Why does this matter?

Because millions of people follow advice that probably does not help many, possibly harm some, and proliferates distrust in science.

What’s next?

When research done on other people doesn’t give reliable information, you need to become your own research subject. Fortunately, the necessary technology already exists.

The Analogy

 
The catholic church institutionalized the belief that earth was at the center of the universe. When Bruno, Galileo and Kepler challenged this belief, the church placed their heretical writings on the index of prohibited books. That is, books that nobody of the church’s flock was supposed to get in front of their eyes.
 
We all know how that exercise in information suppression worked out.
 
Fast forward to today. And to our topic, for which I will use the red-meat controversy as an illustrative example.
 

The Gospel

 
You probably know the gospel about red meat:

Eating it increases your risk of heart disease, cancer, and death. It is enshrined in nutrition guidelines all over the world.

It is based on irrefutable scientific evidence, at least that’s what the gospel’s evangelists are telling the public.

And it is heresy to proclaim otherwise.
 
So much so that when an international group of eminent researchers published a paper that essentially says “forget that gospel” [1], the evangelists’ knee jerk reaction was to call for the editors of the journal to retract the paper.
 
Does that sound familiar?
 
Among those voices were, predictably, the American Heart Association and the American Cancer Society.

The Physicians Committee for Responsible Medicine, a group known for their openly proclaimed anti-meat bias, even filed a petition against the offending journal with the Federal Trade Commission.

Quite obviously, our today’s anti-red-meat evangelists haven’t learnt much from history.

Suppressing evidence has never made anyone the good guy in history, or the winner.
 
Though for heretics to survive it is always a good idea to arm themselves with sterling credibility and irrefutable evidence. That’s what those red-meat heretics did.

The Heretics

 
They were a 14-members panel of eminent researchers from 7 countries, and the offending journal none less than the Annals of Internal Medicine, one of medical sciences’ top journals.
 
What they essentially told the evangelists was that their evidence is too thin and of so low quality that it fails scientific standards.
 
That was not just an opinion.
 
In life science we have the GRADE system (Grading of Recommendations, Assessment, Development and Evaluation) to, well, grade studies. There are 4 grades for evidence: very low, low, moderate and high.

The red-meat gospel failed GRADE miserably. On most aspects the evidence for health and survival effects of red-meat rated as “very low”.

The rest as “low”.

This ”evidence” should not have made it into guidelines. Particularly when those guidelines are lorded over many millions of people.
 
That was in 2019.
 
Since then the chorus of heretics has grown larger and louder. In October 2022 another team of 18 researchers published their “burden of proof” study on the same subject: Health effects associated with consumption of unprocessed red meat [2].

A burden of proof study seeks to methodically address the question: “What evidence is required to support this claim, and can we provide that evidence through rigorous scientific inquiry?”

In other words, the authors started from the evangelists’ claim, and worked their way backwards, first asking what evidence is required to support their anti-red-meat claims, and, finally, whether the presented evidence matches these requirements. Spoiler alert: it doesn’t.

Here is their conclusion:

“While there is some evidence that eating unprocessed red meat is associated with increased risk of disease incidence and mortality, it is weak and insufficient…” [2].

This time the heresy was published in Nature Medicine.

The “Nature” pedigree and an impact factor north of 80 makes this journal one of the top addresses to publish your paper.

To get published there your paper needs to pass not only the editor’s desk but a rigorous peer review process that eliminates 2 of 3 submitted papers.

So, if there had been any flaw in the methods which led the authors to their conclusion, their paper wouldn’t have seen the light of day.
 

Reflective of the evangelists’ desperation about their crumbling credibility is the comment of one of them:

“it’s not only nutritional science that people want to have weighed in the balance — we’ve also got things like climate change, we’ve got things like environmental destruction, we’ve got things like basically humane treatment of animals” [3].

The Moral Issue

 
Aaah, so that’s what it is about: a moral issue.

Now, why would we, the people, need anyone – no matter how academically credentialed he or she may feel – tell us what we “want to have weighed in the balance”.

We can damn well decide that for ourselves.

Moreover, issues of climate change, environmental impact and the morality of animal husbandry are of no concern to the enquiry into the health effects of eating red meat.
 
While I fully accept anyone’s decision to abstain from eating anything that involves the breeding of methane farting cows, I abhor the idea of a scientist injecting their moral compass into an answer to a straightforward science question: does eating red meat make me sick or die early?

So, when will you finally get the correct answer to your question whether you should or should not eat red meat?

After all, this is science, and science can settle such things, right?

Wrong.

Science has discovered that a universally valid answer – one that applies to you and me and everyone else –  is unlikely to be found. Ever. Particularly not if we continue to do nutrition research the way we have been doing it all along.

Because of one phenomenon:
 

Vibration of Effects (VoE)

 
If that does not ring a bell, you are in very good company. Many, if not most, of my colleagues haven’t heard of this effect either.

If I had to describe it in one sentence, here it is:

When using different statistical models to analyze relationships between one variable (such as red meat consumption) and another variable (survival) produces different conclusions, then you have vibration of effects.

It is an academically more palatable definition for whatever-you-want-to-find-in-your-data-you-can-find.

It was John Ioannidis who coined the term [4].

He is a professor of medicine, of health research and policy, and of statistics at Stanford University. And he is director of the Stanford Meta-Research Innovation Center, whose purpose is to advance excellence in scientific research.

Ioannidis has made it his quest to investigate how reliable published research really is. His rather sobering conclusion – that most published research is either false or useless [5] – is based on pure mathematics and statistics.
 

Back to his brainchild VoE.

 
Nutrition research is particularly prone to this phenomenon. Here is why:

There are 2 ways of doing research: interventional and observational.

The latter involves observing and analyzing individuals or groups of individuals in their natural settings without intervening or manipulating any variables.

Observational research aims to understand associations between variables, such as the association between red meat consumption (the predictor variable) and the development of cancer or heart disease or early death (the outcome variable).

The key word here is “association”, as opposed to “causation”. Because no observational study can determine causation.

Causation is the domain of interventional – or experimental – research. It has its own credibility issues, and those I will discuss in another post.
 
Back to observational studies: Nutrition research almost exclusively relies on them to answer questions, such as those about the association between red meat and, say, survival.

If this study design doesn’t uncover causal relationships, simply increasing the number of observed people will not change that.

That’s like turning up the volume on a radio that’s playing static.

And that’s exactly what nutrition researchers do. They create very large participant pools.
 
One of these is NHANES – the  National Health and Nutrition Examination Survey. It is a program conducted by the National Center for Health Statistics (NCHS), which is part of the Centers for Disease Control and Prevention (CDC) in the United States.
 

The NHANES program began in the early 1960s as a series of surveys designed to assess the health and nutritional status of the U.S. population.

Its primary goal is to provide nationally representative data on various health-related topics, including chronic diseases, infectious diseases, environmental exposures, dietary patterns, and health behaviors.
 

To Ioannidis NHANES has become the tool to prove his point of “vibration of effects”. NHANES presented him a treasure trove of data for altogether 417 different variables and their association with survival.
 

Here is what he did

 
When analyzing the relationship between any one variable and survival one must not ignore other variables that also affect survival. Being older or male is always a worse hand in the survival game. But there are lots of other variables, too.

The more variables there are, the more different relationship models you can construct. Ioannidis limited his investigation to just 13 variables.

Using simple combinatorics tells you that 13 variables can be combined in 8,192 different ways. These different ways represent the different statistical models that researchers could apply to the data.
Which is what Ioannidis did for all 417 variables of the NHANES data pool
 

Now here comes the clincher:

 
For most of the variables the different models produced different outcomes. And for fully one third of the variables (Vitamin E being one of them) the different outcomes ranged from extending lifespan to shortening it.

That means: for most of the associations between a variable and survival the conclusion depends very much on what the researchers want to read into the data.

Aren’t there variables for which all models come to the same conclusion?

Yep. Vitamin D was one of those. No matter which model Ioannidis applied, Vitamin D was always associated with a beneficial outcome.

But before you stock up on Vitamin D supplements I want to caution you: It has been called the Charlie Brown of supplements. Remember? Whenever Charlie Brown sets up the ball for the kick of the century, it never happens.

So it is with Vitamin D. In observational studies it often looks like the savior, for anything from heart disease to COVID. But in unbiased randomized controlled interventional studies it always flames out.
 

In a nutshell

 
Ioannidis’ work was an elegant statistical proof that our large observational data sets about nutrition and survival can support the most contradictory claims.
Unfortunately, observational datasets are the only ones we have.
 

Back to the red-meat issue

 
While red meat consumption was not part of Ioannidis’ NHANES data set, his discovery of vibration of effects applies likewise.

Because for practically every nutrient, the amount of intake correlates (positively or negatively) with the intake of multiple other nutrients and with environmental exposures and lifestyle habits [6].

If any one or more of the variables that correlate with red meat intake happen to have a genuine association with the risk of death, cancer or heart disease then red meat will also seem to have that association whether it is real or not.
 

What does all this mean for you?

 
You’ll never get a reliable answer to the question how eating or not eating red meat will affect YOU. Not from any of the studies that nutrition researchers fancy for their claims

Why?

Because the complexity of interactions between variables that we see in population-based studies, does not even come close to the complexity WITHIN YOU:

All of those variables additionally interact with your unique genome, microbiome, and metabolism.

Moreover, none of those variables are stable over time. You change your diet, your activity levels, your environment, whatever.

What does Ioannidis recommend?

 
Given his discovery, he has 2 recommendations (among others, but those are not relevant here):

First, he suggests scientists present VoE checks when they publish nutrition research results.

Against the backdrop of a VoE check it becomes much clearer how reliable the researchers’ claims or conclusions are.

To make that easy he and his team developed a statistical tool to do so, and they put this tool into the public domain for every researcher to use free of charge.
 
Second, he suggests to switch from observational trials to large interventional ones.

He is aware of the logistics challenges and the high costs. But he also envisages compensatory savings from curtailing observational trials and pooling smaller interventional trials into larger ones.
 

So, has anything changed?

 
Not really.

Scientists still slug it out over nutritional recommendations, exemplified by the latest disputes over red-meat consumption.
 
While the latest Global Burden of Disease study inflates the risk from red-meat consumption by a factor of 36 (compared to the previous GBD study of 2 years earlier) [7], the blowback came promptly from another team pointing out methodical issues and saying that

“…it would be highly inappropriate and imprudent for the GBD 2019 dietary risk estimates to be used in any national or international policy documents, nor in any regulatory nor legislative decisions.” [8].

 
The anti-red-meat evangelist have been perfectly aware that, as far as the correlations of red meat with survival, cancer and heart disease are concerned, the evidence is weak and of low or very low certainty [9] [10] [11].
 

So what are you supposed to do?

 
To blindly follow belief-colored nutrition advice, is, in my eyes, not an option.

After all, nutrition guidelines are there to ensure lifelong health and function.

Should a red-meat fan deny him- or herself that pleasure when, in the end, it might not carry any benefit at all?

The solution that I recommend, and practice, is to turn oneself into an interventional trial.
 

You explore how any dietary or other lifestyle habit affects one or more of those biomarkers that tell you today already how your future health will be.

The tool for doing so is called N-of-1 method. It is the method recognized as the gold standard for medical single-case experiments.

This short video explains how it works and here (www.adiphea.com) you find how my team and I have, for the first time, operationalized this method.

It takes all the guesswork out of nutrition advice.

The objective is to make you a professional for your own personal body and health. Independent of all the evangelists, self-proclaimed gurus and experts out there who want to tell you how to live.

Or should you wait until nutritional science has been infused with Ioannidis’ recommendations? And attempts to force morality tainted beliefs on you has become a thing of the past?

If experience serves as an indication, I wouldn’t bank on that. It took the catholic church until 1966 to abandon their Index of Prohibited Books.
 

PS

If you want to try our “Lifestyle Navi”, let me know. If you have a real health concern that fits to what we do at adiphea, and if you have what it takes to address this concern through lifestyle, and you
 

PPS

But whatever you do, stay skeptical when anyone tells you the gospel about what you should or should not eat.

hashtags

• #RedMeatControversy
• #MeatMyths
• #MeatScience
• #NutritionScience
• #NutritionMyths
• #NutritionFacts
• #NutritionGuidelines

References

• [1]      Johnston BC, Zeraatkar D, Han MA, Vernooij RWM, Valli C, El Dib R, et al. Unprocessed Red Meat and Processed Meat Consumption: Dietary Guideline Recommendations From the Nutritional Recommendations (NutriRECS) Consortium. Ann Intern Med 2019. doi:10.7326/M19-1621.
• [2]      Lescinsky H, Afshin A, Ashbaugh C, Bisignano C, Brauer M, Ferrara G, et al. Health effects associated with consumption of unprocessed red meat: a Burden of Proof study. Nat Med 2022;28:2075–82. doi:10.1038/s41591-022-01968-z.
• [3]      Hamilton professor calls criticism of his red meat study “hysterical” and “extreme.” CBC 2019. https://www.cbc.ca/news/canada/hamilton/hamilton-professor-calls-criticism-of-his-red-meat-study-hysterical-and-extreme-1.5307128.
• [4]      Patel CJ, Burford B, Ioannidis JPA. Assessment of vibration of effects due to model specification can demonstrate the instability of observational associations. J Clin Epidemiol 2015;68:1046–58. doi:10.1016/j.jclinepi.2015.05.029.
• [5]      Ioannidis JPA. Why most published research findings are false. PLoS Med 2005;2:0696–701.
• [6]      Trepanowski JF, Ioannidis JPA. Perspective: Limiting Dependence on Nonrandomized Studies and Improving Randomized Trials in Human Nutrition Research: Why and How. Adv Nutr 2018;9:367–77. doi:10.1093/advances/nmy014.
• [7]      Murray CJL, Aravkin AY, Zheng P, Abbafati C, Abbas KM, Abbasi-Kangevari M, et al. Global burden of 87 risk factors in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet 2020;396:1223–49. doi:10.1016/S0140-6736(20)30752-2.
• [8]      Stanton A V., Leroy F, Elliott C, Mann N, Wall P, De Smet S. 36-fold higher estimate of deaths attributable to red meat intake in GBD 2019: is this reliable? Lancet 2022;399:e23–6. doi:10.1016/S0140-6736(22)00311-7.
• [9]      Zeraatkar D, Han MA, Guyatt GH, Vernooij RWM, El Dib R, Cheung K, et al. Red and Processed Meat Consumption and Risk for All-Cause Mortality and Cardiometabolic Outcomes. Ann Intern Med 2019. doi:10.7326/m19-0655.
• [10]    Vernooij RWM, Zeraatkar D, Han MA, El Dib R, Zworth M, Milio K, et al. Patterns of Red and Processed Meat Consumption and Risk for Cardiometabolic and Cancer Outcomes. Ann Intern Med 2019. doi:10.7326/m19-1583.
• [11]    Han MA, Zeraatkar D, Guyatt GH, Vernooij RWM, El Dib R, Zhang Y, et al. Reduction of Red and Processed Meat Intake and Cancer Mortality and Incidence: A Systematic Review and Meta-analysis of Cohort Studies. Ann Intern Med 2019. doi:10.7326/M19-0699.