SCARSDALE DIET, grapefruit diet, Beverly Hills diet, Cambridge diet, liquid protein diet, cabbage soup diet and, of course, Atkins diet and South Beach diet: The names change, the formulas alter, but the gimmick diet never goes away. For that reason alone, there is something pleasantly predictable about the regular reports of the U.S. Dietary Guidelines Advisory Committee, published every five years since 1980. Written under the auspices of the departments of Agriculture and Health and Human Services, the reports are intended to reflect not the latest fad but widely accepted conclusions about nutrition. These then determine the content of school lunches and meals for the elderly and form the basis for the venerable "food pyramid." Historically they have been moderate, ungimmicky and unsurprising.

In recent years, their recommendations have come under challenge from diet gurus, such as the late Dr. Robert C. Atkins, who have argued that they reflect an unjustified "bias" toward low-fat foods. Nevertheless, the committee's preliminary 2005 conclusions are, once again, moderate, ungimmicky and unsurprising. Americans, the guidelines state, eat too few fruits and vegetables, get too little exercise and eat too much sugar. They should consume a wide variety of foods, choose fats and carbohydrates wisely, use little salt, limit calories -- no "eat as much as you want" suggestions here -- and drink alcohol in moderation. All of that, of course, is advice that could have been proffered in Aristotle's time.

Unfortunately, it is mostly being ignored -- which raises larger questions about the purpose of national guidelines. Indeed, the more important question the committee should examine is why so few Americans listen to what it says. Every year, some 400,000 Americans die from illnesses related to poor diet and lack of exercise -- about 16 percent of annual deaths. Obesity is on the rise; so are diabetes, cardiovascular disease, hypertension and osteoporosis, all diseases related to diet, all difficult and expensive to deal with.

If officials are serious about changing America's eating habits, publishing scientific reports isn't enough. New ways have to be found to promote nutritional science, whether through school curricula, doctor's offices, public health campaigns, or Medicare and Medicaid literature. They might also look harder at successful experiments such as the USDA's "Fruit and Vegetable Pilot Program," which, after providing $6 million worth of fresh and dried fruits and vegetables free to children in 107 schools, concluded that regular availability had -- surprise! -- an enormous impact on children's eating habits. If the government's nutritional scientists are being drowned out by soft-drink advertisements and Internet diet programs, then perhaps it's time to look at new and better ways to make them heard.