What's the best diet?

We've heard it time and again. Every day, eat no more than:

* 30 percent of your calories from fat,

* 10 percent of your calories from saturated fat,

* 300 mg of cholesterol, and

* 2,400 mg of sodium. Every day, eat at least:

* 20 to 30 grams of fiber and

* 5 servings of fruits and vegetables.

Does that advice describe the best diet? Or are health authorities tempering what's best with what they think we can achieve?

CSPI associate nutritionist Jayne Hurley called 11 leading researchers and asked them not what's practical, but what diet would lead to the lowest risk of heart disease, cancer, and other health problems. She also asked them what they ate.

SATURATED FAT

Ideally, how much saturated

fat should we eat?

Shekelle: Ten percent seems to be a reasonable, practical target, but going lower will decrease your LDL ["bad"] cholesterol level.

LaRosa: The recommended ten percent of calories from saturated fat is not a target; it's a ceiling. For people who have high cholesterol levels, the American Heart Association's recommendation is seven percent or less.

The real question is whether we should go lower than ten percent for the public. The evidence is scanty, but it suggests that it would be good. It depends on what level of evidence you're willing to base public policy on.

Connor: I'd say five to six percent of calories is ideal. Any more and you start to get changes in the LDL receptors, which means more harmful LDL stays in your blood.

Sacks: None. Saturated fat is not needed. We get nothing beneficial from it, and it raises LDL cholesterol. That means virtually no flesh food or dairy fat.

Barnard: The optimum amount of animal fat on the plate is zero. A study at the University of Massachusetts showed that vegans [vegetarians who eat no dairy or eggs] have lower cardiovascular risk than ovo-lacto vegetarians [who eat dairy and eggs].

What about cancer?

Stampfer: You can eat your ten percent of calories from saturated fat and still have more meat than is healthy. If health is the only consideration, red meat intake should be very low or absent. Meat increases the excess risk of colon cancer quite a bit.

Lean meat might reduce that excess risk, but it might not. I don't know if it's the fat, the protein, or something else in the meat that's causing the colon cancer. This country probably would be healthier if we came closer to a vegetarian diet. But I don't think we have enough solid data to make such a public health recommendation.

CHOLESTEROL

How much cholesterol is ideal?

LaRosa: I can't find a rationale for lowering the recommended level of cholesterol below 300 mg per day. I think we overemphasize dietary cholesterol. It has some effect on blood cholesterol, but more when it is combined with saturated fat.

Having said that, there is still good evidence that dietary cholesterol does affect blood cholesterol levels. And there is a lingering concern that cholesterol intake is directly related to heart disease, independent of what it does to blood cholesterol. That may be because the fats that form after we eat, which disappear in a few hours, may be more artery-clogging if there is more cholesterol in the diet.

Connor: 100 milligrams per day or less. If you eat more, it can raise your LDL cholesterol. Populations that do not have a high rate of heart disease, particularly the Japanese, consume below 100 mg per day.

Sacks: Zero cholesterol, with the exception of fish, at least until we have better data. Fish does have some cholesterol, but it also has other nutrients like the potentially beneficial omega-3 fats and minerals.

Barnard: The optimum is zero.

TOTAL FAT

How much total fat is ideal?

LaRosa: We don't have a lot of evidence from clinical trials. If you're willing to use studies of populations, then, at least for heart disease, you're better off with less than 30 percent of your calories from fat. But how feasible is it for people to lower fat below 20 percent of calories given the foods that are available today?

Also, we don't know about the possible detriments of a diet that supplies only 10 percent of calories from fat. It doesn't seem to have bad side effects, but I'm not certain whether it's good for children's growth and development.

Shekelle: Somewhere between 25 and 30 percent is a good compromise. If you tell people to get under 20 percent, it may be disruptive. I'm concerned that people not move towards faddish or nutritionally inadequate diets.

Connor: I recommend 20 percent of calories from fat to my patients and to the general public. That fits the population data, and when we fed 20 percent to the Tarahumara Indians, who eat a low-fat diet and have very low cholesterol levels, they did not have a rise in cholesterol. This fat should include omega-3 fatty acids, particularly from fish.

Wynder: Countries like China and Japan, which have diets lower in fat and higher in fruits and vegetables, have lower rates of heart disease and certain types of cancer.

Ideally, fat should be ten percent of our calories. While there are peoples with largely vegetarian diets who can reach that goal, we are used to a much higher-fat diet. That's why I suggest 25 percent as our national goal.

Barnard: Some people exclude animal products but include potato chips and french fries. I consider vegetable oil to be like whie bread or sugar. It's a refined, processed product. There is no oil plant.

If people stick to the new four food groups that we've proposed--fruits, vegetables, legumes, and whole grains--their fat intake will automatically be about ten percent of calories.

I don't suggest that people count their fat, fiber, calories, protein, or anything. Those people need to be in an anti-obsessional program with a competent psychotherapist.

Sacks: Vegetable oils could be neutral, or even beneficial, since they raise HDL ["good"] cholesterol. So I'd say that there is no need for total fat to be very low, like 15 to 20 percent of calories. It could be as high as 35 to 40 percent. It's a killer to try to eat under 20 percent fat, and I wouldn't want to impose a tough guideline if it's not clearly to people's benefit.

Given a calorie intake that will maintain a satisfactory leanness--not obesity--I don't think it matters whether calories come from vegetable oils or carbohydrates.

What about cancer?

Sacks: I don't know of any data that says that a vegetable-based, no animal-food diet that has a fair amount of oil promotes cancer. The high-fat diets that are associated with cancer in humans are very low in vegetables that are cancer protective.

Stampfer: When you look at individuals, as we have done in the Nurses Health Study, we find virtually no difference in the risk of breast cancer with fat intake varying between 29 and 49 percent of calories from fat.

Barnard: The people in the Nurses Health Study were eating so much fat that I'd be amazed if going down to 30 percent would reduce the incidence of cancer. An analogous study would ask: "What's the benefit of going from four packs of cigarettes a day to three?"

SALT

How much sodium is ideal?

Stamler: For public education, 2,000 mg is easy to remember. It's close to one teaspoon of salt. In my opinion, though, the lower, the better; 1,800 mg is better than 2,000 mg.

If we cut our national average intake of sodium by about 2,300 mg, from two teaspoons a day to one, it could lower the population's average systolic blood pressure by about 2.5 millimeters of mercury. That may not seem like much, but it translates into a five percent reduction in the coronary death rate and a nine percent reduction in deaths from stroke.

LaRosa: We've been forced into the position of inoculating everyone for the sake of a relatively few people. Sodium limitation for people who have high blood pressure is clearly important, although pressure responds best to weight reduction.

Do we all need to worry

about sodium?

Stamler: Not all of us respond to excess sodium intake by developing hypertension. But most of us respond by having blood pressure above the optimal and by having pressure increase with age, which mean extra risk.

We have worked with non-hypertensive people who were hypertensionprone because they had high normal blood pressures and were overweight and high-salt eaters. By reducing sodium, weight, and alcohol--all modest changes--we reduced the incidence of hypertension by 50 percent.

FIBER-RICH FOODS

How much fiber should we eat?

Jenkins: In areas of the world where people eat large amounts of foods rich in soluble fibers, such as beans, peas, and other legumes, you have lower incidences of colon cancer. And there's good evidence that increased soluble fiber lowers blood cholesterol.

Wynder: The ideal diet may be 40 to 50 grams of fiber per day, but we only recommend 25 grams. And that applies to everyone over the age of two. It's a diet we can get used to, and one the food industry can produce.

Is the soluble fiber most

important?

Wynder: In terms of colon and breast cancer, it's the insoluble fiber in foods like wheat bran that seems to make the very difference.

Block: It's not clear that fiber is the most important factor in these foods. Fruits and vegetables are our main sources of antioxidants like vitamin C and the carotenoids--not just beta-carotene. Antioxidants prevent, and in some cases repair, damage that is caused by oxidation and free radicals.

Tomatoes are good sources of lycopene, which may be valuable in preventing some cancers. And fruits, vegetables, and whole grains have vitamin E as well as fiber. We should eat more beans, perhaps because of their fiber, but also because they are a source of protein that isn't a source of fat.

Are five servings a day of fruits

and vegetables enough?

Block: That's sensible and attainable. We don't know if it's optimal, but it would be a huge improvement over what we currently eat. On any given day, only nine percent of the population eats two or more servings of fruits and three or more servings of vegetables.

SUGAR

Does sugar cause anything but

tooth decay?

Hallfrisch: Most of us don't exercise enough that we can afford to waste a lot of calories on foods that have no redeeming qualities. And even though starch has the same number of calories as sucrose, fructose, and other sugars, sugars are more likely than starch to be converted to body fat.

Barnard: Sugar makes you fat if you really try. But sugar is not to blame for obesity in most people. It's not the sugar in the pie, it's the fat.

Are there other reasons

to avoid sugar?

Hallfrisch: For some of the population, sugar is much more detrimental than for young, fit people. If certain postmenopausal women and middle-aged men with high triglycerides and blood cholesterol consume large amounts of sugar, their glucose tolerance deteriorates and their cholesterol and triglycerides can increase.

And sugar can raise triglycerides in overweight people whose bodies don't use insulin efficiently.