Crunching Numbers on the Definition of Clinical Obesity
Six months have passed since the Lancet Commission on the definition of clinical obesity aimed to “settle the ongoing dispute around the idea of obesity as a disease.” At the time, we had doubts about this new definition “settling” disputes about defining obesity. Indeed, we have seen lots of vigorous discussion, but not a lot of settling.
In fact, two new journal articles crunch the numbers on the Lancet definition of clinical obesity and bring us radically different answers on the prevalence of clinical obesity.
In Lancet Diabetes & Endocrinology, a new paper by Zhiqi Yao et al estimates the prevalence of clinical obesity among U.S. adults to be 35.5%. This is not terribly different from the current CDC estimate of 40.3% using BMI alone.
But a new paper in Obesity gives us very different numbers. Yun Shen and Gang Hu estimate the prevalence of clinical obesity to be only 9.7%.
Uncertainty and Ambiguity
These very different answers come from very different methods for obtaining them. The estimate from Yao et al used data from the All of Us Research Program. Shen and Hu used data from NHANES. Both approaches are reasonable and perhaps there is much to learn from these diverging results.
But one thing is abundantly clear. The controversies about how best to define obesity have not been settled. Instead, we are left with a great deal of uncertainty and ambiguity to resolve. Yao et al describe this quite well:
“Although the Commission introduced a clinically oriented definition of obesity, in our opinion, its practical application remains unclear, creating uncertainty for researchers, clinicians, and policy makers. Our sensitivity analyses suggest that roughly a quarter of individuals could shift between clinical and preclinical obesity depending on the criteria used for metabolic disorders and physical impairments. This ambiguity complicates treatment decisions, policy development, and health-care resource allocation, and sparks controversy, as it remains uncertain whether the new definition will expand or restrict access to obesity treatment.”
The process of working out what all this means has barely begun.
Click here for the paper by Yao et al and here for the paper by Shen and Hu. For perspective on the Lancet Commission from its leader, Francesco Rubino, click here.
Iridescent Interpenetration, painting by Giacomo Balla / WikiArt
Subscribe by email to follow the accumulating evidence and observations that shape our view of health, obesity, and policy.
July 20, 2025

July 21, 2025 at 3:08 am, Mary-Jo said:
I like that attention is being given to finding composite measurements of obesity reflecting the complexities of the disease. For epidemiological purposes, the importance of a measureable parameter cannot be overstated and at least BMI helps us to identify population prevalence, track incidence, helps to point out where in the world obesity is occurring the most (or least) and to, at least, start looking further, where to start allocating efforts resources to. Maybe it is time to use a composite measure, even for that, like the BMI+waist or BMI+WHR composite measurement to pick up adiposity. For treatment priorities and purposes, perhaps a more comprehensive composite ‘score’ can be developed, adding other metrics — comorbidities, disabilities, future risks, genetic profile, etc. — to guide clinicians more easily identify who and where to start delivering best care more timely and efficiently.
July 29, 2025 at 7:42 pm, Eddy said:
This article brings clarity to a very misunderstood topic. The way it breaks down the complexities of defining clinical obesity with data-backed reasoning is both informative and eye-opening.