Can we trust research?

Understanding different types of studies is crucial when evaluating scientific research, or even when we ourselves are evaluating a bold headline in Dagbladet. Not all studies are equally good, and each type has its strengths and weaknesses.

Observational studies, for example, which nutritional science is literally drowning in, can show possible patterns and connections, but they cannot prove causal relationshipsImagine this:

You are walking down the street, and someone is walking behind you. It may seem like they are following you, but this is just your interpretation—an observation. Does this prove that they are following you? No. What you have done is made a hypothesis based on something you have seen. You have no proof.

This is a clear example of the thinking error known as post hoc ergo propter hoc – the assumption that because one event follows another, the first must have caused the second.

This thinking error is common because humans naturally look for patterns and connections, even where they don't exist. It's part of how we understand the world, but it often leads to erroneous conclusions.

The next step is then to find good ways to prove it, and test it out thoroughly. Not least, sit down and ask yourself; could there be other reasons why this person is following me? Maybe you're just going the same way?

These questions seem to have been completely omitted from modern dietary research, and it really takes very small magnifying glasses to find all these errors, intentional or unintentional, that are hidden behind the headlines.

Science and the art of disproving oneself

At the heart of the scientific process lies an often misunderstood principle: Having formulated a hypothesis, the goal of the researcher is not to prove the right one, but to actively try to disprove it.

A hypothesis is, after all, just a personal creation – the researcher’s own idea of ​​how the world might work. But ideas, no matter how compelling they may seem, must be thoroughly tested, and the process of science is designed to challenge hypotheses in all possible ways.

This is where some fields, such as nutritional science, have been criticized. Too often, studies in these areas seem designed to confirm hypotheses rather than critically test them.

For example, a researcher might hypothesize that a specific food reduces inflammation and then structure a study to show this effectInstead of searching for evidence that might disprove the hypothesis, they can focus on confirming it—a practice that introduces confirmation bias and undermines the reliability of the findings.

True scientific rigor requires experiments that expose hypotheses to their greatest vulnerabilities. By attempting to disprove their own ideas, scientists gain a deeper understanding of whether their ideas actually hold water.

If a hypothesis survives repeated attempts to disprove it, we can have greater confidence that it is valid. This principle is the foundation of all robust science.

How controlled is the study?

To move beyond hypotheses and test causal relationships, we use more robust study designs, such as randomized controlled trials (RCTs). RCTs are considered the gold standard in research because they use randomization to minimize bias and isolate variables.

For example, in a clinical trial testing a new drug, participants may be randomly assigned to either receive the drug or a placebo. Neither the participants nor the researchers know who is in which group, a process known as double-blinding.

This eliminates bias, and ensures that any differences can be attributed to the treatment itself, rather than external factors, or human expectations.

But the controls in these studies have their limitations. In real-world scenarios, it is often difficult to perfectly isolate variables. A study may control for age, gender, or health status, but unexpected factors, such as diet or stress levels, can still affect the results.

Furthermore, the more controlled a study is, the less it may reflect real-world conditions, which limits how broadly the results can be applied.

And in nutrition, it gets really complicated, because even if you make just one change that you want to study in your subjects, humans will have thousands of variables that you can't control for. Adjust a little, yes, but don't control.

Systematic reviews and meta-analyses

At the top of the evidence pyramid we find systematic reviews og meta-analysesThese combine data from many studies to provide a broader understanding of a topic.

They are often considered very reliable because they summarize large amounts of research, but they are only as good as the quality of the studies they are based on. If the included studies are of poor quality – or simply “garbage” – we just end up with a big pile of bad data. It may look so big and impressive, with perhaps dozens or hundreds of studies, but it all depends on all the included studies being of high quality themselves.

This is the classic principle of “garbage in, garbage out.” A meta-analysis cannot therefore compensate for errors, biases or weaknesses in the underlying studies. The original studies must therefore be carefully assessed to determine whether the meta-analysis actually has value.

Especially in nutritional science, there is a form of “inbreeding,” where old, corrupt studies are constantly being returned to and recycled as the basis for “new studies.” These new studies are often based solely on old numbers and findings—numbers that may not even have been reliable to begin with.

Finally, such consonant articles point to hverandre as "sources", forming a real echo chamber, and everything is left hanging in the air, without any real anchoring.

An illusion of consensus is created – a convincing facade, but one that lacks any solid basis.

The role of ethics in study design

Another critical part of research is ethics. In the past, studies were often conducted under conditions that we now consider deeply unethical. For example, research was previously conducted on vulnerable groups, such as prisoners or institutionalized individuals, who were unable to give truly informed consent.

The "positive" thing was that they actually had full control over the individuals, and that they followed the study's principles to the letter. But it's clear that it's not okay to treat people like guinea pigs and laboratory rats.

These practices were condemned and led to the development of modern ethical standards in research, such as the Declaration of Helsinki and the Belmont Report. Fortunately, today, researchers are bound by strict ethical guidelines that require informed consent, protect the rights of participants, and ensure that studies do more good than harm.

Still, it's worth reflecting on what this means for research on humans – and especially research on the food we eat. Because such studies must respect ethical boundaries, they will always be weaker than other types of research.

Human behavior, diet, and lifestyle are simply too complex and uncontrollable for us to ever achieve the same precision as in laboratory studies. Perhaps answers about the human animal and our food would be more solid and correct if we found them in evolutionary history instead?

Peer review

Even with the best questions and ethical guidelines, science relies on a system of checks and balances – peer reviewThis process involves other experts in the field reviewing the study's methods, results, and conclusions before publication.

Peer review acts as a safety mechanism, but it is not perfect. Poorly designed studies can slip through, and journals with lower standards may accept work that would not have been accepted in higher-ranked publications.

You can also end up, in this dietary echo chamber, acknowledging each other's studies because they coincide with an opinion you already had, and it becomes a self-fulfilling "truth" that is never actually tested.

Asking the right questions

When evaluating scientific evidence, it is crucial to ask critical questions:

• How was the data collected?
  
• Was the study well designed?
  
• Were ethical guidelines followed?
  
• Were the best questions asked – and for the right reasons?
  
• Whose financed the study, and what were their reason?
  
• Was the research subject thoroughly peer review?
  
• Did the researcher try to disprove his hypothesis, that is test it thoroughly, or just confirm den?

Science thrives on transparency, scrutiny, and a willingness to challenge assumptions. By understanding the complexities of study design and execution, we can navigate the often contradictory world of research with greater confidence and clarity.

SOURCES

Photo: Shutterstock license

Anyone can comment by logging in with their Facebook or Google accounts.. Norwegian Carnivory is not health care. You must seek medical attention for medical problems.

Written 2025 © Norwegian Carnivory