The American obesity epidemic might be much worse than many experts believe because of the limitations of the Body Mass Index, which is the most popular number used to diagnose the condition.
Body mass index—a rough estimate of body fat using a person's height and weight—may be missing nearly 40 percent of obesity cases, according to a study from the New York University School of Medicine. These so-called "false negatives" are caused because BMI simply estimates body fat and does not directly measure it. People who have lost a lot of muscle mass—often older women—may be classified as "overweight" by BMI, when they are actually obese, says Eric Braverman, one of the authors of the study.
"BMI is the least accurate test in medicine," he says. "It's been around since 1832 and hasn't changed. People are being told their BMI is [a healthy] 24, when their body fat is actually at 34 percent, which is obese."
The Centers for Disease Control estimates that 35 percent of American adults and 17 percent of American children are obese. According to Braverman, that number is likely much higher.
His team used BMI and two other measures—one called Dual Energy X-Ray Absorptiometry, which tests body fat, muscle mass, and bone density and measures of leptin, a protein that regulates metabolism—to test a nationally-representative cross section of 1,400 patients. The BMI of those patients underestimated their body fat by an average of 4 percentage points. According to the study, 39 percent of patients were misclassified as non-obese according to their BMI but were considered obese when their body fat was measured directly using a DXA test.
The number of missed diagnoses jumped to nearly 50 percent in women, who often begin losing muscle mass as they age.
"Patients are being misinformed that they are fit and thin, and then they bump into heart attacks and diabetes as they age," Braverman says. "If they were told the truth when they were young, they could do something about it."
The CDC and the National Institutes of Health say that BMI is a relatively good indicator of obesity, but that it does have limitations. One of the strongest arguments for using BMI is its cost—if a patient knows their height and weight, it's free. But leptin can be measured with a simple blood test, Braverman says, and the cost of performing that test is rapidly dropping. DXA, on the other hand, can be added to any bone density scan at minimal cost.
"This should be the beginning of the end for BMI," he says. "Leptin levels will eventually cost no more than a dollar to do, it'll be measured like cholesterol."
That attitude is generally becoming more popular among physicians, says Robin Blackstone, president of the American Society for Metabolic & Bariatric Surgery.
"I think this study confirms what a lot of us believe—that BMI should not be used as a strict cut off for who should be treated for obesity," she says. "A lot of the literature from the NIH and CDC reflects an older point of view, but now we have a shifting paradigm. There shouldn't be an arbitrary [BMI] cutoff before someone considers treatment."
Blackstone says more doctors are using what's called a Body Adiposity (body fat) Index, which measures a person's hip circumference and height. Some research suggests that the measure may be more accurate than BMI, and is just as inexpensive to perform.
James Zervios, a spokesperson for the Obesity Action Coalition, says the organization has always recommended people check their BMI because it's easy to do.
"It's just a simple tool—we've pushed it as something individuals can do on their own and figure out easily," he says. "Until the cost comes down [with the other tests], it's going to be difficult for something else to take center stage."
Zervios says that BMI is a "decent measurement to start with," but does agree that it can trick people into thinking they're at a normal weight.
"At the end of the day, it's important to go to your doctor and have them take a look at your weight," he says. "There's a lot of undiagnosed cases out there."