Why BMI Is Not the Best Measure of Health

BMI, or body mass index, has long been a part of health evaluations. A calculation that takes into account a person’s weight and height, BMI is supposed to reflect how much excess fat a person might be carrying and give doctors a sense of a patient's risk for obesity-related health conditions.

In a study published Nov. 11 in Nature, researchers led by Francesco Rubino at King’s College London report that a new framework for assessing obesity that combines BMI with other measures can dramatically improve doctors’ ability to determine whether a person’s excess weight is a health concern or not.

The study included more than a decade's worth of data from more than 844,000 people in China, Spain, the U.K., and the U.S. The scientists compared determinations of obesity calculated by either BMI alone or by the new combined framework, which includes BMI along with a person's waist circumference or their waist-to-height ratio. They found that using only BMI missed diagnosing obesity in nearly 9% of people and misdiagnosed 3% as having obesity when their body weight wasn’t an immediate reason for concern.

“The scale of the problem with BMI is beyond what we imagined,” says Rubino. “We all know that BMI doesn’t represent health accurately. But the scale of the problem is concerning because of the number of people it misclassifies, by misdiagnosing in both directions. This is dozens of millions of people in the large countries we studied, and across the world, we’re talking many, many more—so it’s really a concern.”

Even when excess weight is correctly identified, not all experts agree on the appropriate next steps. The new framework, first proposed in 2025 by the Lancet Diabetes and Endocrinology Commission—which Rubino led—argues for a new classification called preclinical obesity, in which patients with excess weight do not necessarily require immediate treatment (or treatment at all). This condition, Rubino and others say, is distinct from clinical obesity, a medical diagnosis that requires more immediate interventions like diet, exercise, or drug options such as GLP-1 medications. In the current study, about 77% to 91% of people who had preclinical obesity did not progress to clinical obesity over 10 years.

In the study, about 5% to 11% of people in the general population were determined to have clinical obesity; their risk of dying during the study was 69% higher than people without excess fat. For those with preclinical obesity, the mortality risk was 19% higher than that of those without excess fat. Distinguishing clinical from preclinical obesity, therefore, “allows doctors to triage and treat people differently, because the two groups are meaningfully different and the difference in mortality is quite obvious,” says Rubino. “All in all, [the study results provide] quite a robust level of evidence to say that the old system of BMI is something that is unsustainable in the clinic.”