Why is nutrition so difficult to study?

Is dairy good or harmful to health? Is cholesterol evil? Does red meat kill or cure? May be the ketogenic diet a godsend or a health hazard? Can the vegan, vegetarian, pescatarian, or raw food diet extend disease-free life?

Nutrition is wrapped in multiple confusions. Exactly why is it so hard to determine whether a food is good or harmful to health?

In medical science, proving any theory is difficult. The science of nutrition is no different, but it addittionally has some unique challenges. In this feature, we outline just many of these stumbling blocks.

Despite the many conditions that nutrition scientists face, understanding which foods benefit or harm health is vital work.

Also, the public keeps growing increasingly thinking about finding ways to boost health through diet. Obesity and diabetes are actually highly prevalent, and both have nutritional risk factors. It has sharpened general interest further.

All areas of scientific research face the following issues to a greater or lesser degree, but because nutrition is indeed on top of people’s agenda, the issues appear magnified.

Although the water is muddy and difficult to traverse, there were substantial victories in neuro-scientific nutrition research. For example, scientists have determined that vitamin C prevents scurvy, that beriberi develops because of a thiamine deficiency, and that vitamin D deficiency causes rickets.

In all of the cases, there exists a link between a specific compound and a particular condition. However, the picture is rarely so clear-cut. This is also true when investigating conditions wherein multiple factors are in play, such as for example obesity, osteoporosis, diabetes, or cardiovascular disease.

Also, nutrition-related conditions have changed as time passes: The most frequent threats to health used to be deficiencies, whereas in Western countries today, overeating is commonly the primary concern.

Understanding the role of food in health insurance and disease is vital and deserves attention. In this feature, we discuss a number of the reasons that nutrition research appears to be so indecisive, difficult, and downright confusing.

In a perfect world, to understand medical impact of confirmed food - goji berries, for example - an experiment would go something similar to this:

Scientists recruit 10,000 participants (both males and females, from a variety of nationalities and ethnicities) and house them in a laboratory for 10 years. The scientists feed each individual the exact same diet throughout their stay, with one difference: Half of the participants consume goji berries surreptitiously - perhaps blended right into a mixed fruit smoothie.

Alcohol and tobacco are banned throughout the study.

The participants must exercise for the same amount of time every day; if some persons exercised more, they could become healthier, regardless of their goji berry intake. This would skew the data.

Neither the researchers nor the participants are aware of who is acquiring the goji berry smoothie; if the participants knew these were receiving a “superfood,” they could take advantage of the placebo effect. This so-called double-blinding is vital when running clinical trials.

Through the decade-long study, the scientists monitor the participants’ health intensively. This may involve running regular blood tests and medical imaging.

Of course, the astronomical cost of this sort of study is the initial obstacle. Also, ethics and common sense say that is beyond impossible.

Nutritional research must make some concessions, as an ideal study is unachievable. So, in “observational studies,” nutrition scientists search for links between just what a person consumes and their current or future state of health.

Observational studies could be incredibly useful. Like this, scientists proved that tobacco causes lung cancer and that exercise is wonderful for us.

However, these studies are definately not perfect.

One issue with observational studies may be the researchers’ reliance on self-reported food intake. They ask participants to note down everything they eat for a set period of time, or even to remember what they ate previously. This could make reference to yesterday or months earlier.

However, human recall is definately not perfect. Also, some persons might purposely miss certain foods, such as their third candy bar of the day. Furthermore, participants usually do not always know the precise size of their portions, or the entire list of ingredients in restaurant or take-out foods, for example.

Studies often ask questions about the long-term impact of a nutritional component on health. However, researchers have a tendency to take dietary information at only a couple of points in time. The truth is, people’s diets can change substantially over the course of a decade.

The issues connected with measuring nutrient intake are so ingrained that some authors have described self-reporting as a pseudoscience.

These issues prompted an extremely critical study, which appeared in the journal PLOS One, to pull apart data from the National Health insurance and Nutrition Examination Survey (NHANES).

The NHANES, which started in the 1960s, “is a program of studies made to assess medical and nutritional status of adults and children in the United States.” Experts utilize the findings to steer public health policy in the U.S.

The primary approach to data collection for the NHANES are 24-hour dietary recall interviews. Researchers utilize this information to calculate energy intake.

The authors of the critical paper conclude that “the opportunity to estimate population trends in caloric intake and generate empirically supported public policy relevant to diet-health relationships from U.S. nutritional surveillance is extremely limited.”

In an judgment piece, lead author Edward Archer pulls no punches, explaining that their paper demonstrated “that about 40 years and several huge amount of money of U.S. nutritional surveillance data were fatally flawed. In […] nutrition epidemiology […], these results are commonplace.”

Here, we meet up with the double-edged sword of industry: The PLOS One paper declares that funding for the critical study “was provided by an unrestricted research grant from The Coca-Cola Company.”

Industry funding certainly will not invalidate the findings of studies, nonetheless it should prompt us to wonder what the funder might gain from such research. In cases like this, a company that produces sweet drinks might benefit from destabilizing people’s faith in the research which has deemed their products unhealthful.

Perhaps this example is a little unusual; more commonly, an industry with a vested interest will fund studies that demonstrate the benefits associated with a product.

As an example, the California Walnut Commission regularly fund research concluding that walnuts are good for us. Meanwhile, one study supported by the U.S. Highbush Blueberry Council proudly states in its abstract:

“[I]t is widely agreed that the regular consumption of tasty, ripe blueberries can be unconditionally recommended.”

For more on this topic, read our article on the sugar industry and how it manipulated scientific discourse in its favor.

To reiterate, if a report secures industry funding, it generally does not mean that persons should dismiss the findings out of hand. However, it should provide food for thought.

Another study in the journal PLOS Medicine looked at the impact of industry funding of research into carbonated drinks, juice, and milk.

The authors conclude, “Industry funding of nutrition-related scientific articles may bias conclusions in favor of sponsors’ products, with potentially significant implications for public health.”

To increase the confusion, media outlets tend to amplify these findings. If, for example, a report funded by a chocolate manufacturer concludes that chocolate extends life, media outlets will faithfully reproduce the conclusions, often with out a reference to the funders or a discussion of the study’s limitations.

Another issue that appears to plague nutrition research is complexity. Sometimes, a report will focus on the impact of just one specific meal or compound on a health outcome. This is slightly much easier to manage.

Often, however, studies will attempt to investigate the impact of a specific diet. The Mediterranean diet, for example, has gained a lot of attention over recent years.

The difficulty here's that one person’s version of the Mediterranean diet could be very different from another person’s version. For instance, one person might have one small glass of red wine, 25 olives, and an artichoke every day, while another person - who may be in the same experimental band of a report - might drink no wine and eat neither olives nor artichokes.

A related issue is that of replacement: If someone eats no meat, for example, they will probably replace meat with other sources of protein, such as for example beans or pulses. So, when comparing diets which contain meat with those that usually do not, any health effects might not be due to the insufficient meat but the addition of other foods.

All sorts of fruit and vegetable contains a vast array of compounds, and the sort and amount of the can vary based on where they grow, how persons transport and store them, and how they process and cook them.

There are so many variables to take into account that even when a report does find a statistically significant result, it really is difficult to determine if it actually originated from the meals under investigation.

Of course, humans are simply as different as the foods they consume. Eating an individual peanut might provide one person with beneficial nutrients, while that same peanut could possibly be fatal for someone with an allergy.

One 2015 study makes the same point however in a more subtle way. The scientists consistently measured 800 participants’ blood glucose levels and found “high variability in the response to identical meals.”

The authors explain that suggests “that universal dietary advice may have limited utility.”