January 1, 2024
3 min read
The body mass index is flawed, and medicine now has better options to measure obesity
![Illustration of a woman being given a physical examination, including blood samples, blood pressure and waist measurements](https://static.scientificamerican.com/sciam/cache/file/ED53E1A8-31A5-4BBF-9F7A1F5E0F0B5197_source.jpg?w=600)
According to standard calculations, my husband’s body mass index (BMI) is too high. Yet he is the fittest person I know—an athlete carrying plenty of muscle and very little fat.
Therein lies the problem with BMI. Derived by dividing someone’s weight in kilograms by the square of their height in meters, a BMI number classifies a person as underweight (less than 18.5), normal weight (18.5 to 24.9), overweight (25 to 29.9) or obese (30 or more). But that simple formula obscures critical details such as the difference between muscle and fat. When it comes to individual health risks, those details tell the real story.
The shortcomings of BMI have been recognized for decades. Yet physicians kept using it as a quick way to diagnose obesity and as a proxy for overall health. “It made life really easy,” says clinical psychologist Cynthia Bulik, founding director of the University of North Carolina Center of Excellence for Eating Disorders. “It also led to a type of tunnel vision.”
Equating a slightly high BMI with poor health isn’t always accurate. The American Medical Association acknowledged as much when it announced last June that BMI alone is an imperfect measure and that clinical practice needs to change.
The new thinking on BMI does not negate the need to address the health risks associated with a high body-fat percentage. In 2013 the AMA recognized obesity as a disease and noted that it often leads to many dangerous conditions, including cancer, diabetes and heart disease. The risks of obesity haven’t changed. For the time being, however, insurers still rely on BMI to determine people’s eligibility for bariatric surgery and most weight-loss drugs, including popular new options such as Wegovy.
A recent study showed just how imprecise BMI can be. Yftach Gepner, a physiologist and epidemiologist at Tel Aviv University, and his colleagues looked at data on about 3,000 Israeli men and women. Roughly one third of those whose BMI placed them in the normal range were found to be obese when their actual body fat was measured. And a third of those who were identified as overweight by their BMI had normal amounts of body fat. “If you are combining the misclassification on both sides,” Gepner says, using BMI to determine obesity “is like flipping a coin.”
Not only does BMI fail to distinguish between muscle and fat, but it says nothing about where that fat sits in your body, says Priya Jaisinghani, an endocrinologist and specialist in obesity at N.Y.U. Langone Health in New York City. With fat, as with real estate, location matters. Abdominal fat confers higher risk, as does fat around vital organs. A 2018 study looked at magnetic resonance imaging scans—the gold standard for body-composition research—collected by the U.K. Biobank, a large biomedical database. The results showed that people with fat concentrated in their abdomen had higher risk for type 2 diabetes, heart disease and metabolic disease than did people with the same BMI and of the same age whose fat was spread through other parts of their body.
A further problem is that BMI is based on height and weight tables developed using data from non-Hispanic white people, mostly men. Yet researchers now know that race, ethnicity, sex and age affect body composition and health risks differently. For instance, Black people tend to have greater muscle mass and thus may be misclassified as obese on the basis of BMI. The opposite is true for Asians, who tend to have more body fat at lower BMIs, so their actual disease risks may be missed.
Although BMI has value for assessing obesity at a population level, better methods exist for individuals. “The key is not to use BMI on its own as an index of health,” Bulik says. To properly assess health, doctors should combine the index with measures such as waist circumference, blood pressure and cholesterol levels. Bioelectrical impedance analysis, which uses electric signals to tell fat from muscle in the body, is becoming more common in medical offices. “In a very few years it is going to become standard,” Gepner says. DEXA scans, a type of x-ray imaging that can distinguish between muscle and fat, and MRI also can be used to measure body fat, although they tend to be more costly and are therefore less accessible.
With so many alternatives available, no one, including health-care workers, should give BMI too much weight.
This is an opinion and analysis article, and the views expressed by the author or authors are not necessarily those of Scientific American.