distorts the reading. BMI cannot distinguish muscle from fat, so athletic builds get flagged routinely. Body-composition analysis resolves this in minutes. It is available at most Korean gyms and clinics, and it is one reason the country’s medical and wellness sector leans so heavily on body scanning.
Do not shop for a threshold. The wrong response here is deciding your numbers do not count because a foreign standard is looser. Korean researchers adopted BMI 25 for a defensible reason. The metabolic risk it captures at lower body weights is real.
Two points of BMI. That is the entire distance between the government’s insurance researchers and the country’s obesity specialists. Yet those two points would reclassify millions of people overnight. They would redraw the size of a multi-billion-dollar industry. They would also change how a nation talks about itself.
The most striking fact about the Korea obesity paradox is how rarely it gets discussed as a choice. The 34.4 percent headline circulates as though the underlying number were a fixed property of Korean bodies. So do the international comparisons built on it. It is not. Instead, it is an artefact of a decision made in 2000 and defended in 2024. A global framework quietly overtook it in 2025.
Korea will probably resolve this the way it resolves most institutional disagreements — slowly, and then all at once. The 2024 KSSO guidelines already gesture toward clinical obesity. The screening infrastructure will eventually follow. For now, though, a foreign resident in Seoul can hold a paper that calls them obese. Their doctor back home would call them fine. Both documents are telling the truth about a body that never changed.
The number moved. The person did not.
It usually starts as indigestion. That is the detail cardiologists repeat most often. It is…
A stroke gives you about four and a half hours. After that, the treatments that…
It is 11:58 p.m. in Hongdae, and the queue at the convenience store counter has…
The Night the Lights Went Out For fifty years, the sea off Ulleungdo glowed after…
Walk into any Korean supermarket in winter and you will find a chilled case of…
Korean Car YouTubers: Watchdogs of a 92% Monopoly A new Hyundai lands in Korean showrooms…