Adding a waist measurement to BMI did a better job of spotting obesity that raises the risk of death, a study of 844,000 adults finds
People whose BMI put them in the obese range but who had normal waists showed no extra risk of dying. People labeled merely overweight who had large waists did.
Checking waist size along with body mass index, the standard height-and-weight score, did a better job of finding people whose body fat raises their risk of dying, according to a study of 844,108 adults published Nov. 11 in the journal Nature.
Body mass index, or BMI, is weight divided by height squared. A BMI of 30 or more usually counts as obesity. But BMI can’t tell fat from muscle. Waist measurements are more likely to reflect extra fat, the authors wrote. The study tested a definition proposed in 2025 by a Lancet Commission, an expert panel organized by the journal The Lancet Diabetes & Endocrinology, that confirms obesity with waist measurements as well as BMI.
What it found
- BMI alone missed people with risky body fat. Between 5.2% and 8.6% of people in the four groups studied had an “overweight” BMI but large enough waists to count as obese under the new definition.
- BMI alone also wrongly flagged some people. Between 1.2% and 3.1% had an obese-range BMI but normal waist measurements.
- Obesity by BMI alone, without a large waist, was not linked to a higher risk of death. Obesity confirmed by waist size was, even in people whose BMI was only in the overweight range.
- The new definition splits obesity into two stages. “Clinical obesity” means extra fat is already harming the body, through one of 18 problems such as heart failure, sleep apnea or knee arthritis. It carried a 69% higher risk of death. “Preclinical obesity,” extra fat with no such harm yet, carried a 19% higher risk.
- Most people with preclinical obesity, between 77% and 91% depending on the group, did not move on to clinical obesity over 10 years.
What readers can check
You can check the two waist measures the study used with a tape measure and, for the second one, a calculator. The study used them to tell whether a high BMI reflects extra body fat.
Measure 1: waist size. The study counted a waist as large if it was at least 102 centimeters (about 40 inches) for men or 88 centimeters (about 35 inches) for women. For Chinese adults, the cutoffs were 90 centimeters (about 35½ inches) for men and 85 centimeters (about 33½ inches) for women.
Measure 2: waist-to-height ratio. This compares your waist with your height. Measure both in inches, then divide your waist by your height. For example, a man 70 inches tall (5 feet 10 inches) with a 40-inch waist divides 40 by 70 and gets about 0.57. That means his waist is a little more than half his height. The study’s cutoffs for a large ratio depend on sex and race or ethnicity:
- White men 0.57; white women 0.58.
- Black men 0.55; Black women 0.57.
- Hispanic men 0.57; Hispanic women 0.59.
- Asian men 0.56; Asian women 0.59.
A shortcut: multiply your height in inches by the number for your group to get your cutoff waist size. For a white woman 64 inches tall (5 feet 4 inches), 64 times 0.58 is about 37, so a waist of 37 inches or more is large by this measure.
Putting it together with BMI. Under the definition the study tested:
- A BMI of 40 or more counts as obesity by itself, with no waist check needed.
- A BMI of 30 to 39.9 counts as obesity only if at least one of the two waist measures is at or above its cutoff.
- A BMI of 25 to 29.9, the range usually called overweight, counts as obesity only if both waist measures are at or above their cutoffs.
For example, take that same white woman, 64 inches tall, with a BMI of 32 and a 36-inch waist. Her waist is over the 35-inch waist-size cutoff but under her 37-inch ratio cutoff. Because her BMI is between 30 and 39.9, one large measure is enough, so she would count as having obesity. If her BMI were 28 instead, she would need both measures to be large, so she would not.
A common rule of thumb says a waist should be less than half of height, a ratio of 0.50. That lower cutoff flagged far more people, but those flagged only by it showed no extra risk of death, the authors wrote. These measures are screening clues, not a diagnosis. A doctor decides whether extra fat is harming organs.
How the study was done
Researchers led by Huazhong University of Science and Technology in Wuhan, China, and King’s College London analyzed health records from four large groups: the Kailuan Study in China, the UK Biobank, the U.S. National Institutes of Health’s All of Us Research Program and the SIDIAP primary-care database in Catalonia, Spain. They tracked deaths for 11 to 15 years in three of the groups; the U.S. group was left out of the death analysis because its follow-up was too short.
Limits
The study looked back at existing records, so it shows links, not cause and effect. Health problems were identified from billing codes and records, not standardized exams, and not every group had data on all 18 problems. No industry money paid for the study, but several authors have been paid consultants or speakers for drugmakers including Novo Nordisk and Eli Lilly, which make obesity drugs.