Why Does Nutrition Advice Always Seem to Be Changing?
It can be hard to keep up with ever-changing guidelines. Marion Nestle shares a few principles to help steer you right.
Everyone eats. Everyone can claim firsthand experience and expertise. Whose experience and expertise should you trust? “Mine, of course,” is my standard (slightly facetious) answer. I can understand why people trust celebrities more than scientists or nutritionists; they feel like friends, even if the relationship is unreal. It doesn’t help that nutritionists have impenetrably confusing credentials, ranging from none beyond personal experience to years of graduate and post-graduate study.
It also doesn’t help that nutrition science is so extraordinarily difficult to do. Just think of what it would take to show whether eggs, the largest dietary source of cholesterol, raise the risk of heart disease. To achieve definitive results, you would need to put large numbers of people matched in age, gender, and risk on one of two defined diets, the same except for whether eggs are included. To make sure your study subjects stick to the diet, you would have to confine them under close supervision for expensively long periods to see whether eggs induced symptoms.
Humans are not lab rats. We make terrible experimental animals. This forces nutrition scientists to resort to indirect measures, such as blood cholesterol levels, that do not always relate clearly to disease risk.
I’ve long said that the most intellectually challenging problem in nutrition is to figure out what people eat. Diets vary from day to day, and yours differs from mine. Scientists ask us to write down everything we ate yesterday (24-hour recall), or to keep track of what we eat in a day (24-hour dietary record), or to fill out a survey of how often we ate a given food in the last week, month, or year (food-frequency questionnaire). The accuracy of these methods depends on how well we remember what we ate. They are, to say the least, imprecise. At the moment, better methods are too expensive and difficult to use with large numbers of people, leaving researchers to do the best they can with whatever information they can get.
Humans are not lab rats. We make terrible experimental animals.
Linking dietary intake to disease risk is also difficult. Suppose you want to know whether people who eat eggs have higher levels of cholesterol in their blood. Food cholesterol raises blood cholesterol, but saturated fat raises it even more. If your blood cholesterol is high, it’s hard to know whether this is due to eggs, your level of physical activity, or something else you ate, took, did, or were born with.
That is why the best that most studies can do is to show some kind of association or link between what you ate and the likelihood — your risk — of developing a disease. They cannot prove that what you ate caused a disease. Depending on how studies of eggs and health are designed, the results could go one way or the other — and they do, with no resolution in sight.
A further complication is that the egg industry pays for research about eggs and health. When scientists first identified cholesterol as a risk factor for heart disease, they advised limiting egg intake to no more than one a day. Alarmed, the egg industry commissioned research to cast doubt on that advice. As with much industry-funded research, the egg industry’s studies showed “funding effects” — results that favor the sponsor’s interests — in this case, that eggs pose no risk to health. Egg research results are now so muddled that the 2015 Dietary Guidelines for Americans judged a limit on dietary cholesterol as unnecessary but advised eating as little cholesterol (meaning as few eggs) as possible. I have no idea what to make of this.
It’s not that all industry-funded research is biased; it’s just that most of it appears to be. Because nutrition research is so complicated — and because everyone feels like an expert — biases get into this research in all sorts of ways. My molecular biology training has proved useful for dealing with them. As a graduate student, I was taught to distrust my own results and to overcompensate for biases, recognized and not. Scientists are people too. We have hypotheses we want to prove. Some of us may be so wedded to hypotheses that we discount studies we don’t like and “cherry-pick” the data. But rigorously trained scientists go out of our way to control for our conscious and unconscious biases.
Any nutritionist who thinks critically knows that diets of enormous variety are capable of supporting health and longevity.
As it happens, I just love the complexities of nutrition research and findings that are gray, rather than black or white, and that inevitably require interpretation. It is human nature to think that other people are biased, but never ourselves. It is human nature to trust the nutrition opinions of celebrities, even if a moment’s thought ought to reveal the absurdity of that trust. A mere trace of critical thinking should make you surprised (if not outraged) that celebrities who know nothing of the complexities of nutrition science are taken seriously as diet gurus, that food companies pay for studies that give them the results they want, and that scientists wedded to their own hypotheses insist that only one diet — the one they recommend — is appropriate for health.
Any nutritionist who thinks critically knows that diets of enormous variety are capable of supporting health and longevity. Except for avoiding food allergies, a claim that one diet, one food, one food product, or one supplement is the solution to your health problems — no matter who that advice comes from — requires skepticism. Critical thinking is especially in order whenever a recommended diet:
- Involves a product: Here, the claims are likely to be about marketing, not science. Food companies fund research because it helps them sell products. Beware of anyone hawking products of one kind or another.
- Excludes whole categories of foods: This may help reduce calorie intake but limits variety and pleasure and could lead to reduced intake of certain nutrients. Animal food products are the main source of vitamin B12, for example.
- Is announced with absolute certainty: That’s not how science works. Nutrition science is inherently uncertain due to the complexity of everything we eat, drink, or do.
- Claims to prevent a wide range of conditions: I get suspicious whenever I see a claim that one food or product will not only prevent Type 2 diabetes and heart disease, but also conditions without known cures, such as autism, Alzheimer’s, or other cognitive problems.
- Is advertised as a breakthrough: This too is not how science works. Nutrition science builds on previous studies and makes incremental progress.
- Insists that everything you thought you knew about nutrition is wrong: Science rarely works this way, so it’s best to read the fine print. Case in point: even when health authorities advised cutting down on eggs, they said one a day was acceptable. They still give the same advice.
These same principles apply to studies of coffee, wine, and any other food or drink. As is ever true, the basic principles of nutrition are easily summarized:
- Eat a wide variety of foods. Because food plants and animals differ in nutrient composition, variety ensures the full complement of needed nutrients.
- Eat relatively unprocessed foods. These contain nutrients but do not have excessive amounts of added sugars, fats, salt, and calories.
- Eat in moderation. The word “moderation” may sound like a nutritional joke — what does it mean, exactly? — but the advice aims to keep calorie intake balanced with calorie needs.
To these principles, I would add one more: Enjoy what you eat.