A figure on the scales does not, on its own, tell us what proportion of a person’s weight is fat, or whether an organ or daily activities are affected. A study published on 11 November in “Nature” examines what a more complete assessment of obesity changes, using data from 844 108 adults. Its findings point to the need to distinguish more clearly between screening, adiposity and clinical disease, without making BMI a useless tool.
BMI is a guide, not a complete assessment
Body mass index, or BMI, is calculated by dividing weight in kilograms by the square of height in metres. For adults, the World Health Organization defines overweight as a BMI of 25 or more and obesity as a BMI of 30 or more. These thresholds are useful for identifying trends and organising screening. However, they do not directly measure fat mass and do not, on their own, describe how the body is functioning.
Two people with the same BMI can therefore have different body compositions. A large amount of muscle can contribute to a high BMI without representing the same amount of fat. Conversely, a BMI in the “overweight” category can mask a build-up of abdominal fat. Waist circumference and its ratio to height provide other indicators, but they too are indirect measures.
The distinction matters for a practical reason: a statistical category is not enough to determine a person’s health status. It can help decide whether a more detailed examination is appropriate, without establishing on its own whether excess adipose tissue is already affecting the body’s functioning.
Four cohorts, more than 844 000 adults
The study published in “Nature” brings together data already collected in four cohorts from China, the United Kingdom, the United States and Spain. The researchers analysed 844 108 adults aged 18 to 74. This reanalysis concerns populations selected according to the availability of their data, and people who had already had a stroke, heart attack or cancer at the start of follow-up were excluded.
The authors compared a classification based on BMI alone with an approach that adds waist circumference and the waist-to-height ratio. This approach is part of the framework proposed in 2025 by a “Lancet” commission: confirm excess adiposity using additional measurements, then examine whether it is accompanied by functional impairment.
The additional measurements changed the classification in both directions. In the same data from the Chinese cohort, for example, the proportion of adults classified as having obesity rises from 17,8 % with BMI alone to 21,3 % with the combined criteria. This comparison shows how the choice of measurement framework can change the identification of those concerned. It applies only to this cohort and the thresholds used.
The central finding, then, is not that one measure replaces another. BMI serves as a starting point; waist circumference and its ratio to height can make the assessment more precise, particularly when the first indicator does not reflect how fat is distributed. The thresholds used in the study vary between populations. This method cannot be summed up as a single rule to apply at home.
From excess fat to clinical obesity
The framework under study distinguishes between two situations. Preclinical obesity is excess adiposity without established functional impairment. Clinical obesity requires, in addition to this excess, signs of organ dysfunction or limitations in daily activities attributable to adiposity. The second step therefore requires more than a waist measurement or a figure on the scales: it relies on a clinical assessment.
Attributing such effects requires caution. A disease can coexist with excess fat without necessarily being caused by it. In the study, the researchers used a statistical estimate at group level to take this difficulty into account. It cannot determine, for any given person, whether a particular impairment is due to their adiposity. Medical records and the indicators available also varied between cohorts.
Over the ten-year follow-up, between 77,4 % and 91,3 % of participants initially classified as having preclinical obesity did not progress to the clinical category in the cohorts with this length of follow-up. This describes the observed progression of a subgroup, not an absence of risk throughout life or a reason to forgo follow-up. In the mortality analysis, clinical obesity was associated with a higher risk than preclinical obesity. These are statistical associations between groups; they do not predict an individual’s future and do not, on their own, demonstrate an individual cause.
What the findings change in the medical discussion
For readers, the value of this approach is that it helps frame the questions to ask during an assessment: is BMI enough in this case? Do the additional measurements indicate excess adiposity? Is there organ impairment or difficulty with daily activities, and what is causing it? The answers require a professional assessment that takes the overall situation into account rather than relying on a single threshold.
The study does not automatically turn the “Lancet” commission’s framework into a definition adopted everywhere. Its findings support the value of a more nuanced classification, but the authors highlight limitations linked to retrospective analysis, administrative records and the diversity of indicators between cohorts. Prospective studies using standardised clinical diagnoses will help clarify the model’s scope.
The WHO also points out that overweight and obesity are linked to biological, social and environmental factors. A body measurement tells us neither about a person’s living conditions nor about their health as a whole. Making better use of BMI therefore means putting it into the context of an assessment, not reducing diagnosis to a number or passing judgement on the people concerned.
References
- Study published in “Nature”: validation of the clinical and preclinical obesity framework
- World Health Organization: overweight and obesity
- L’essentiel: overview of the study and its findings
A flexible tape measure. Illustrative photograph.
