Can ‘Body Roundness Index’ Replace BMI? That Depends, Say Experts

During the 2024 Olympic Games in Paris, France, Ilona Maher was a breakout star, an athletic marvel who possessed an astonishing combination of speed and strength. As the 27-year-old member of Team USA’s women’s rugby sevens rushed downfield, stiff-arming opponents into oblivion, it was unfathomable to think that Maher, and other athletes like her, could be considered overweight or obese.

And yet on TikTok, Maher revealed that she weighs 200 pounds, and at 5 feet 10 inches tall, that makes her overweight by traditional metrics, specifically the body mass index (BMI).

Obesity specialist Fatima Cody Stanford, MD, MPH, MBA (Massachusetts General Hospital/Harvard Medical School, Boston, MA), said BMI—which tells physicians about height and weight—reveals very little. “With an athlete, BMI might be elevated, but they’re carrying very little central adiposity—so that’s not very helpful to me,” she told TCTMD. Another person might carry all of their weight in the midsection, with visceral adiposity wrapped around vital organs, and yet the BMI is “normal” given their height, she said.

In that case, “BMI totally misses the boat,” leaving the door open for other metrics that might offset BMI’s known limitations, said Stanford.

One new anthropomorphic metric currently having a moment is the body roundness index (BRI). Developed a little over 10 years ago by mathematician and obesity researcher Diana Thomas, PhD (United States Military Academy, West Point, NY), BRI takes body weight out of the equation. Instead, waist circumference (or hip circumference) is measured along with height and plugged into a formula to estimate the amount of visceral adipose tissue.

BRI ranges from 1 to 20, with 1 being a narrow body and 20 being round.

“When we started discussing some of the limitations of BMI, we asked what other ways are there to quantify body shape and how could those specific shapes better predict health risks,” said Steven Heymsfield, MD (Pennington Biomedical Research Center, Baton Rouge, LA), who worked with Thomas to develop BRI. “Then [Thomas] came up with this idea that only a mathematician could think of, which is basically the idea of thinking of the body being a 3D elliptical shape, instead of using BMI, which resolves people into a much simpler shape, more or less a cylinder.”

Since its development in 2013, studies have linked BRI to risks of metabolic syndrome, type 2 diabetes and prediabetes, hypertension, colorectal cancer, osteoarthritis, and depression. Most recently, researchers showed that a higher longitudinal trajectory of BRI was associated with an increased risk of cardiovascular disease. Another found a U-shaped association between BRI and all-cause mortality among nearly 33,000 US adults.

Apples vs Pears

This all begs the question: is it times to ditch BMI for BRI?

Despite the latter overcoming some of the drawbacks of the former, experts say it’s not that simple, or maybe even necessary, to swap out BMI for BRI in clinical practice. One advantage of BMI is that height and weight are simple measurements, and most people know their numbers off the top of their head, said Heymsfield.  

“BMI is an excellent screening point,” he said. “It’s the entry node into a discussion and evaluation of obesity. If you pass through that gate, there’s no question that adding waist circumference or hip circumference improves the differentiation of shape into the risk that we’re concerned about.”

If the BMI is normal, Heymsfield said there is little justification for a full weight/obesity evaluation.

“If BMI is borderline or high, then getting a waist circumference and thus BRI is indicated,” he said. “BRI would then establish risk and level of importance in losing weight and implementing lifestyle measures. This approach is more or less what has been in place for quite a while. The problem is that very few clinical settings/clinicians follow it.”

For Stanford, BRI doesn’t solve all the problems of BMI.

“With the body roundness index, we get to look at where the weight is distributed,” said Stanford. “If you carry it in your midsection, kind of an android distribution, that’s when we get that high risk of cardiometabolic disease. Carrying the weight around your gut, this visceral adipose tissue can be wrapped around your organs like your liver or heart. However, BRI alone can’t distinguish whether it’s visceral or subcutaneous adipose tissue.”

Stanford said both visceral and subcutaneous adiposity are problematic. Still, visceral fat, deep in the abdominal cavity surrounding the stomach, liver, intestines, and other vital organs, is more metabolically active and is associated with more significant health risks. For this reason, BRI alone is not a good tool for identifying disease risks in individuals with android fat distribution, she said.

A History of Healthy, White Males

As Heymsfield noted, BMI’s sheer simplicity has helped cement its role, despite long being recognized as an imperfect, crude tool for assessing obesity.

Its use dates back to 1832 when Belgian statistician, mathematician, and astronomer Adolphe Quetelet, looking for a way to characterize “normal” Belgian soldiers, proposed weight divided by height squared. The Quetelet index was later adopted by the Metropolitan Life Insurance Company, who wanted to establish tables for normal weights of the mostly white men and women they insured. In 1972, physiologist Ancel Keys, PhD, introduced the modern term, body mass index, in a study of 7,000 healthy middle-aged men.

Let's say we captured everyone in the world with a BMI of 25, how much variation would there be in their percent fat? Steven Heymsfield

“The simplest way to understand BMI is if everybody in the world was the same height, then weight would be a measure we could use to differentiate people in terms of their adiposity,” explained Heymsfield. The problem is, if two people are the same weight but one is muscular and one is not, BMI can’t make that distinction because there’s a built-in assumption that everybody carries the same proportion of adiposity, he said.

“Let's say we captured everyone in the world with a BMI of 25, how much variation would there be in their percent fat?” continued Heymsfield. “There's an amazingly large variation, and that variation is caused largely by differences in muscle mass.”

In 2023, the American Medical Association (AMA) formally recognized the issues around BMI as a measurement, noting that its development was based on data primarily collected from previous generations of white men. The AMA concluded the metric is not the best tool to assess obesity across multiple groups because it doesn’t account for different racial/ethnic backgrounds, sexes, genders, and ages. While BMI correlates with the amount of fat mass in the general population, that relationship breaks down at the individual level, according to the medical group.

Instead, the AMA said BMI should be used alongside other valid measurements, such as assessments of visceral fat, body adiposity index, body composition, relative fat mass, waist circumference, and others.

Midsection Adiposity

There are several steps to consider when encountering a patient with midsection adiposity, said Stanford. These include a careful clinical assessment and patient history as well as the use of other anthropometric measurements. BMI and BRI are useful, she said, but waist circumference and waist-to-hip measurements should be considered as they provide better insights into abdominal fat distribution and associated risks.

If necessary, imaging with MRI or CT can differentiate between visceral and subcutaneous fat, though these are not commonly used in clinical practice due to costs and accessibility, said Stanford. The evaluation of metabolic biomarkers (fasting glucose, lipids, and inflammatory biomarkers) and calculation of cardiovascular risk should also be performed.

The key thing with regard to all of this is to evaluate the patient and recognize that there is no cookie-cutter approach to the individual in front of you. Fatima Cody Stanford

With gynoid adiposity—the so-called pear shape—fat tissue is distributed in hips, buttocks, and thighs, areas of the body without vital organs. It’s here that BRI is a much better predictive measure of health outcomes than BMI, said Stanford.

However, it’s still important to “thoroughly evaluate overall health and risk factors using similar steps [as with android adiposity],” she said.

Other Metrics Being Used

Stanford said BMI is still a ubiquitous surveillance metric, noting that it is automatically calculated in most patient charts. She, however, measures waist circumference in all her patients.

“I think the key thing with regard to all of this is to evaluate the patient and recognize that there is no cookie-cutter approach to the individual in front of you,” she said. “This is where we often fall short in medicine, trying to extrapolate one cookie-cutter approach to a set of patients and expecting the same outcomes without recognizing the heterogeneity of each patient.”

This includes thinking not only about the obesity assessment but also response to various treatments, whether it’s lifestyle interventions, surgery, or medical therapy with glucagon-like peptide 1 receptor agonists or other agents, she said.

To TCTMD, Heymsfield noted that some European countries, including the United Kingdom, recommend that BMI be supplemented with the waist-to-height ratio to get a better picture of the distribution of body fat. A waist-to-hip ratio exceeding 0.5, for example, provides an estimate of greater central adiposity and can help predict higher risk of type 2 diabetes, hypertension, and cardiovascular disease, according to the National Institute for Health and Care Excellence. 

Heymsfield said primary care clinics are pretty lax when it comes to the basics, including taking blood pressure correctly. From his own experience as an obesity specialist, Heymsfield joked that he’s been asked to step on a scale wearing his coat and shoes while carrying a briefcase.

“Then you say, can you also measure a waist circumference or a hip circumference?” he said. “There have been a lot of studies in healthcare showing that offices don't really quantify circumferences very accurately, so we’d need to have a transformation in that setting.”