January 18, 2025 - 3:00pm

A new definition of obesity has just dropped. It’s set out in a paper published this week by The Lancet Diabetes and Endocrinology journal. Authored by a “commission” of 58 experts drawn from leading medical institutions in several countries, it’s likely to prove influential. Indeed, it has already been covered by Nature, New Scientist and the BBC. What’s more, the new definition has now been endorsed by the Royal College of Physicians, among others.

In short, this has the potential to change the way that obesity is diagnosed and treated — which matters to millions given that 29% of Britons and 38% of Americans are currently defined as obese. However, we’re in danger of making a big mistake, primarily because the paper’s argument is badly flawed.

It begins by making a perfectly valid point, which is that the standard way of defining obesity — Body Mass Index (BMI) — leaves a lot to be desired. As a simple formula based on height and weight, BMI is a convenient way to categorise individuals and entire populations. Perhaps that’s why health bureaucracies are so stubbornly attached to it.

A key issue is that BMI disregards body type, as if a broad-chested guy and a beanpole of the same height are meant to weigh the same. Athletes of various builds are also messed around, because BMI makes no distinction between muscle mass and adipose tissue — i.e. fat.

So, yes, we need a better framework. Unfortunately, what the Lancet study authors propose really isn’t it. They want a distinction between “clinical obesity” and “preclinical obesity”. The first is defined as a “chronic systemic illness […] due to excess adiposity”; the second is a “state of excess adiposity with preserved function of other tissues and organs”.

This implies that there is a fundamental distinction between the two states. Indeed, the paper goes on to characterise the preclinical/clinical divide as “health vs illness”. Imagine a paper authored by leading medical experts that proposed a similar distinction between preclinical and clinical smoking. It just wouldn’t happen, because there is no state of being a smoker that can be described as healthy.

The same should apply to obesity. Though the paper allows that the preclinically obese are at a “varying, but generally increased risk” of becoming clinically obese, it underplays the links between the two. One doesn’t suddenly flip from one state to the other: rather, the damage done by “excess adiposity” accumulates to the point at which it becomes diagnosable as disease.

The worst thing about the new definition is that it runs counter to a long overdue shift in how society views medicine. It shouldn’t just be about treating people when they get sick, but intervening earlier to halt the process by which most of us are getting sick. This applies as much to poor diet, lack of exercise and obesity as it does to smoking. The last thing we need now is a new definition that reinforces an outdated paradigm.


Peter Franklin is Associate Editor of UnHerd. He was previously a policy advisor and speechwriter on environmental and social issues.

peterfranklin_