BMI might not tell the whole truth


Sometime later this year, the Centers for Disease Control and Prevention will release its latest accounting of the nation’s weight problem, as measured by the body mass index, or BMI.

This much we know already: It won’t be a pretty picture. The last census reckoned that a third of American adults were overweight, meaning their BMIs (calculated by dividing a person’s weight by his height, squared) lay between 25 and 30. About another third weighed in with BMIs over 30 -- the demarcation line that brands them as obese.

With BMIs like that, experts tell us, the prognosis for the nation’s health is grim. In recent decades, researchers have amassed a fearsome litany of health woes for those whose BMI exceeds 30: higher rates of cardiovascular disease, Type 2 diabetes, a passel of cancers, arthritis, infertility, liver disease, sleep apnea, dementia and depression. Victimization to workplace harassment and bullying is more likely, the brain suffers higher rates of shrinkage with age, and the risk of erectile dysfunction surges.


It’s enough to scare you off your next cookie. But here’s the thing: For how often doctors, researchers and politicians lean on this quirky, almost-two-centuries-old mathematical formula to browbeat us into shaping up, a number of experts caution that the BMI is a fairly imprecise tool -- and often wrong -- for helping individual patients appraise their health prospects.

The BMI is easy to measure, cheap to compile and track, and simple to convey to patients, said Richard N. Bergman, an obesity specialist at USC’s Keck School of Medicine; but it was never meant to be a predictor of an individual’s health risks. It was intended as a useful metric to track changes in the health and nutrition of large populations (including the 40-year run-up in Americans’ weight that it has documented so well).

For patients who are very muscular, and for African Americans, body mass index is often a poor (and unflattering) gauge of body composition, Bergman added. For Asians and people who are sedentary but slim, a reliance on BMI can lead a physician to overlook signs of elevated disease risk.

A child’s BMI, which is calculated differently than that of an adult, is also an imperfect predictor of illness or early death.

One of the most serious blows to the authority of BMI as an omen of premature death has come with a string of studies showing that among patients with established heart disease, those who are overweight or obese by BMI standards fare better and survive longer than heart patients of normal, “healthy” weight. Dubbed the “obesity paradox,” these findings have sparked wide debate over the benefits of encouraging dramatic weight loss in heart patients -- and in elderly patients in general. And they have prompted questions about what protective factors some larger patients may have that BMI fails to measure.

The BMI continues to dominate research on obesity and guide physicians’ advice to patients. But some researchers have launched a rear-guard effort to knock the measure from its place atop the public-health pedestal. To replace -- or at least supplement -- the BMI, they are searching for measures that might offer individual patients a better gauge of their health risks, as well as a wider range of options to better their odds of staying healthy.

“BMI is a screening tool, and all screening tools are controversial: They miss some who are at risk, and they give false positives on others” who are not, said Becci Twombley, UCLA’s director of sports nutrition. It plays no part in UCLA’s athletics program, Twombley added, which relies instead on a machine called the “Bod-Pod,” which measures the ratio of fat to lean tissue and detects whether the mass of a large athlete -- say a football lineman who would probably log a BMI in the obese category -- is healthy or not.

Beyond athletes, the BMI routinely sends up red flags over people of any age who are “fit but fat,” critics say. And it clearly fails to raise concerns about skinny weaklings across the spectrums of age, ethnicity and gender. If physicians really want to give the right health advice to patients, these experts say, they need measures more directly tied to diseases -- such as the proportions of muscle and fat a patient carries, and the location of that fat. They will need to account for factors, notably exercise, that can offset some of the harmful effects of excess blubber.

They’ll also need gauges and tools to do so that are as cheap and simple as the BMI. Most techniques that measure body composition, including the Bod-Pod, the dual-energy X-ray absorption test (DEXA) and hydrostatic underwater weigh-in, are costly and time-consuming.

Bioelectrical Impedance Analysis, an inexpensive and reliable gauge of body composition increasingly used in weight-management programs, may be practical in some general medical practices, said Dr. Richard L. Atkinson, co-editor of the International Journal of Obesity.

But researchers are increasingly looking for ways to capture such information with tools as simple and inexpensive as tape measures.

One of the easiest alternatives to the BMI is waist measurement. With growing evidence that fat girdling the waist and visceral organs disturbs metabolism, the circumference of a patient’s midsection has been shown to be a better predictor of Type 2 diabetes risk. Several large studies have linked a waist circumference greater than 40 inches for men and 35 inches for women to a higher risk of cardiovascular disease and Type 2 diabetes, as well as earlier mortality -- with every inch beyond those thresholds adding risk.

A twist on that approach, described last month in the journal Obesity, is a measure of hip circumference. When combined with a person’s height, the “body adiposity index,” proposed by Bergman at USC, offers a simple gauge of how much of a person’s body is made up of fat, and what its proportion is to muscle, bone and water.

Finally, an increasing number of researchers cite a major gap in the BMI: its inability to reflect the health effects of a person’s exercise habits. At least seven high-profile studies in the last decade have established that even for people with high BMIs, cardiovascular risk and the likelihood of early death are driven down significantly by maintaining a high level of fitness or at least regular physical activity. The risk of Type 2 diabetes also falls, although in that case, studies suggest that whittling waist circumference is a better strategy.

The bottom line: “It’s important not only to look at the weight status of a patient, but other factors such as physical activity” in assessing his or her health prospects, said Dr. Amy Weinstein, an internal medicine physician at Harvard University’s teaching hospital, Beth Israel Deaconess Medical Center.

Even as they gain more and better tools to assess their patients’ health prospects, physicians will -- and should -- continue to use the BMI as well, Weinstein said. It gives a rough approximation of a person’s fat, which clearly plays a role in promoting certain diseases. And for busy doctors, she added, the BMI is easy, familiar and cheap.