"Is a BMI of 25 for overweight and 30 for obese the right ones to use with 20- to 29-year-olds in terms of disease risk?" he asked. "Maybe it could be as simple as this: If you're a regular exerciser or an athlete, maybe 28 is overweight for you and 33 is considered obese."
BMI, said Pivarnik, is used as a "surrogate" for percent fat, even though body fat doesn't really fit into the equation.
"What if you take fat out of the equation?" he said. "Is there something unique about BMI that may increase a person's risk of chronic disease, unrelated to fatness? Are people with high BMIs but low fatness as much at risk of heart disease as other people? The answer is no one knows."
In their research, Ode and Pivarnik used carefully measured height and weight to calculate BMI in more than 400 college-age people. They then used a standard method of measuring body fatness and found that BMI was not providing an accurate portrayal of the amount of fat a student had.