BMI may not be the best way to predict a person's likelihood of dying… (Lori Shepler / Los Angeles…)
A new way to measure and categorize an individual's body shape appears to predict more accurately whether he or she is in greater danger of premature death, says a pair of scientists in a new look at alternatives to the body-mass index (or BMI).
The proposed new measure is called "A Body Shape Index," or ABSI, by the father-and-son team that has devised and tested it, Dr. Jesse Krakauer, an endocrinologist at Middletown Medical in Middletown N.Y., and his Nir Krakauer, an assistant professor of engineering at City University of New York. Their study was published this week in the open-access journal Public Library of Science One.
The ABSI measure uses just three easily captured metrics -- a person's height, weight and waist circumference -- to determine whether an individual has what they grimly characterize as a "hazardous body shape." Their equation manipulates those three measures in a relatively simple and compact way: Take your waist circumference (in centimeters) and divide that by the product of two variables. The first is the square root of your height (in centimeters); the second is the square of the cube-root of your BMI.
The result promises to predict the impact of body fat and body shape on an individual's health prospects more accurately than BMI alone or waist circumference alone. It comes at a time when physicians and public health officials are eager to find a better way to detect which patients are most vulnerable to fat-related health consequences and to guide those patients in ways they can improve their odds of staying well.
Since it was invented in the 19th century by mathematician Adolphe Quetelet, the BMI has been the mainstay of medical research on body size and its relationship to health. But while the BMI is a fine measure of a large population's tendency toward illness and early death, it's a poor predictor of those outcomes in individual patients.
Waist circumference, a measure that has gained adherents more recently, may be a better predictor of health, since it is more likely than BMI to identify the presence of dangerous fat deposits around the visceral organs. But waist circumference is still not perfect, since a proportionate waist on a 5'11" woman will probably be larger than that of a woman who's 5'2".
The ABSI is, to be sure, tied to BMI and waist circumference in that both are critical components of the measure. Losing weight can lower one's BMI, and thus improve one's ABSI. An intensive exercise regimen can whittle the waist and redistribute fat away from the middle, and that too will improve one's ABSI. But the authors stressed that further research must answer two fundamental questions about ABSI: what a patient with a "hazardous body shape" can do to change it, and whether changing it can boost his or her longevity or quality of life.
Recent decades have made it even more urgent to find a better prognostic measure: Today, 78 million American adults and 12.5 million children are considered obese by BMI standards -- and therefore at higher risk of developing such illnesses as type 2 diabetes, cardiovascular disease and certain cancers. It's an overwhelming number to try to motivate to weight loss. So physicians and public health officials must consider whether all with a BMI over 30 (the definition of obesity) are in equal need of aggressive treatment.
They might also need to know which among those of normal, "healthy" weight has a body shape that puts them at higher risk -- in other words, which patients the BMI or waist circumference is missing.
The Krakauer brothers calculated the ABSI of more than 14,000 Americans, and then tracked those participants to see how many died within five years of the time they were measured. They found that their ABSI metric bested both waist circumference and BMI in predicting whether men, women, whites, blacks and adults across the age spectrum are likely to die of any cause at some point in the next five years. When they took account of factors other than fat that significantly affect a person's risk of dying -- cigarette smoking, diabetes status, high blood pressure, for instance -- the ABSI remained an accurate way to distinguish which patients were more likely to die.
The ABSI measure was not a useful predictor of death in the study period among Mexicans.
A logical next step, the authors said, would be to track subjects for a period longer than five years to see if the ABSI measure remained a good predictor both of health conditions and of death.