Bruno Gualano: Why does the egg that makes you feel good today kills you tomorrow

Sorry for the pun, dear reader, but many studies in the field of nutrition are indigestible. There are those who suggest that consuming an egg a day would shorten our life expectancy by six years, and, on the other hand, those who point out that eating 12 hazelnuts daily would make us 12 years longer (one year for each berry!). Are these results minimally plausible?

With this rhetorical question in the background, John Ioannidis, a prominent scientist at Stanford University, has harshly criticized not only the far-fetched findings of famous population-based nutritional studies, but especially their methodological flaws.

According to Ioannidis, the exorbitant conclusions generated by these studies are due, above all, to a failure of analysis and interpretation. In general, population studies are associative in nature. And the association between two variables does not necessarily imply causality between them. Let’s look at this example. In the past, coffee consumption has been correlated with increased cancer risk. At the time, many concluded that drinking caused the disease. Suspicions would only be ended with the finding that the habit of drinking a cup of coffee often accompanies that of smoking —voilà!, this is indeed the causal factor behind the spurious association.

Confounding factors (smoking in the previous example)—whose name does justice to the effect they have on the interpretation of results—abound in nutrition science. Some classic California studies from the 1960s to 1980s attributed the consumption of tomatoes, vegetables, grains, whole foods, etc. to the greater longevity of Adventist believers compared to non-religious ones.

However, surveys have also revealed that Christians, on average, drink and smoke less, engage in more physical activity, and have lower rates of obesity. Certainly, the combination of these factors that make up lifestyle—rather than the consumption of a punctual food—explains much better why Adventists tend to live longer.

In observational nutritional studies, such confounding factors are difficult, if not impossible, to be controlled, often generating exaggerated or ambiguous results, which only confuses the population. The thing about the egg that makes you feel good today, but kills you tomorrow…

As Ioannidis reminds us, there are more than 250,000 foods that can be consumed daily in millions of different combinations, from different culinary preparations, resulting in virtually unpredictable biological responses. But this “detail” is often ignored by researchers who, based on the assessment of the diverse and complex diet of populations, choose a single food (eg egg) or, even worse, one of its nutrients (eg cholesterol) as capable to modulate the risk of illness or death.

Ioannidis is surgical in revealing the wounds from which nutritional studies suffer, calling into question worldwide dietary guidelines, which are based on the existing literature. But what would come of the implosion of the area? In the scientist’s view, the “radical reform of nutrition science” would comprise two fronts: the broad and transparent reanalysis of published observational studies, considering the real influence of confounding factors on the results; and the concentration of efforts and funding on large-scale studies, with randomized designs and a focus on nutritional patterns, rather than a single food or nutrient.

There is no doubt that the proposed reform has the potential to generate more qualified evidence to support the construction of better public nutrition policies. However, it is always good to remember that scientific findings are not readily interchangeable between populations, given the profound particularities of food in each part of the world.

The efficient confrontation of our current public health nutritional problems, which involve, among others, obesity and food insecurity, depends mainly on knowledge produced by robust local studies. It is to deal with this type of social challenge that the Brazilian scientific community has relentlessly demanded adequate funding.