Skip to main content
Season 3 ends in 25 d 11:42:31 $2,000 10,000 pts = 1 month Premium
Health & Medicine

BMI vs Waist Size: New Obesity Definition Tested

Published 12 min read
Open the post and read it to claim 30 points
A coiled measuring tape, a glass bathroom scale and a wall height rod in a dim clinic, illustrating BMI vs waist size for diagnosing obesity

Sign in and earn +30 points by reading this

Read to the end, wait a few seconds, then tap the claim button below.

In short

A Nature study published on 11 November 2026 reports that adding waist size and waist-to-height ratio to BMI aligns obesity diagnosis more closely with disease status than BMI alone [1]. Across 844,108 adults in four countries, BMI underestimated obesity in 5.2-8.6% of participants and overestimated it in 1.2-3.1% [1][2].

Key takeaways

  1. The study tested the Lancet Commission's 2025 obesity framework in cohorts from China, the UK, the USA and Spain [1].
  2. Combining BMI with waist circumference and waist-to-height ratio raised obesity prevalence by 3.3-7.4 percentage points versus BMI alone [1].
  3. Clinical obesity, where excess fat already harms organs or daily life, made up 22.7-31.3% of obesity cases [1].
  4. Compared with people with no combined-measure obesity and no organ dysfunction, clinical obesity carried a hazard ratio for death of 1.69 and preclinical obesity 1.19 [1].
  5. 77.4-91.3% of people with preclinical obesity did not progress to clinical obesity over 10 years [1].
  6. The study is retrospective and based on health records; the Commission framework keeps BMI as the first screening step [1].
On this page 12 sections
  1. What did the new obesity study find?
  2. What is BMI and why is it criticised?
  3. What is the new definition of obesity?
  4. How does the new diagnosis work step by step?
  5. Is waist-to-height ratio better than BMI?
  6. How risky is clinical obesity compared with preclinical obesity?
  7. What are the limits of the study?
  8. What does this mean for you?
  9. Common mistakes
  10. Frequently asked questions
  11. Glossary
  12. Sources

What did the new obesity study find?

The study, published in Nature on 11 November 2026, found that BMI alone both over- and under-identifies obesity, and that higher mortality was linked to a definition adding waist measurements but not to BMI-only obesity [1]. The research was led by King's College London; Francesco Rubino of King's is the senior author and An Pan of Huazhong University of Science and Technology in Wuhan the first author [2].

It analysed records from 844,108 adults in four cohorts: the Kailuan Study in China (137,795 people), UK Biobank (436,203), the US All of Us programme (218,841) and Spain's SIDIAP database (51,269) [1].

  • BMI classed 1.2-3.1% of participants as obese although they did not have raised waist measures (the authors call this overestimation) [1].
  • 5.2-8.6% of participants had a BMI in the overweight range but met both waist criteria for obesity-level fat (underestimation) [1].
  • Obesity defined by the combined measures was linked to a modest but higher mortality (pooled hazard ratios 1.06 and 1.07 for overweight and obesity by BMI with combined-measure obesity); people with BMI-defined obesity but no raised waist measures had no higher mortality (0.93) [1].

"Body mass index alone cannot tell us whether someone is already ill or at risk of illness, or what care they need," Rubino said [2].

What is BMI and why is it criticised?

Body mass index is weight in kilograms divided by height in metres squared; for adults, WHO defines overweight as a BMI of 25 or more and obesity as 30 or more [5]. WHO notes that other measures, such as waist circumference, can also be used [5].

The criticism is that BMI is a proxy for body size, not body fat. The Nature authors note that a higher BMI may also reflect greater lean (muscle) mass, while raised waist measures are more likely to reflect fat-mass expansion [1]. Euronews summarises that BMI does not account for how much fat a person has, where it sits or whether it is affecting health [4].

The scale is large. WHO estimates that in 2024, 2.6 billion adults were overweight, including 940 million living with obesity, about 16% of adults [5]. The authors say that even small misclassification rates translate into tens of millions of people in large countries [1]. For a broader look at how health tools are judged, see AI in Healthcare: What It Can Do Today, Where It Fails and How It Is Regulated.

Translucent blue torso with glowing gold belly fat and a gold ring at the waist, illustrating waist size
Waist measures are more closely linked to abdominal fat than BMI, the authors say.

What is the new definition of obesity?

The Lancet Commission framework, proposed in 2025, confirms obesity with body measurements beyond BMI and then separates "clinical" from "preclinical" obesity [1].

  • Clinical obesity: excess body fat that is already impairing organ function or daily activities, an active disease [1][2].
  • Preclinical obesity: excess fat with increased health risk but preserved organ and whole-body function [1].

In the study, clinical obesity meant combined-measure obesity plus at least one of 18 adult diagnostic criteria, covering conditions such as heart failure, atrial fibrillation, high blood pressure, sleep apnoea, liver disease with fibrosis, chronic kidney disease, osteoarthritis and limits on daily activities [1].

The Guardian gives an example: three people with a BMI of 32 could receive three different diagnoses, from no excess fat at all, to preclinical obesity, to clinical obesity [3]. BMI stays the first screening step; the Commission retained it as the initial tool [1].

Two white torso mannequins, one with defined abs and one with a glowing gold belly, showing differing fat distribution
People with the same BMI can have very different amounts of body fat.

How does the new diagnosis work step by step?

Under the framework as applied in the study and described by the Guardian, BMI is still checked first, then waist measures confirm or rule out obesity, and finally organ function decides whether it is clinical or preclinical [1][3].

  1. Very high BMI: someone with a BMI of 40 or more is considered to have obesity without further body measurements [1][3].
  2. BMI of 30 to 39.9: counts as obesity when paired with a raised waist measurement or a raised waist-to-height ratio [1][3].
  3. BMI of 25 to 29.9: counts as obesity only if both waist circumference and waist-to-height ratio are high [1][3].
  4. Second step: clinicians look for signs that excess fat is already impairing organs or daily life, separating clinical from preclinical obesity [1][3].

The researchers followed participants' records for a median of 15.3 years in the Kailuan Study, 13.8 in UK Biobank and 11.2 in SIDIAP; All of Us was left out of the mortality analysis because its follow-up was short [1]. The framework itself was proposed by a Lancet Commission published in The Lancet Diabetes & Endocrinology in January 2025 [4]. Rubino told the Guardian that BMI misclassification "can account for tens of millions of people, so the inaccuracy of BMI is a big problem at scale and also for individuals" [3].

The Guardian notes that, by BMI alone, about 1.6 billion people worldwide are classed as overweight and nearly a billion as obese, and that two-thirds of UK adults and more than 70% of US adults are overweight or obese [3]. WHO's own 2024 estimates are 2.6 billion overweight adults, including 940 million with obesity [5]. Under the new definition, the study found a modest net increase in obesity prevalence, because people were reclassified in both directions [1][3].

Is waist-to-height ratio better than BMI?

In this study, using waist circumference and waist-to-height ratio together with BMI improved obesity classification and risk prediction compared with BMI alone [1]. The authors stress that waist measures are, like BMI, indirect measures of fat, but more closely tied to fat mass [1].

The thresholds used for obesity-level adiposity were [1]:

  • Waist circumference: 102 cm or more for men and 88 cm or more for women (90/85 cm for Chinese participants).
  • Waist-to-height ratio: ethnicity-specific cut-offs, for example 0.57 for non-Hispanic white men and 0.58 for non-Hispanic white women.

The authors found that the conventional waist-to-height ratio cut-off of 0.50 corresponded roughly to the BMI-defined overweight threshold, and that obesity-level values were generally above 0.55 (0.55 to 0.60 across the four cohorts) [1]. These are research cut-offs for diagnosis, not personal targets.

How risky is clinical obesity compared with preclinical obesity?

Clinical obesity was linked to a substantially higher risk of death from any cause (hazard ratio 1.69, 95% CI 1.64-1.73), while preclinical obesity carried a smaller increase (1.19, 1.14-1.24) [1]. These are pooled observational results from three cohorts (Kailuan, UK Biobank and SIDIAP). A hazard ratio of 1.69 means a 69% higher risk over the follow-up period compared with people with no combined-measure obesity and no organ dysfunction.

Clinical obesity accounted for 22.7-31.3% of all obesity, or 5.6-10.9% of the whole population, depending on the cohort [1]. Between 77.4% and 91.3% of people with preclinical obesity at the start showed no progression to clinical obesity over 10 years [1].

"Preclinical obesity should therefore be neither treated automatically as active disease nor dismissed as harmless," Rubino said, adding that some people may need preventive treatment but "its intensity should match their individual risk" [2]. First author An Pan said the consistent differences across countries "support global adoption of the new framework" [2].

Server racks beside a glass world map with glowing points over the US, Europe and China, evoking multi-country study data
Data came from 844,108 adults in China, the UK, the USA and Spain.

What are the limits of the study?

The study is retrospective and relies on electronic health records, which the authors say could limit diagnostic precision and introduce differences between cohorts [1].

  • Organ problems were identified from administrative codes, not standard clinical checks, and not all 18 indicators were available in every cohort [1].
  • The method used to attribute organ problems to obesity works at population level and cannot decide it for an individual [1].
  • The authors call for prospective studies with clinically confirmed diagnoses [1].
  • Other studies that used a single waist-to-height cut-off of 0.50 found much larger increases in obesity, showing that results depend on how the framework is applied [1].

The approach is also more complex for clinicians, the Guardian notes [3]. It is a research validation of a proposed framework, not a change to any guideline, and the framework keeps BMI as the first screening step [1]. Health claims shared online about "the end of BMI" go further than the paper; AI Hallucinations Explained: Why Chatbots Make Things Up and How to Get Accurate Answers explains how summaries can distort findings.

What does this mean for you?

In the proposed framework BMI stays the first screening step, with waist measurements and checks of organ health added before obesity is diagnosed [1][3].

Rubino argues diagnosis should happen at the individual level so care, whether lifestyle advice, medicines or surgery, matches need, telling the Guardian the approach should "absolutely" be used in clinics [3]. Rubino also said people with obesity "deserve care matched to their needs: no less, and no more" [2][4].

If you are unsure what your BMI or waist size means for you, discuss it with a doctor, who can look at your full health picture rather than one number.

This is not medical advice. This article explains a study and does not recommend any weight target or treatment.

Obesity prevalence by BMI vs combined measures (Nature, 11 November 2026)
CohortParticipantsObesity by BMIObesity by combined measures
Kailuan (China)137,79517.8%21.3%
UK Biobank436,20324.0%27.3%
All of Us (USA)218,84142.5%48.1%
SIDIAP (Spain)51,26918.7%26.1%

Common mistakes

  • Saying BMI has been abolished.

    Fix: The framework keeps BMI as the first screening step and adds waist measures [1].

  • Treating a 0.5 waist-to-height ratio as the obesity line.

    Fix: The study found 0.50 corresponded roughly to the BMI overweight threshold; obesity-level values were generally above 0.55 [1].

  • Assuming all obesity is the same disease.

    Fix: Clinical and preclinical obesity had very different death risks [1].

  • Reading observational results as proof of cause.

    Fix: The study is retrospective and based on health records [1].

Frequently asked questions

What did the new obesity study find?

BMI alone underestimated obesity in 5.2-8.6% of participants and overestimated it in 1.2-3.1%; higher mortality was linked to the combined definition but not to BMI-only obesity [1].

What is BMI and why is it criticised?

Weight divided by height squared; it does not show how much fat a person has or where [4][5].

How does the new diagnosis work step by step?

BMI first, then waist measures to confirm obesity, then a check of organ function [1][3].

What is the new definition of obesity?

BMI plus waist measures to confirm obesity, then clinical vs preclinical based on organ function [1].

Is waist-to-height ratio better than BMI?

Used together with BMI and waist size, it improved obesity classification in this study [1].

How risky is clinical obesity compared with preclinical obesity?

Hazard ratio for death 1.69 for clinical vs 1.19 for preclinical obesity [1].

What are the limits of the study?

It is retrospective, uses health records and cannot attribute causes for individuals [1].

What does this mean for you?

In the framework BMI stays the first check, and waist measures and organ health are added before diagnosis [1][3].

Glossary

BMI
Body mass index: weight (kg) divided by height (m) squared [5].
Waist-to-height ratio
Waist circumference divided by height, a proxy for abdominal fat [1].
Clinical obesity
Excess fat that already impairs organ function or daily activities [1].
Preclinical obesity
Excess fat with raised risk but preserved function [1].
Hazard ratio
A comparison of risk over time between groups; 1.0 means no difference [1].

Sources

  1. Validation of clinical and preclinical obesity across global cohorts Nature · 2026-11-11
  2. New clinical definition outperforms BMI in diagnosing obesity King's College London · 2026-11-11
  3. BMI alternative for identifying obesity 'could prevent misdiagnosis in millions' The Guardian · 2026-11-11
  4. BMI alone misses obesity diagnoses, international study finds Euronews · 2026-11-12
  5. Obesity and overweight (fact sheet) World Health Organization · 2026-10-07
Share this guide Telegram WhatsApp X
0 likes

Comments

Top comments

About the Pouyam Team Editorial policy

On Pouyam

Explore Pouyam

Guides are one corner of Pouyam. Here is where to go next.

Quick navigation