Stanford Report Online

Stanford Report, September 27, 2000
Obesity guidelines could be simplified, researchers find


Simplify, simplify was Henry David Thoreau's credo. His advice applies to the latest government guidelines that help doctors determine who needs to lose weight, say two researchers from the Stanford Center for Research in Disease Prevention. Their study shows that a simpler protocol that omits one tricky measurement works just as well and can save busy doctors time.

As the American waistline balloons, physicians are seeing more and more patients who need to do something about their weight. The question the doctor must answer is whether a particular patient should aim to shed pounds or to stop gaining. Two years ago, the National Heart, Lung and Blood Institute (NHLBI) tried to make this decision easier by issuing evidence-based guidelines for weight-control treatment.

The guidelines rely on three variables. The first is body mass index, or BMI, an indicator of fatness obtained by dividing weight in kilograms by the square of height in meters. People with a BMI of 25 and above are considered overweight, while those with a BMI over 30 are classified as obese.

The second variable is waist circumference, which gauges the amount of fat stored in the abdomen -- the greater the amount, the higher the chances of a heart attack. Even people of normal weight (a BMI of under 25) who store a lot of abdominal fat are at higher risk. The final variable a doctor should consider, according to the NHLBI guidelines, is the presence of two or more risk factors from a list that includes familiar cardiovascular risk factors like high blood pressure, smoking and a family history of early heart disease.

By plugging these values into a 16-step algorithm devised by the Institute's expert panel, a doctor can quickly decide which patients should immediately begin a weight-loss program and which should try to stabilize their weight.

When Stanford researchers Michaela Kiernan, PhD, and Marilyn Winkleby, PhD, began looking into the guidelines, their initial goal was to estimate the future need for weight-loss treatment, assuming that doctors began using the new algorithm. To peg that value, they turned to a massive data set known as NHANES III, a national health survey of nearly 34,000 adults and children conducted between 1988 and 1994. Participants were interviewed about their health and diet and also underwent physical examinations.

From this pool, Kiernan, a social science research associate, and Winkleby, a senior research scholar, gleaned information on more than 9,100 black, Mexican-American and white adults. They then calculated how many of these adults would fall into the "weight-loss treatment" and "weight maintenance" categories under the new guidelines and under a simplified set of guidelines that use only BMI and presence of cardiovascular risk factors.

They found that 98 percent of the patients would get the same treatment recommendation using the simpler algorithm, the researchers wrote in the July 24 issue of Archives of Internal Medicine. The results suggest that the simpler procedure is equally effective for screening patients, and it takes less time to perform.

Eliminating waist measurements from the algorithm made little difference because BMI and the cardiovascular risk factors together identify nearly all the people in need of treatment, Kiernan said. She added that this may be because the waist cutoffs were set so high -- 88 centimeters (35 inches) for women, 102 centimeters (40 inches) for men. Virtually no one of normal weight had a waist that large, while virtually everyone with a BMI above 30 surpassed the cutoff, so the measure doesn't provide any further guidance for doctors.

The study isn't saying that measurements of abdominal fat aren't an important gauge of cardiovascular risk, Kiernan said. In fact, one recent study found that women with waist sizes above 76 centimeters (30 inches) had twice the risk of heart disease. "It means our measures of abdominal fat need some work, and these cutoffs may not be appropriate," Kiernan said.

Using the more complicated algorithm won't reduce the quality of care for patients, Kiernan said. But when doctors are required to see more and more patients in a day, why should they have to spend time taking a measurement that doesn't help them make a decision? And it does take time. As any tailor can testify, the diversity of body shapes makes it difficult to accurately measure waist size, even for people who aren't overweight. It's even harder when the person is heavy, and patients often find the process uncomfortable and embarrassing.

"The study speaks to the need for more scientific evaluation before implementing health care policy," Kiernan said.