Flawed formula: BMI has holes

As a measure of obesity, body mass index could improve

4/21/2014
BY JANE E. BRODY
NEW YORK TIMES NEWS SERVICE

In July, 1998, the National Institutes of Health changed what it means to be overweight, defining it as a body mass index of 25 or greater for adults. The cutoff had been 28 for men and 27 for women, so suddenly about 29 million Americans who had been considered normal became overweight even though they hadn’t gained an ounce.

The change, based on a review of hundreds of studies that matched BMI levels with health risks in large groups of people, brought the country in line with definitions used by the World Health Organization and other health agencies. But it also prompted many to question the real meaning of BMI and to note its potential drawbacks: labeling some healthy people as overweight or obese who are not overly fat, and failing to distinguish between dangerous and innocuous distributions of body fat.

More recent studies have indicated that many people with BMI levels at the low end of normal are less healthy than those now considered overweight. And some people who are overly fat according to their BMI are just as healthy as those considered to be of normal weight, as discussed in a new book, The Obesity Paradox, by Dr. Carl J. Lavie, a cardiologist in New Orleans, and Kristin Loberg.

Unlike readings on a scale, BMI is based on an individual’s weight in relation to that person’s height. It is calculated by dividing weight in kilograms by height in meters squared (or, for those not metric-savvy, weight in pounds divided by height in inches squared and the result multiplied by 703).

According to current criteria, those with a BMI below 18.5 are underweight; those from 18.5 to 24.9 are normal; those from 25 to 29.9 are overweight; and those 30 and higher are obese. The obese are further divided into three grades: Grade 1, in which BMI is 30 to 34.9; Grade 2, 35 to 39.9; Grade 3, 40 and higher.

Before you contemplate a crash diet because your BMI classifies you as overweight, consider what the index really represents and what is now known about its relationship to health and longevity.

The index was devised in the 1830s from measurements in men by a Belgian statistician interested in human growth. More than a century later, it was adopted by insurers and some researchers studying the distribution of obesity in the general population. Though never meant to be an individual assessment, only a way to talk about weight in large populations, BMI gradually was adopted as an easy and inexpensive way for doctors to assess weight in their patients.

At best, though, BMI is a crude measure that "actually misses more than half of people with excess body fat," Geoffrey Kabat, an epidemiologist at the Albert Einstein College of Medicine, has noted. Someone with a "normal" BMI can still be overly fat internally and prone to obesity-related ills.

Calling BMI an imperfect predictor of a person's health risks, the Centers for Disease Control and Prevention cautions doctors against using it as a diagnostic tool.

For one thing, body weight is made up of muscle, bone and water, as well as body fat. BMI alone is at best an imprecise measure of how fat a person may be. When Arnold Schwarzenegger was Mr. Universe, his BMI was well in the obese range, yet he was hardly fat.

Another problem: the distribution of excess body fat makes a big difference to health. Those with lots of abdominal fat, which is metabolically active, are prone to developing insulin resistance, elevated blood lipids, high blood pressure, diabetes, premature cardiovascular disease, and an increased risk of erectile dysfunction and Alzheimer's disease.

But fat carried in the hips, buttocks or thighs is relatively inert; while it may be cosmetically undesirable, it is not linked to chronic disease or early death.

Furthermore, a person's age, gender and ethnicity influence the relationship between BMI, body fat and health risk. Among children, a high BMI is a good indicator of excess fat and a propensity to remain overly fat into adulthood. But for an elderly person or someone with a chronic disease, a BMI in the range of overweight or obesity may even be protective. Sometimes — after a heart attack or major surgery, for example — extra body fat can provide energy that helps the patient to survive. An added layer of fat can also protect against traumatic injuries in an accident.

On average, women have a higher percentage of body fat in relation to total weight than do men, but this does not necessarily raise their health risks. And African-Americans, who tend have heavier bones and weigh more than Caucasians, face a lower risk to health even with a BMI in the overweight range.

Physical fitness, too, influences the effects of BMI. In an editorial in JAMA last year, Dr. Steven B. Heymsfield and Dr. William T. Cefalu of the Pennington Biomedical Research Center in Baton Rouge, La., noted that "cardiorespiratory fitness" is an independent predictor of mortality at any level of fatness.

While experts continue to debate whether a person can be "fit and fat," Keri Gans, a dietitian in New York and former spokeswoman for the Academy of Nutrition and Dietetics, points out that physical activity and a healthy diet tend to offset the risks of being overweight.

"You don't need to be thin to be fit," she said.

At any weight, fitness can reduce the risk of developing heart disease, lung disease, diabetes or high blood pressure.

At the other end of the weight spectrum, people with a low-normal or below-normal BMI (less than 18.5) face a different set of health risks. They may lack sufficient reserves to survive a serious health problem, and they are prone to osteoporosis, infertility and serious infections resulting from a weakened immune system.

Last year a widely publicized meta-analysis covering more than 2.88 million people and 270,000 deaths found that those whose BMI indicated they were overweight and those with Grade 1 obesity were not at a greater risk of death than those in the normal range. And a new analysis of 32 studies by researchers in Australia concluded that for older people, being overweight did not increase mortality, but the risk rose for those at the lower end of normal, with a BMI of less than 23.

 

An alternative to BMI

Although BMI is a reasonable measure of fatness for the average sedentary person, there are several potentially more accurate methods. But doctors rarely use them because they are more time-consuming and thus more costly, and some require equipment rarely found in medical offices.

The simplest is a tape measure around the waist and hips. Divide the waist measurement by the hip measurement; a result above 0.9 for men or 0.85 for women indicates abdominal obesity and an elevated health risk.

Almost as simple is a skin-fold caliper, used to measure fat beneath the skin. It updates the old pinch at the waist; more than an inch between the fingers, and you're too fat.

A more sophisticated measurement, called bioelectric impedance analysis, uses a special scale that measures resistance to the flow of an electric current to estimate fat-free body mass and body fat percentage. Its accuracy depends on how much water you retain in your body.

The so-called gold standard of body fat measurement, hydrostatic weighing, takes about half an hour, costs up to $150 and involves being weighed while submerged in a tank of water.

Finally, a DEXA scan, like that used to measure bone density, improves on the accuracy of water weighing. It divides the body into fat and fat-free tissue, and can analyze the distribution of body fat.