How can we make good food choices? Is it enough to follow common sense, the advice of a family member, to listen to our body, or to eat like our ancestors? In fact, nutrition researchers have been eagerly trying to answer all of the questions that we could want to know about food for a long time. But can we trust nutrition science? The field of nutrition has been wrong before, so why should we trust current nutrition advice?
What is the scientific method?
Nutrition is a type of science and like any other kind of science, nutrition follows the scientific method. The scientific method is a continuous cycle involving observation, questioning, research, hypothesis, experimentation, analysis, conclusion, and reporting. This is a systematic method to gain knowledge. The observation step involves seeing or sensing an existing phenomenon. The question step is the subsequent curiosity following that observation. For example, why did that observation occur? The research step involves learning as much as we can about the subject. This includes reading related published research or textbooks. In order to prevent us from reinventing the wheel and redoing what has already been discovered, we need to check what work has already been completed in that field. We endeavour to identify and fill in the the knowledge gaps that exist about that topic. The hypothesis step involves making informed guesses about what will happen based on what we have learned. What do we expect the answer to our research question to be based on current knowledge?
The experimental step involves designing an experiment or simulation to answer our question. We need to control for confounding variables as much as possible. If we don’t control these confounding variables, then we won’t be able to determine the true cause of the results that we will observe. For example, in one experiment, if we give a person some broccoli and steak at the same time and that person becomes unwell, we cannot from that experiment alone identify which food may have caused the sickness. The analysis step involves thinking carefully about the results of the experiment and interpreting their meaning. If our results did not match with our hypothesis, we may not conducted the experiment well or maybe we discovered something new. In many cases, the experiment needs to be redesigned to produce better results. But if our results still don’t match with the hypothesis, then we have to create and test a new hypothesis. The conclusion step involves summarising what we have found. The reporting step involves presenting the results in a report, article, or presentation. People can then judge our work, suggest improvements, and learn from the findings. This cycle goes on and on repeatedly. The results of our experiment raise new questions. Questions which we can try to answer with the next experiment.
What is a journal article?
Scientists usually try to publish their work in academic journals. This work is usually published as a journal article. When a scientist publishes an article, other scientists can learn from their work and design experiments to try to answer the subsequent research question. Journal articles usually consist of an abstract, introduction, methodology, results, discussion, conclusion, and references section. The abstract summarises the objectives, methods, and results of the article. This section is important, but if we only read the abstract and nothing else, we may miss important details or nuances within the article. The introductory section introduces and justifies the problem, question, or topic to be examined. It also defines important terms, and cites and summarises previous research in that area. After reading the introduction, we should have a good understanding of why the research was done. The methodology section describes and justifies the process that was followed. If someone else wanted to replicate the experiment, they should be able to do so by using the described method. The results section typically involves presenting the measured data in graphs or tables. The discussion sections talks about the implications of the results and the flaws in the experiment. Usually, the results are compared to previous journal articles and the significance of the results is judged. The conclusion section summarises the most important findings and suggests further research. The references section lists the sources cited within the article.
What is peer review?
If a scientist wrote an article and published it immediately, a lot of misinformation could become public. That is why we need a fact checking process. Peer review is a process in which several researchers working in a similar area but at a different academic institution judge the article. Respectable journals will always request peer review before publishing an article. The reviewers give advice to the journal regarding whether to publish or reject the article, and give recommendations to the authors on how to improve it. The editor makes the final decision on whether or not to publish the article based on the peer review outcome.
The goal of the peer review process is to reduce bias and ensure the scientific quality of the article. Usually, the writer and the reviewer are anonymous to each other. Still, peer review isn’t perfect. The reviewer may also be biased, or they may not pick up on the bias of the article’s authors. There are still articles out there today with some misinformation. But that is not a good reason to not believe in science in general. So, instead of focusing on a few bad articles, looking at the quality of all the available evidence can help us find more useful information, identify trends, and get closer to understanding reality. Science is not based purely on one article. Science is the collection of knowledge about a subject, ranked by quality. Science is more valid than evidence that has not been systematically examined.
The quality of an article and the reputation of a journal are not always related. Researchers may seem completely objective, but at the end of the day they are still human and academia can be subjective and political. Some journals prioritise publishing articles that contain interesting results. Some articles may attract less readers and fewer citations from other scientists if they don’t find interesting results or a large effect, even if they follow the scientific method perfectly. However, not finding an effect is just as important as finding an effect. We need to be aware of variables that don’t matter too. In general, papers from highly reputable journals are of higher quality and are evaluated through a more rigorous peer review process. However, instead judging an article based on the journal that it is published in, it is a better approach to evaluate the quality of the paper itself.
Predatory journals do not use peer review. If a scientist pays the journal, the article will be automatically published. Of course, this is unethical. However, scientists are aware of these journals and usually do not cite them. You can see a list of potential predatory journals here.
After an article is published, an author, reader, or editor may find mistakes or misinformation within the article and hence the article may be retracted. Experts usually have a sense of what results are possible and if they come across questionable results, they can report this to journal. Additionally, if the results of an article cannot be replicated, it may indicate that the results are incorrect.
Can we trust science?
Before discussing whether we can trust nutrition science, we must first discuss if we can trust science itself. You can typically find some flaws in any study, scientific predictions have been wrong in the past, and scientific advice keeps changing. So can we trust science at all?
A lot of people do not trust science these days. Especially during the pandemic, people were suspicious of the government’s science based decisions. Is vaccination helpful or dangerous? Is it necessary to wear a mask? If we didn’t isolate in our homes, would a lot of people have died? Because of this, many people may began to doubt governments, scientists, doctors, the pharmaceutical industry, health research, and epidemiology.1 Social media and the internet have also increased our ability to access a variety of information. This is a good thing. However, many people now also have access to misinformation, and may believe it if they cannot identify that it is false. The way in which the pandemic was managed is not representative of science as a whole. During the pandemic, many decisions were made based on non peer-reviewed pre-print articles.1 Because it was a relatively new field, there were no strong reviews or meta-analyses summarising all of the available evidence. Scientists and politicians had to make complex choices quickly based on limited data. In contrast, in areas of science such as nutrition, scientific advice comes from a wealth of long-term data, and many scientists have been judging each other’s work for a very long time.
Of course, we should always question science. Asking questions is a fundamental aspect of science. But we should also be suspicious of our own challenges and questions if we don’t have better evidence or information than current science. Until we learn a lot about a subject, we should have low confidence in the validity of our doubts. Only after we have examined the evidence in its entirety can we come to appropriate conclusions. Before then, our thoughts don’t take into account every aspect of the field, so they are just an opinion, and could very well be wrong.
The current state of science is usually different from the state of science reported by the media. One day the media claims that bacon is healthy, but another day the media claims that bacon is unhealthy. Of course, this is very confusing for the viewers. The average person may think that the scientists simply do not know what foods are healthy, or that most scientists disagree. The media’s recommendations about nutrition and science are always changing, but the reality is that scientific advice is much more stable and consistent. The media typically promotes interesting or controversial studies in order to attract more viewers, but one study does not represent the whole of nutrition. A true scientist looks at the evidence as a whole, considers the quality of the evidence, and comes to an overall conclusion. There are some corrupt scientists who spread misinformation to the media, typically because they are paid by industries or because they are heavily biased. However, these people do not represent scientists as a whole.
The constantly changing nature of science or the existence of studies with conflicting results, is not a good justification to neglect science as a whole. In fact, the ever changing nature of science is not a disadvantage, but rather an important advantage. Science is a process that seeks to continuously improve. We can always expand on the knowledge we have discovered before, or we can add more nuance. Over time, theories will be proven wrong and new theories will emerge. That is progress. Science does not guarantee 100% accuracy, but it is the process of systematically reducing uncertainty. People may prefer black and white, all or nothing, or certain and clear recommendations, but that is not reality. The goal of nutrition is not to give absolute advice, but to provide us with ways to reduce our chances of developing disease. The more we want simple answers and the more we try to reduce nutrition into simple rules, the more we miss the complex and nuanced nature of reality.
Science may not be perfect, but it is the most useful guide we have. Our intuition and personal anecdotes do not control confounding variables very well. Without science, we would just be doing everything subjectively and doctors would give us advice based only on what they think is best. In contrast, science seeks to answer a question by systematically reducing bias. Of course, modern science does not know everything and cannot answer all questions. But the more we experiment, the more we will discover later. Following scientific advice is the safest approach rather than making decisions based on more biased sources.
Our thoughts and attitudes toward science do not change the validity of science. Our thoughts and attitudes may change how we respond to scientific information, but the facts are still facts. Overall, whether we believe science or doubt it, science is our best estimate of reality. In many cases, reality and science are quite close. If intuition and common sense were enough, scientists would already know everything. While our personal thoughts and feelings are important and should be considered, they should not replace scientific evidence.
Can we trust you?
I’ll admit, I am not a nutritionist or a nutrition researcher. So, can you believe a thing I say about science and nutrition? I am not a nutrition expert, but I have gained the skills to understand nutrition research through a lot of self-study and by working as a chemical engineering researcher. I always check what I have learned with expert advice. However, the information regarding nutrition in this blog is not my personal advice, but rather a summary of the results from the nutrition research. As I mentioned, information from research is the most trustworthy information we have. So instead of trusting me, trust the many thousands of nutrition researchers who worked to produce those results.
The concept of cherry picking is to only present evidence that supports your position and simultaneously ignoring papers that do not align with your position. To prevent cherry-picking, I endeavour to read and discuss papers that challenge my position. But at the end of the day I am only human and have my own set of biases. I try to be aware of these biases and reduce them while writing. So if you would like to challenge any of the information in this blog, please do so. An important feature of science is checking each other’s results. It is always a good approach to directly check any person’s claims with other sources. This blog is no exception. It is also a good approach to consult a health professional whenever making health related choices.
Does it matter who funds a study?
If an industry provides funding for an experiment, should we automatically doubt or dismiss it? In fact, a lot of useful research has come from articles that have been funded by industries. Without industry funding, science would not have developed to the level that it is today. So funding is not necessarily a bad thing. Instead, whether a study is funded by the plant based industry or the livestock industry, we should always take a closer look at the methodology, analysis, and conclusions of an article more carefully. In particular, if there are many absolute claims that support the funding industry’s interests, the paper should be examined in more detail. In the field of nutrition, there are few absolute results and usually more conditional or nuanced results, so absolute claims are a red flag. We also need to make sure that the experiment is well-designed and that there is no bias in the methodology that favours results that an industry wants to achieve. We also need to make sure that the authors have fairly compared their results, included all of the important statistics, and that the abstract and conclusions are truly representative of all the results from the study.
Papers that are funded by industry are more likely to show positive outcomes for the industry’s interests than papers that are not funded by industry.2,3 If an industry doesn’t like the outcome of a study, they can choose not to publish it. So there is a positive bias in the articles that do get published. Fortunately, many journals and research grants these days do not tolerate this kind of interference, and the authors of an article have to declare that the industry that funded them did not affect the outcome or the decision to publish. In addition, the authors of the article must also declare any conflict of interests. For example, if a scientist works for the livestock industry or owns a farm, they must declare that within the article. Every researcher has personal biases. Vegan researchers also have their own personal biases. However, a researcher with a conflict of interest can still conduct science objectively. So, we shouldn’t automatically ignore the work of these scientists.
But I am different from the participants in nutrition studies, right?
We all have a unique biology and research does not typically test our exact individual circumstances. So, is this a good justification for disregarding experiments conducted on others? The results from high-quality journal articles provide general information about which variables will affect our health. From these results, we cannot accurately predict the risk of an individual developing a disease due to a specific variable, but we can make reasonable predictions about a larger population. If we don’t possess specific knowledge about our individual biology, it is reasonable to assume that we are just like the average population. We may feel healthy, but we may not actually be healthy on the inside. It is important to listen to the signals from our bodies, but they can be misleading and ignoring the science altogether is a bad approach. We may not be able to follow all general nutrition advice if we suffer from allergies, food intolerances, or other specific conditions. However, we can consult with a dietitian to create the most evidence based diet that takes into account our particular case.
Should we just do our own research?
Of course, it is a good habit to check information and maintain a healthy level of suspicion. However, nutrition science is complex, and if we don’t understand the nuances of the field, if we don’t know how to judge the quality of a journal article, if we get our information from sources other than journal articles, and if we don’t understand or interpret an article correctly, we can easily come to incorrect conclusions. We can also easily fall into the trap of following false logic or try to find evidence that confirms our bias. If we want to question science, we first have to take into account all of the current evidence and provide higher quality counterevidence. Therefore, the most reliable approach may be to follow expert advice or read consensus journal articles. In later posts, I will explain how to do your own research properly and the basic skills required. If anyone wants to invest enough time, they can do their own research well. However, if we don’t invest enough time, we can easily be misled.
Conclusion
In this post, I introduced the scientific method, the structure of a journal article, and the process of peer review and publication. Even though science is not perfect, it systematically attempts to reduce bias and uncertainty. It is important for us to do our own research, but if we don’t have the time, knowledge, skills, or tools, we can unintentionally make misinformed choices. The following post will introduce different kinds of experiments and journal articles common to nutrition sciences and answer the following important questions. What is sufficient evidence? Are all studies equally valid and useful to us?
References
1. Kreps SE, Kriner DL. Model uncertainty, political contestation, and public trust in science: Evidence from the COVID-19 pandemic. Science Advances. 2020;6(43):eabd4563.
2. Fabbri A, Lai A, Grundy Q, Bero LA. The Influence of Industry Sponsorship on the Research Agenda: A Scoping Review. American journal of public health. 2018;108(11):e9-e16.
3. Lesser LI, Ebbeling CB, Goozner M, Wypij D, Ludwig DS. Relationship between funding source and conclusion among nutrition-related scientific articles. PLoS medicine. 2007;4(1):e5.