rewrite this title Study Suggests BMI Can’t Predict Your Risk of Death—But This Other Metric Can

Yet another study has found that body mass index, or BMI—a measure of body fat based on height and weight—is a poor indicator of overall health.

The report, published in the Annals of Family Medicine in June, found that BMI is unreliable at predicting a person’s risk of death. But according to the researchers, there’s a more effective measurement that can be easily assessed during routine checkups: body fat percentage, or the proportion of your body that’s made up of fat.

That metric is determined via bioelectrical impedance analysis, a technique that sends a painless electrical current through the body to estimate the proportions of body fat, lean muscle mass, and water.

Here’s what to know about the growing concerns around BMI, and why doctors say body fat percentage may be a smarter metric.

Because body composition is linked to a higher risk of type 2 diabetes, cardiovascular disease, and early death, physicians have traditionally relied on BMI to evaluate a person’s fat, bone, and muscle health.

Calculated by dividing weight in kilograms by height in meters squared, BMI can offer some sense of a person’s health, said Wajahat Mehal, MD, DPhil, the director of the Yale Metabolic Health and Weight Loss Program.

“The higher it is, the more likely that it will be associated with metabolic diseases,” he told Health. 

But growing evidence suggests BMI is a flawed indicator because it doesn’t account for age, biological sex, or the distribution of fat and muscle.  

For example, elite athletes often have high lean muscle mass and elevated BMIs, according to Shiara Ortiz-Pujols, MD, MPH, the director of obesity medicine at Northwell Health’s Staten Island University Hospital. “These athletes are healthy, but if the focus is solely on BMI, they would be mislabeled as having obesity,” she told Health. 

On the flip side, some people with “normal” BMIs may still have a higher risk of diabetes, hypertension, and nonalcoholic fatty liver disease, Arch Mainous, PhD, lead author of the study and a professor in the Department of Community Health and Family Medicine at the University of Florida College of Medicine, told Health.

“These people are sometimes called ‘normal weight obesity’ or, colloquially, ‘skinny fat,’” he said.

With the limitations of BMI in mind, researchers from the University of Florida Health set out to investigate how strongly the measure is linked to mortality. They also wanted to know whether body fat percentage—“an easy, reliable, and inexpensive measure of body composition,” they wrote—could better predict the risk of death.

The team analyzed health data from 4,252 adults between the ages of 20 and 49. The data, collected between 1999 and 2004, included height, weight, waist circumference, and body fat percentage (BF%). Researchers calculated BMI and tracked mortality through 2019 using death certificate records.

After adjusting for multiple factors, including age and race, the scientists concluded that:

There is no significant link between a BMI categorized as obese—defined as 25 kg/m2 or higher—and a higher risk of death from any cause compared to those with a “healthy” BMI. 
BF% offers insight into mortality risk. Those with a high BF%—27% or higher for men and 44% or higher for women—were 78% more likely to die from any cause.
Waist circumference is less predictive than BF% but still associated with a higher risk of death.

According to Mainous, the study “shows that a direct measure of body fat percentage that is relatively inexpensive and easy to obtain in a doctor’s office is superior in predicting downstream mortality risk versus the indirect measure of body fat yielded by BMI.”

Still, the research doesn’t suggest that doctors should never use BMI, Mainous said. It can still serve as a simple starting point to determine whether a patient needs to lose weight or gain lean muscle mass to prevent chronic diseases like diabetes or heart disease. But it’s just that, he said—“an easy first step.”

Ideally, other metrics should be measured as well. “The more data points we use—BMI, waist circumference, body fat, blood pressure, blood cholesterol—the more complete a picture we will have of patient health,” Mehal said.

Modern machines that measure BF% using bioelectrical impedance analysis can produce results in less than a minute, the study authors wrote.

Mainous hopes more physicians will begin measuring BF%. We need to focus on disease prevention, he said, and a direct measure of body fat percentage can help us do that.

“We now have nationally representative evidence that a strategy to do that in an inexpensive and viable measure for the doctor’s office can do just that,” he said.