Which would you choose? Although many questions remain, current research suggests that a low-fat diet is best.

Food Pyramid on Stable Ground

After 3 years spent sifting through thousands of studies, 21 researchers appointed by the National Academy of Sciences (NAS) released a report today cataloging how much fat, protein, and carbohydrates healthy North Americans should eat. Their endorsement of a low-fat, high-carbohydrate diet reflects the prevailing scientific view that certain fats, particularly saturated fats and trans fatty acids, pose health risks even in low doses. But panel members also expressed frustration at the dearth of research in certain areas and conflicting results in others.

Since the 1970s, public health officials have exhorted Americans to eat less fat. Well, it's not that simple. Whereas certain fats, such as the omega-3 fatty acids found in fish, are deemed beneficial, others are blamed for chronic diseases, such as heart disease, diabetes, and hypertension. On the other hand, it's not certain whether a low-fat diet actually prolongs life (Science, 30 March 2001, p. 2536). And if you eschew the fat, you are probably eating more carbohydrates. The trouble there is that certain carbohydrates may cause problems too, among them obesity and diabetes. So where should one get the majority of calories?

Today, the NAS panel announced its conclusions: Existing research, it said, supports a diet made up of 45% to 65% carbohydrates, 20% to 35% fat, and 10% to 35% protein. Ranges had not been suggested in the past (previous recommendations called for 50% or more carbohydrate and 30% or less fat). The ranges--within which the vast majority of Americans fall--are meant to give the public flexibility and reflect the belief that only outside that range will one risk ill effects. The panel, chaired by Joanne Lupton of Texas A&M University in College Station, also recommended an hour of exercise a day--double the amount suggested by the U.S. Surgeon General in 1996, and well beyond what most North Americans currently manage.

Still, the panelists confessed that they were unable to spell out the risks of relying heavily on a single macronutrient, such as protein--the recipe for certain popular diets--and that the data failed to answer all their questions. "Although the reference values are based on data," the report reads, "the data were often scanty or drawn from studies that had limitations."

"The biggest list [in the report] was all the answers we don't have," says George Blackburn, associate director of the division of clinical nutrition at Harvard Medical School in Boston. With public debate raging over the relative value of fats and carbohydrates, resolution remains a long way off.

NAS Report