November 11, 2026

expert reaction to study looking at a new clinical definition versus just BMI in diagnosing obesity

Scientists react to study published in Nature* looking at a new clinical definition versus just BMI in diagnosing obesity.

Dr Marie Spreckley, Weight Management Researcher, Prevention of Diabetes and Related Metabolic Disorders in High Risk Groups, Institute of Metabolic Science, University of Cambridge, said:

“This is a substantial and carefully conducted study that provides important evidence for moving beyond BMI alone when assessing obesity.  A particular strength is the inclusion of more than 840,000 adults across four countries, with longitudinal follow-up allowing the researchers to examine how different classifications relate to longer-term health outcomes.  The findings suggest that BMI alone does not adequately capture differences in health risk, and that incorporating measures of abdominal adiposity and organ function provides a more informative assessment.

“Importantly, the researchers have attempted to distinguish organ dysfunction associated with obesity from conditions that may arise independently.  This is a methodological strength, although their statistical approach cannot establish whether obesity caused organ dysfunction in individual participants.  The reliance on existing health records, differences in how conditions were assessed across cohorts, and the potential for residual confounding are also important limitations.

“While the findings support the prognostic value of distinguishing clinical from preclinical obesity, some of the observed differences in mortality risk are unsurprising, given that existing organ dysfunction forms part of the definition of clinical obesity.  The study also does not establish that adopting this framework will improve patient outcomes.  Further prospective research is needed to determine how reliably these classifications can inform treatment decisions and whether they lead to better care in practice.

“The implications are nevertheless important.  Recognising that people with obesity have different health risks and support needs could help move care towards a more individualised approach.  However, the distinction between clinical and preclinical obesity must not become a barrier to preventive treatment.  People without established organ dysfunction may still benefit from timely support, and decisions about care should reflect their individual health risks, circumstances and preferences rather than diagnostic categories alone.”

—

Comments from our friends at the German SMC:

Prof. Dr Alexandra Kautzky-Willer, Head of the Gender Medicine Unit, Department of Endocrinology and Metabolism, Clinic for Internal Medicine III, Medical University of Vienna, Austria:

COI: “No conflicts of interest.”

On the current study

“Four very large and diverse datasets from China, the UK, the USA and Spain were analysed to examine whether the new Lancet definition of obesity, based on additional anthropometric parameters such as waist (waist circumference; Editor’s note) or WHtR (waist-to-height ratio; Editor’s note) can better predict mortality over a ten-year period. Furthermore, a distinction was made between pre-clinical and clinical obesity based on information included in the datasets regarding obesity-related organ dysfunction. Overall, the study supports the call to use additional anthropometric parameters to determine obesity. Similarly, most professional societies and experts highlight the importance of fat distribution and the adverse effects of visceral fat in particular, which can be better estimated using waist circumference or WHtR.”

Weight-to-Height Ratio

“For the WHtR, new ethnicity-specific cut-off values were determined instead of the usual cut-off of 0.5 – taking into account the BMI cut-off for obesity in the respective group. Although this new classification showed a clearer association with mortality rates, in my opinion this should first be verified in other cohorts. It would also be important to investigate whether sex-specific cut-off values for the WHtR might be more sensitive.”

Body fat mass

“Furthermore, it would be beneficial to have actual measured values for fat mass as a reference, which was not possible here. With retrospective datasets, there is always the problem that parameters were measured and documented differently and are incomplete. In the study, comorbidities and organ dysfunction not caused by obesity itself were adjusted for population-related factors – to what extent this is possible in principle seems questionable to me, as causality is rarely proven and correlations are usually indirect. Temporal relationships can only be reliably assessed in prospectively collected datasets.”

“In this respect, a prospective longitudinal cohort study measuring body fat mass across different ethnic groups and for both men and women would be important. In postmenopausal women in particular, a BMI below 30 underestimates actual body fat mass. For a large percentage of them, this would already be in the 35 per cent range – that is, in the obese category.”

Added value for individual groups

“Overall, the study also stratifies data by sex, though this is not discussed in further detail. It appears that hip circumference – which already had different cut-off values for men and women, as well as for different ethnic groups – would define the highest rate of obesity. Particularly among the Chinese and Spanish populations, there are marked differences here compared with the prevalence defined by BMI alone.”

“In the Women’s Health Initiative study, in which women were followed for over 20 years, the assessment of mortality risk was improved by using BMI combined with waist circumference. Compared with women whose weight and waist circumference were within the normal range, women of normal weight but with a larger waist circumference had a 17 per cent higher mortality rate [1]. In this respect, it would be interesting to investigate whether the inclusion of anthropometric parameters might also be useful within a lower, normal-weight BMI range. Particularly in women, I consider hip circumference to be a more important cardiometabolic risk marker than BMI.”

GLP-1-based medicines

“Furthermore, the influence of medicines, particularly GLP-1-based medicines, on the progression of obesity and mortality could not be investigated. It is reasonable to assume that significant differences will emerge in future, depending on the use of these medicines.”

Clinical and preclinical obesity

“Between a quarter and a third of obese individuals had clinical obesity, which was also associated with a 70 per cent higher mortality rate over ten years. The Lancet definition of obesity combined with clinical obesity was associated with higher mortality in both sexes, but particularly in women. I am surprised that, in the group with preclinical obesity, at least three-quarters did not develop clinical obesity over a ten-year period, but this was not systematically investigated. However, mortality was also almost 20 per cent higher in the group with preclinical obesity. It is at least pointed out that preclinical obesity must also be taken seriously from a medical perspective, and treatment is recommended depending on the risk profile and individual circumstances. After all, our aim must be to prevent long-term damage rather than waiting to treat it until irreversible consequences have already set in. We must invest much more in prevention and treat even overweight individuals with an increased waist circumference, WHtR or directly measured fat mass on an individual basis. ”

Added value and additional effort

“In my view, the additional effort involved in the new diagnostic approach is entirely justified. Measuring hip circumference and the WHtR is very simple and inexpensive. Body fat measurement, on the other hand, is primarily carried out in the context of scientific research. Sometimes it is also carried out on an individual basis, depending on the availability of a measurement method and following a cost-benefit analysis.”

[1] Aragaki AK et al. (2025): Development and Validation of Body Mass Index–Specific Waist Circumference Thresholds in Postmenopausal Women: A Prospective Cohort Study. Annals of Internal Medicine. DOI: 10.7326/ANNALS-24-00713.

—

Comments from our friends at the Spanish SMC:

Diego Bellido, Head of the Endocrinology and Nutrition Department at the Ferrol Hospital Complex (CHUF) and President of the Spanish Society for Obesity (SEEDO), said: 

Is the study of good quality? Are the conclusions supported by sound data?

“It is a robust study, involving more than 844,000 participants from four countries, including Spain. Its results support the view that combining measures of adiposity with clinical assessment allows for a better distinction to be made regarding prognosis. However, it does not yet prove that using this classification improves treatment or reduces complications. The proposal to adopt it universally requires further evidence.”

How does it fit in with existing evidence?

“It reinforces something we have been advocating for years: BMI is useful, but insufficient for assessing a person with obesity. Fat distribution and its impact on health are crucial. What is new is the provision of follow-up data supporting the distinction between clinical and preclinical obesity, although the debate on how to use it remains open.”

Are there any significant limitations?

“Yes. It is an observational study and clinical information was not collected in a standardised manner across all cohorts. Furthermore, the attribution of certain health conditions to obesity was estimated statistically, without individual confirmation. This means the progression figures must be interpreted with caution: the fact that a person does not develop clinical obesity according to these criteria does not mean they remain free from health problems.”

What are the advantages of incorporating other body measurements and organ function rather than just BMI?

‘BMI does not distinguish between fat and muscle, nor does it reflect where fat is accumulated. Measuring waist circumference and assessing organ function helps to better understand each patient’s situation. Two people with the same BMI may have very different risks and treatment needs.’

How important is it to distinguish between clinical and preclinical obesity?

“It can help to tailor the intensity of monitoring and treatment. When organ damage is already present, there is a greater urgency to act. When it is not detected, we must focus on prevention and assess individual risk. The ‘preclinical’ category should not be interpreted as the absence of risk, nor should it be used to delay appropriate treatment.”

Could this lead to over-diagnosis or over-medicalisation? What implications would it have for prescribing medication?

“There is a risk if a new diagnostic label automatically translates into a prescription. However, these criteria can also correct imprecise diagnoses based solely on BMI. The prescription of medicines must be based on the expected benefit, the risks and the patient’s circumstances. This study does not determine who should receive them nor does it alter their authorised indications.”

Are they applicable in routine clinical practice?

“Supplementing BMI with waist circumference and a clinical assessment is feasible and should form part of good care. The greatest challenge is to rigorously establish which abnormalities are due to adiposity and to do so using standardised criteria. Their incorporation requires training, time and straightforward protocols, without turning the assessment into an indiscriminate battery of tests.”

He does not state whether he has any conflicts of interest.

Ana Belén Crujeiras Martínez, Director of the Epigenomics Research Group within the Endocrinology and Nutrition Group at the Santiago Health Research Institute (IDIS) and a member of the CIBER Network on the Pathophysiology of Obesity and Nutrition (CIBERobn), said: 

“This is a high-quality article, as it evaluates a large number of patients from different countries. The analysis carried out provides solid evidence that body mass index, which is widely used in clinical practice, does not correctly classify patients with obesity, leading to overestimation or underestimation. It is proposed that additional measures be used, such as the waist-to-height ratio and other measures related to organ function, which provide a more accurate classification and enable a treatment strategy better tailored to each patient’s individual situation. These measures are easily applicable in routine clinical practice and would help to avoid the over- or under-diagnosis of obesity.

The study distinguishes between clinical and preclinical obesity, demonstrating that patients classified as having preclinical obesity have lower mortality rates and that only a small number of those with preclinical obesity progress to clinical obesity within 10 years. This classification would enable the establishment of a preventive treatment strategy for patients with preclinical obesity and help determine the urgency of treatment, without implying inaction in the management of these patients. However, the fact that organ function was not assessed using biochemical or molecular measures constitutes a limitation of the study and may lead to an underestimation of the true clinical status. In this regard, I believe that for the future of obesity classification, in addition to incorporating various anthropometric and body composition measures, it would be necessary to carry out biochemical and molecular measurements in order to adequately quantify organ function. In fact, there is an epigenetic memory in adipose tissue that indicates dysfunction in that tissue, and it is possible that this is not being adequately assessed in patients classified as having preclinical obesity, who may present biological abnormalities that we are not yet detecting using conventional diagnostic tools.

The major challenge is not merely to classify obesity more effectively, but to identify biological dysfunction at an early stage so that we can anticipate its complications and move towards truly personalised medicine”.

She does not state whether she has any conflicts of interest.

—

* ‘Validation of clinical and preclinical obesity across global cohorts’ by An Pan et al.  published in Nature on Wednesday 11 November 2026

DOI: 10.1038/s41586-026-11237-5

Declared interests

Dr Marie Spreckley: “Dr Marie Spreckley is a Senior Research Fellow and Programme Lead at the Precision Healthcare University Research Institute at Queen Mary University of London.  She has no financial relationships with food or beverage companies, pharmaceutical companies, or other commercial organisations, and has received no industry funding for her research.”

For all other experts, no reply to our request for DOIs was received.