We could categorize our time as the age of information. Every day, a modern human consumes, on average, the equivalent of the information a person during the 15th century would be exposed to throughout their entire life. From many perspectives, this is a wonderful thing. For example, knowledge has been democratized, and now anyone can access it from their smartphone. However, the abundance of contradictory ideas and content makes it difficult to discern what is real or true from what is not, and here, science has played a complicated role. As an example, in the field of nutrition, it is common to find visceral clashes between diametrically opposed groups claiming to have the absolute truth, supported by "irrefutable scientific evidence." This reminds me of past wars between European states where both sides prayed to the same God to help them defeat the enemy. In fact, this analogy is much deeper than it seems at first glance. My idea with this article is to demonstrate that although science is an excellent tool for approaching the truth, it is also an inherently human activity with flaws, biases, and problems that must be taken into account when analyzing the validity of its claims. Therefore, using it as the unquestionable basis for a set of dogmatic ideas not only represents a complete contradiction of the scientific method and a major barrier to expanding the boundaries of knowledge but also a source of radicalization, polarization, and misinformation.
On one hand, we can define science as a branch of knowledge that encompasses the set of objective and verifiable knowledge about the natural or physical world obtained through observation and experimentation, the formulation of hypotheses, and their verification through further observation and experiments. This procedure is known as the scientific method and is what sets science apart from other human disciplines. One of its basic assumptions is that no hypothesis can be accepted as absolutely true. In this sense, experiments or observations that support a theory simply reaffirm its validity as a plausible model of reality. On the contrary, strong negative evidence is enough to discard a hypothesis and seek a better one. Evidently, having a healthy dose of skepticism and accounting for different positions and interpretations is essential when reviewing experimental data and validating a theory.
On the other hand, and I believe many people forget this, science is an inherently human activity. As such, it is susceptible to distortions caused by personal, political, economic, and social interests, which means that any assertion coming from science should be taken with caution and not considered an absolute and irrefutable truth. We must start by remembering that science is an expensive activity that does not bring immediate rewards. This implies that there has to be funding and, therefore, a conscious or unconscious pressure to obtain certain results aligned with the financier's agenda, be it the state, the private industry, or a wealthy individual. Secondly, science occurs in laboratories and institutions operated by human beings with biases, interests, and imperfections. The selection of an experimental design and data collection involves subjective decisions, and it is very difficult (if not impossible) for the experimenter to distance themselves from the expected results, even if they have good intentions. Third, although science uses the objective language of mathematics to analyze its results, the choice of analysis methods and their interpretation is highly subjective. While statistics provide an objective probability that the hypothesis is false given the obtained results, the decision of whether that number represents a significant value ends up being more subjective and highly determined by the quality of data collection, experimental design, and the truthfulness of model assumptions and distributions used.
It is true that science has mechanisms to counteract these problems, such as conflict of interest declarations, peer review processes, double-blind studies, and independent verification of results in other laboratories. However, given the large number of scientific articles published and the inevitable diversity in their quality, it is clear that a significant portion will have some of the mentioned flaws. Furthermore, we cannot forget that even at the institutional or societal level, widespread paradigms and biases can be generated that are difficult to identify and can lead an entire discipline toward an incorrect model of reality. In these cases, peer review is not useful, as most reviewers agree with the same idea, and the few who dare to question it are relegated to ostracism and public ridicule.
My intention is not to promote relativism and invalidate all scientific claims. It is evident that science has taken us to a level of development and civilization never before seen and that, despite the mentioned flaws, it is an extremely powerful and useful tool. Just look at the advancements in semiconductor physics and the consequent rise of computing, the precise theoretical predictions about the fundamental properties of subatomic particles, or the understanding of how cells, DNA, and evolution work, to name just a few examples. However, there are some areas of science that are particularly susceptible to the mentioned distortions due to their complex nature and direct connection to humans. A clear example of this, which I will focus on in this text as a case study, is human nutrition, although other examples could be climate change or other areas of medicine.
The human body is an incredibly complex system, where an astronomical number of molecules interact in multiple ways to generate various emergent phenomena impossible to study solely at the molecular level given our limited theoretical and computational capabilities. For this reason, disciplines such as psychology, medicine, or nutrition emerged, seeking to understand humans at different levels of complexity. The compartmentalization and simplification generated by these branches proved to be very useful for many years, but the risk of oversimplification and losing sight of humans as a whole whose parts interact is very high. Although it is very tempting to seek simple explanations for complex problems, cause-and-effect relationships in human physiology are difficult to establish, and often a phenomenon is due to multiple interrelated causes. Additionally, humans are diverse and complex, and a change in one parameter can imply unexpected and varied consequences among individuals, so scientific conclusions must be understood in the context of statistics as averages and not as specific recommendations.
In particular, human nutrition is a field where finding conclusive and high-quality scientific studies is really difficult. To understand why, it is important to define what a controlled, randomized, double-blind experiment is; the gold standard for testing a scientific hypothesis. Controlled means that a portion of the study subjects is not given the treatment to be able to compare their results with those who do receive it. Randomized means that the assignment of who receives the treatment and who belongs to the control group is entirely random. Finally, double-blind means that both the subjects and the experimenters do not know who is receiving the treatment and who is in the control group.
While it is possible to conduct controlled and randomized nutritional experiments, comparing, for example, different dietary interventions, it is by no means easy to carry them out. Firstly, there is an ethical limitation since human subjects cannot be arbitrarily administered compounds that could potentially put them at risk, and they cannot be kept in controlled environments for an extended period. Additionally, many of the phenomena studied can take several years to manifest, making the studies impractical and very costly. Furthermore, it is virtually impossible to control and keep constant the rest of the variables with which the subjects interact during the study, guarantee adherence to the program's recommendations, or prevent premature dropouts. On the other hand, it is impossible to blind the subjects to prevent them from knowing whether they are receiving the treatment or not. After all, they are fully aware of what they are eating. Finally, in nutrition, you cannot simply remove a food; you have to replace it with another to meet the daily calorie and nutrient requirements, making it complicated to discern whether the effects were due to the elimination of the first food or the inclusion of the second.
This causes most controlled nutrition studies to be of short duration, involve very few individuals, and/or have significant flaws in the implementation of the experimental design. It also makes it necessary to use animal experimentation as points of comparison, which can provide clues but not necessarily reflect human reality. All of this has led many scientists to turn to another type of study known as epidemiological or observational studies. In these studies, large groups of people are surveyed about their diet and lifestyle, and these data are compared with the incidence of certain diseases to determine whether there is a correlation. These studies are much more practical and economical from a practical point of view since they do not require strict control of variables, and no treatment needs to be administered to people. The problem is that they lose the essential control and randomization elements mentioned earlier. This means that any result coming from observational studies cannot be established as a cause-and-effect relationship but only as a correlation, meaning that when one is high, the other is too, but not necessarily that one caused the other. Additionally, it has been shown that the accuracy of survey formats filled out by individuals is very poor since it is generally difficult to remember the foods consumed and their quantities, and people tend to unconsciously fill in values biased toward what they consider healthy.
As an example, let's take the purely epidemiological result that correlates the consumption of red meat with a higher incidence of colon cancer, widely publicized in the media and official recommendations. Essentially, what these studies show is that, on average, people who eat more red meat also have a slight increase in the risk of colon cancer. Now, the key question is whether this is because some compound inherent in meat causes the disease or if there is another logical explanation for its correlation. One possible answer comes from what is known as the healthy user bias vs. unhealthy user bias. Since the 1950s, the idea that animal products are not good for health has been introduced into North American culture, and by extension, the rest of the world. Because of this, people today who identify as healthy have mostly adopted a diet high in vegetables and have greatly limited red meat consumption. These same people probably exercise more, do not smoke, drink moderately, do not consume processed products, have access to better health services, and are more obedient to health recommendations. What's interesting is that in Asia, where the cultural paradigm presents meat as a symbol of social status and abundance, epidemiological studies show the opposite: people who consume more meat are on average healthier. On the other end of the spectrum are people who do not care much about general health recommendations. They eat more red meat, but they also smoke, drink, probably engage in riskier activities, and, in general, care less about their health. Furthermore, they probably consume meat as part of a very unhealthy diet, rich in vegetable oils and processed cereals (McDonald's!). Is it the higher probability of death due to meat? Or maybe everything associated with its consumption? These studies cannot tell us, but it is most likely that simply because culturally in the West we were convinced that meat was bad for health, healthier people eat less meat, and hence the statistics show the aforementioned correlation.
While epidemiology is not entirely useless, it can never provide information about causation, only correlation. This can serve to formulate hypotheses and then test them in a controlled randomized study. In the case of meat, the first controlled and randomized study concluding that meat causes colon cancer has not been published, and yet epidemiological studies have been used over and over again by vegan advocates, government entities, and the processed food industry to justify general recommendations and dietary changes without having solid evidence, a clear display of irresponsibility.
Finally, it is important to highlight that the results arising from these studies directly affect industries and public policies and therefore face significant pressure from different social sectors. It is very common to find studies funded by industries or specific associations with a clear agenda behind them. Given their economic power, they can influence the design of studies, manipulate results with statistical tricks to make them appear relevant, inflate results using relative percentages instead of absolutes, and decide not to publish results if the study eventually turns unfavorable.
Returning to the broader point, it is evident how biased, misinterpreted, or scientifically motivated results can be passed off as objective and unquestionable under the protective cloak of science. The human desire to belong to a tribe, to share ideals and fight for them, has led many groups with extremist ideologies to misuse science as the foundation of their unquestionable truths. They forget that the very nature of science goes against any form of dogmatism, and that scientists are essentially skeptical, curious, and open-minded. The biggest problem is that most people are ignorant of the different types and qualities of studies, so they either believe everything that claims to be scientific, or they fall into relativism, thinking that all science is useless because there is an article to support every claim.
In the face of this landscape of excessive contradictory information and the difficulty in finding quality science, it is crucial to question the validity of any result, no matter how scientific it may be, especially in the media, social networks, or government recommendations. It is important to consult other sources, read the original articles, determine the type of study, its methodologies, sources of funding, etc., to form an educated opinion. We cannot forget that science has been wrong time and time again; this is normal and necessary. For this reason, we must maintain an unrelenting quest to expand our knowledge and question current theories and reevaluate them in the light of evidence with an open mind and without falling into dogmas.
In any case, high-quality scientific studies do exist if one looks for them, and it is valid to rely on their results to make better decisions and recommendations. In the field of nutrition, science will show us more and more interesting results every day, and we will be able to better understand the role of food in our bodies, but in some cases, this probably will take many years due to the complexity of the object of study and the bioindividuality of people. Therefore, the key is to take responsibility for our own health and do what is natural and evolutionarily sensible. We must connect with our bodies and listen to what feels good and what doesn't. We do not need a study to know what gives us diarrhea or headaches or what fills us with vitality and makes us feel radiant. Just listen to our bodies, use common sense, and look back to see how our ancestors lived before the advent of all modern diseases. When in doubt, I suggest that any recommendation involving a substance or activity that has not been tested by evolution for thousands of years should be considered potentially harmful until proven otherwise. In this regard, the ancestral lifestyle and diet are an excellent starting point to reconnect with our most human essence and from there conduct our own scientific experiments to find our truth. If you want me to accompany you on this journey, do not hesitate to schedule a free coaching session. I'll be waiting for you!
Bibliography:
Comentários