Learn more about Peter Attia’s digital program
A few months ago, the Annals of Internal Medicine published 6 papers by the same group of investigators — part of the Nutritional Recommendations and accessible Evidence summaries Composed of Systematic reviews (NutriRECS) — investigating the impact of unprocessed red meat and processed meat1“Red meat was defined as mammalian meat, and processed meat was defined as white or red meat preserved by smoking, curing, salting, or adding preservatives,” the investigators wrote. on cardiovascular disease, cancer, and all-cause mortality.
The first paper, “Unprocessed Red Meat and Processed Meat Consumption: Dietary Guideline Recommendations From the NutriRECS,” was a summary article about the dietary recommendations from the NutriRECS consortium. To support this position, the investigators published 4 parallel systematic reviews, and a final paper on attitudes toward meat consumption.
Three of the 4 reviews (here, here, and here) looked at observational studies, with the remaining paper reviewing randomized controlled trials.
This group of investigators ultimately recommended that adults continue eating the current levels of red and processed meat while acknowledging that this was a “weak recommendation,” based on evidence that provides little certainty. Much of that evidence is in the form of observational studies, and are limited because of confounding (e.g., the healthy user bias), lack of accuracy in design (e.g., the use of unreliable food frequency questionnaires), and because they inherently cannot determine causation. Essentially, the investigators concluded that the evidence for adverse outcomes associated with meat consumption is too weak to recommend reducing its consumption.2“Although statistically significant,” the investigators wrote, “low- to very low-certainty evidence indicates that adherence to dietary patterns lower in red or processed meat is associated with a very small absolute risk reduction in 9 major cardiometabolic and cancer outcomes (range, 1 fewer to 18 fewer events per 1000 persons), with no statistically significant differences for 21 additional outcomes observed.”
As you might expect, this recommendation faced criticism from public health authorities who have previously taken a hard line on meat avoidance, including the Harvard School of Public Health (HSPH). But what was unexpected, and reported in a perspective in JAMA, was that the editor-in-chief of the Annals, Christine Laine, received about 2,000 “vitriolic” emails before the papers were published. According to the perspective, David Katz, Walter Willett, and Frank Hu (the latter two both at the HSPH) contacted Laine asking her to pre-emptively retract the papers. (Here is the letter.)
Rather than repeat myself on the limitations of nutritional epidemiology and the behind-the-scenes ridiculousness that is going on with groups that are supposedly promoting good health and science, I highly suggest you read the perspective in JAMA, the Annals papers (at least the recommendations paper), and the HSPH’s response to the recommendations.
One more thing worth adding to this discussion: the perspective points to financial ties to industry as the exclusive conflict of interest, whether pro- or anti-meat. While this is certainly a problem, what often isn’t discussed are the nonfinancial conflicts of interest these authors have in terms of their ideological, intellectual, and allegiance biases. Most of us are biased when it comes to nutrition. We all are supposed experts who have countless hours of putting food into our own bodies. (Give credit to the authors for disclosing their eating habits in a supplement to the recommendations paper. You can find it in the “Supplements” section of the online paper.) It’s virtually impossible not to be biased in this field. Which is all the more reason we must raise the standard in terms of the rigor of research required for public health authorities to base guidelines and recommendations on. If we want this to be more like science and less like religion, we cannot rely on observational epidemiology.
In a viewpoint in JAMA, David Ludwig and his colleagues provide some suggestions for improving the quality of dietary research that I think make sense. I can’t help but agree with people like John Ioannidis who argue (as he did in JAMA) the field of nutritional epidemiology needs radical reform. In an interview with CBC News, Ioannidis summed it up succinctly: “Nutritional epidemiology is a scandal. It should just go to the waste bin.”
– Peter
Assorted delicious grilled meat with vegetable over the coals on a barbecue
Link to authors’ eating habits doesn’t work. “Access denied”
Good catch. It looks like the link expires, so your best bet is to go to the recommendations paper online and see the PDF in the “Supplements” section. I updated the post to reflect this. Thanks.
It is sad we have come so far and know so little about human nutrition. A bit strange anyone would be bias in this field as we should all want to eat what is best for us. Stranger still, I think the truth is known but too many careers would be shattered if it were widely known. Ugh!!!
“The truth is known” ..? Which truth? And how do you know it’s the truth? What evidence do you have for this ‘truth’
In my opinion there is no such thing as Nutritional Science currently – as the so called scientists seem to completely disregard any black swans linked to their firmly held beliefs.
A single black swan is enough to render any, say, physics theory dead and abandoned, but not so in Nutrition…
Until this can be addressed I will pay no attention at all to their biased ramblings 😀
Is there any study or case that shows reversal of blocked arteries on a diet containing animal products?
That is exactly what he points out. Arriving at decisions about cause and effect of different foods is impossible with observational comments. Randomised controlled trials are near impossible to carry out with humans as ,controlling ‘ a group is impossible , unless they are in a prison … and it has to be for years.. !
“ (Give credit to the authors for disclosing their eating habits in a supplement to the recommendations paper.)”
Link is bad for me. I just get a page that says, “AccessDeniedAccess denied”
Good catch. It looks like the link expires, so your best bet is to go to the recommendations paper online and see the PDF in the “Supplements” section. I updated the post to reflect this. Thanks.
I would love to read the perspective in JAMA, but it’s not available without a price to us lowly ones.
I really wish there was a solid, definite answer about meat consumption, especially as it pertains to those of us with MBC. I still eat red meat, albeit sourced from local ranchers, most of whom I know on a first name basis. I do avoid grilling the meat because I read that the “bark” of the meat is where to potential for cancer lies, and my gut tells me this is true. Whether my gut is correct or not, I can live without eating the crusty (burnt) bark of BBQ despite it’s incredible deliciousness. But my gut also tells me that the meat I eat is highly nutritious and is easy for me to digest, despite the medications I’m on that attacks the gut lining.
The “Annals” and the response from Harvard are both available to those of us without subscriptions.
Thank you for all you do! I first “met” you on Dave Asprey’s podcast and your reasoning with him on why he probably will not live 180 years was brilliant. I immediately hunted you down on the interwebs and have followed every since.
Try putting the paper title into Sci Hub. I found the JAMA perspectives article there – full access.
Thanks, Peter!
Do you know if it is true, and if it is why, the PREDIMED and the Diabetes
Prevention Program (DPP) studies were not included in the paper? And if these actually “satisfy fully the stated inclusion criteria” of the paper? Also, do you know which of the studies authors were “among the signatories below is one of the study authors” which disagreed with the conclusions and guidance and why they were in diasagreement?
Here are my best guesses:
DPP: NutriRECS excluded “Studies in which more than 20% of the participants were pregnant or had cancer or a chronic health condition, other than cardiometabolic diseases.” BMI is about 34 on average at baseline in all 4 groups. They are also on average prediabetic at baseline, based on their fasting glucose and OGTT.
However, I think obesity and T2DM can be classified as cardiometabolic diseases. It may have been excluded for the same reason PREDIMED was probably excluded. Subjects may not have been “allocated to consume diets that included varying quantities of unprocessed red meat (measured as servings or times/week, or as g/d) or processed meat…”
PREDIMED (original paper and republication after retraction): I think PREDIMED fails this criteria: “We included … reports of randomized trials of adults allocated to consume diets that included varying quantities of unprocessed red meat (measured as servings or times/week, or as g/d) or processed meat (meat preserved by smoking, curing, salting, or adding preservatives)…”
In PREDIMED, red meat was discouraged in both diets (and looks like more strongly discouraged for the control vs Mediterranean diet).
From the supplement in Estruch et al., 2013: “In the Control group, advice on vegetables, red meat and processed meats, high-fat dairy products, and sweets concurred with the recommendations of the Mediterranean diet, but use of olive oil for cooking and dressing and consumption of nuts, fatty meats, sausages, and fatty fish were discouraged.”
Probably not excluded for the following reasons, but noteworthy (and perhaps should’ve been excluded for these reasons): The original publication and design essentially didn’t give the control group the proper control/placebo for the first 3 years (i.e., a comparable treatment via low-fat diet, individual and group dietary-training sessions at the baseline visit and quarterly thereafter) of a 4.8 year study; Some villages were randomized collectively (i.e., a cluster randomized trial) in an individual randomized trial. This was not disclosed (or perhaps even realized) in the original publication (Estruch et al., 2013) nor was it accounted for in the analysis. In addition, they were assigning households to one group as well rather than individual randomization.
Thanks, Bob Kaplan!
From the JAMA article:
“What’s more, THI member John Sievenpiper, MD, PhD, also signed the letter to Laine even though he coauthored the NutriRECS systematic review about the relationship between meat consumption and all-cause mortality and the risk of cardiovascular disease, heart attack, and type 2 diabetes.
“Laine said she contacted Sievenpiper, a nutrition scientist at the University of Toronto, after receiving the letter and pointed out that he had signed a standard form affirming his agreement with his paper’s conclusions. That had not changed, he told her, but he did not agree with the guideline paper, of which he was not an author.”
And again – Thanks, Bob Kaplan. I appreciate your time and effort!
Thank you for this thoughtful article on the intellectual context of current nutritional research.
On July 11, 2018, after watching a conversation between Jordan Peterson and Joe Rogan, I spontaneously decided to try the carnivore diet. I was primarily attracted by Peterson’s claim that it had resolved a long-standing anxiety issue. In spite of a lifetime sugar addiction, I found it ridiculously easy to adopt and sustain. For 17 weeks I ate nothing but meat, salt and water, and about halfway through it became obvious that my own anxiety issues were completely gone. This wasn’t a marginal improvement, which I would have been grateful for, but a total eradication. I was astounded. Twenty months later, though I’ve since modified the diet, those gains are still intact.
Far from becoming an evangelical supporter of the carnivore diet, I look back on my experience with a renewed agnosticism. In truth, it was impossible for me to isolate all the variables in play. Which addition, or which subtraction, was decisive? The result was obvious, but the cause could only be guessed at.
I follow the current nutritional debates fairly attentively, but completely free of conviction about this or that approach. I’m only desperate to be right, and would like to think of science as an impartial ally on my journey. It’s both illuminating and disheartening to hear otherwise.
Thank you, Peter, for ringing the bell (one more time).
After reading almost all the referenced sources and links that you proposed in your article, I still have a mixed opinion about why this panel of “investigators” published this controversial “dietary guidelines recommendation”.
If there is something we should keep in mind, it is certainly the GRADE criteria used to evaluate all these studies (as hsph.harvard did mention) and that this panel declared “considerations of environmental impact” out of the scope of their recommendations.
For me, the negative impact of red meat and processed meat on our health is a fact but I’m getting really worried if and when the majority of the population when reading the press or watching the news, will NOT look beyond the headlines and abstract conclusions.
Have a great day,
Alexandra
I don’t think it’s that hard. There is no study that shows you can reverse clogged arteries when you put someone on a grass fed beef diet. Put them on a whole food plant based diet and studies show actual reversal. Should that not be the default diet? I mean, I love me some filet mignon with butter, but only when I switched to a whole food plant based diet did I see my t2 diabetes disappear, blood pressure disappear and cholesterol tank to healthy levels.
Great comment Sven. 100%.
I’d love to see the reference(s) for the reversal of clogged-up arteries on a plant-based diet. I’m 66 and despite following a very ‘healthy’ plant-based diet for >40 years, plus modest amounts of pastured eggs and cheese, I have pre-diabetes and slightly high blood pressure. I’m not yet on any medication but this may change. The blood pressure was above average even at age 40.
I think that some of these reactions may be so specific to individuals that it will be another >50 years before the medical profession/NHS works out a regime of how to test patients and what advice to offer on how that person can live a longer and healthier life.
Just testing blood glucose after meals of different carbs., e.g. wheat, chickpeas, potatoes, shows startling differences between individuals. That’s clearly within the limits of a ‘plant-based diet’, so some plants may be better for some people but not for others.
Financial conflicts of interest are usually construed as meaning that a scientist works for or gets money from some industry or stands to profit from some finding. However, there are other harder-to-quantify types of financial interests, involving things like ability to get very large government grants, grants from other entities, and even speaking fees and honoraria. If you want to get a lot of money from NIH, it is probably best to be a thought-leader and a widely acclaimed expert yourself. If someone publishes something that somehow threatens or undermines or challenges your views or your expertise, that can damage your ability to raise the large sums of money needed to support your research.
I think we all agree there are challenges in nutritional research. I think what’s frustrating is that articles like Ioannidis’ point out all the flaws and say it needs reforming but they don’t offer solutions or alternatives. I’m not an academic researcher myself but I’m pretty sure most people in that field aren’t driven by egos or self-interest and many want to do what’s right.