A recent study analyzed the impact of red meat on human health, finding that the increased risks associated with red meat are small, with weak evidence to support them. Cutting down on red meat wouldn’t really do much for most people, the study reported.
The media loved it. “Red meat is back on the menu” and “Stop worrying about red meat” were just some of the headlines flown around, validating what meat lovers all around the world wanted to hear. But things aren’t really that simple. In addition to some questionable reporting by some parts of the media, there is also great dissent in the scientific world about the results of this study, and a group of researchers believes it shouldn’t have been published in the first place.
Red meat is popular for one simple reason: it tastes good. It’s also easily available in the developed world and has become a staple in many cultures. But let’s go back to existing science on red meat.
Study after study has found that red meat is bad for you, in a number of ways. It raises the risk of cancer, diabetes, obesity, cardiovascular disease — you name it. Furthermore, studies have consistently found that replacing red meat with plant-based foods increases longevity and improves health.
We’ve written about it extensively. Studies seem to be in general agreement. The evidence seems so convincing that health and dietary guidelines in many countries have adapted and are now recommending a lower intake of red meat.
But this study claims otherwise.
This new review, on the other hand, found that the association between red meat and all these problems was very weak — so weak, in fact, that the differences might be attributed to something else.
From the get-go, this study was unusual. Not just because it was controversial, but because a group of unrelated researchers took the extremely unusual step of trying to stop the paper before it was published, claiming that the methods are flawed and the published results can have significant and negative consequences.
Professor Walter Willett is Professor of Epidemiology and Nutrition at the Harvard School of Public Health and lead author of the EAT-Lancet Commission, which advocates a plant-based diet for both environmental sustainability and health. He called this report “the most egregious abuse of evidence that I have ever seen,” adding that it has “has layers of flaws”.
Nevertheless, the authors of the new study stand by their work — at least some of them. First, let’s see how the study was conducted.
What the study found
First of all, it wasn’t a singular study with one piece of evidence. Instead, it was a series of systematic reviews that analyzed the results of previous studies and tried to draw one common conclusion. The authors worked to gauge the impact that eating red meat has on health, particularly in regard to the 2-4 weekly servings average in the US and western Europe.
In one review of 12 studies on 54,000 people, researchers did not find any significant association between meat consumption and the risk of diabetes, cancer, and heart disease. In three other reviews of cohort studies following millions of people, they did find a reduction in risk for people eating 3 or fewer servings a week, but they note that the association is very weak. For instance, cutting three servings a week would result in seven fewer cancer deaths per 1,000 people (which is still something, but was regarded as weak evidence).
In the fifth review, they analyzed why people eat meat, finding that people eat meat because it’s tasty, they are reluctant to change their diet, and for some, because they view it as healthy.
“Thanks very much, but I’m going to keep eating my meat,” said co-author Dr. Gordon Guyatt of McMaster University in Canada.
The other authors were, while less assertive, also supportive of their results. Bradley Johnston, PhD, corresponding author on the reviews and guidelines and an associate professor of community health and epidemiology at Dalhousie University, said that the team realizes that their work is contrary to many nutritional findings, but he still sticks by it.
“This is not just another study on red and processed meat, but a series of high quality systematic reviews resulting in recommendations we think are far more transparent, robust and reliable,” he said.
Johnston also added that there are other concerns regarding red meat — particularly environmental and ethical — which were not considered in the study.
“We focused exclusively on health outcomes, and did not consider animal welfare or environmental concerns when making our recommendations,” Johnston added.
“We are however sympathetic to animal welfare and environmental concerns with a number of the guideline panel members having eliminated or reduced their personal red and processed meat intake for these reasons.” Guyatt himself noted that he usually avoids red meat for these concerns, and 3 of the 14 authors said they support reducing red and processed meat consumption regardless of their health impacts.
Why the differences
There are two main reasons why the findings of this series of reviews are so different from previous studies.
The first has a lot to do with the very nature of the reviews. Such reviews, while very important to scientific progress, are also very delicate. Gathering different studies and putting them in the same pot isn’t exactly a straightforward process. Different studies can have varying methodologies, varying data quality, and they may follow somewhat different parameters. Bringing them all together in a common language is a very challenging task — but that’s just part of the problem. In addition, although they are essentially statistical measures, systematic reviews are also a lot about how the data is interpreted. It’s not unheard of to take several studies which individually found that a drug is effective and produce a review which concludes that the drug isn’t effective. Going into the complexity of such results happen goes way beyond what can be explained here, but let’s just say that while reviews work to eliminate the biases in individual studies, they are themselves subject to inherent biases.
The second reason for these differences is still a matter of contention — basically, a group of researchers believes this is some shoddy science. This group, which features several high-profile US researchers, was so firm in its reaction that it demanded a delay in the paper’s publication.
The main point of debate here is whether or not the evidence is strong enough. Previous studies say it’s strong enough, while this one says it’s too weak to say anything. The evidence here hasn’t changed one bit — it’s how the evidence is interpreted.
“What we need to do is look at the weight of evidence – that’s what courts of law use,” said Dr. Walter Willett, a professor of nutrition at Harvard University who was among those calling for the papers’ publication to be postponed.
It’s extremely uncommon for researchers to ask that a paper’s publication be delayed and further reviewed, even after the official peer review has passed. Usually, when something like this happens, researchers wait for the paper to be published and then try to disprove it. But in this case, scientists feared that the message will dilute healthy nutritional advice and cause significant damage. Even one of the reviews’ 14 co-authors asked for a delay.
The biggest issue here is why some studies were included and why others were excluded from the analysis. The analysis also ignores studies finding that switching from a meat-based diet to a plant-based diet offers significant health benefits — which is an important piece of evidence here.
The journal, Annals of Internal Medicine, defended the publication of the study, saying that preventing an article from being published is not how scientific discourse should be carried out, but this did not dissuade the researchers combatting the results of this study.
Harvard’s Dr. Frank Hu noted that about a third of American adults eat at least one serving of red meat a day — questioning why 2-4 servings a week was chosen as a comparison baseline. Dr. John Ioannidis, a professor of medicine at Stanford University, said such advice (to not cut down on red meat) can distract from clearer and more effective nutritional advice. Willett was extremely blunt in his statement. Dr. Giota Mitrou, Director of Science External Relations at the World Cancer Fund, also rejects this study.
“It could be putting people at risk by suggesting they can eat as much red and processed meat as they like without increasing their risk of cancer.”
“The message people need to hear is that we should be eating no more than three portions of red meat a week and avoiding processed meat altogether. We stand by our rigorous research of the last 30 years and urge the public to follow the current recommendations on red and processed meat.”
So where does this leave us?
We all want a simple and clear take-home message — but in modern studies, especially in nutrition, that’s rarely the case. Science is rarely black and white, and it’s rarely about simple answers. Instead, science is more about different shades of gray and asking better questions — that is the main takeaway here.
The study paints an interesting counterpoint to existing studies on red meat. It represents a huge and robust analysis which will, no doubt, be discussed for years to come. However, some of its premises and interpretations are, at the very least, debatable.
The authors themselves acknowledge that the evidence is “low” or “very low certainty”, which means that the results could be unreliable. Furthermore, since an important part of the society eats much more than 2-4 servings a week, even a small reduction in risk of common health outcomes such as cardiovascular disease and cancer could make a big difference at a population level.
A Professor of Medicine and Human Nutrition at the University of Otago, New Zealand, offers a balanced opinion which sums things up nicely:
“In my opinion, the weak recommendation based on low-certainty evidence that adults continue current consumption of unprocessed red meat and processed meat is potentially unhelpful and could be misleading.”
OK, a take-home message for real now
Let’s try another conclusion.
Eating meat is bad for the environment, that much is clearly true. It produces a lot of emissions, uses up a lot of water, and it kills billions of animals each year. Eating meat also promotes the development of drug-resistant antibiotics. This alone is reason enough to reduce or eliminate red meat from your diet.
Meat can be a significant source of protein, vitamins, and minerals. However, most guidelines still recommend not consuming more than 70 grams a day — and that’s not likely to change soon. A balanced diet can include protein from meat, as well as from non-animal sources such as beans and pulses. Plants are definitely safe.
Health-wise, red meat is still probably bad for you. It’s not exactly clear just how bad it is — it is probably a substantial risk of cancer or CVD diseases, but there is a chance that the risk is extremely low. The weight of arguments leans towards the former, but the latter just go an important boost. If you do eat red meat, don’t eat too much of it, and especially be sure to avoid processed meat.
Later edit: conflict of interest
After the publishing of this article, a NY Times investigation revealed some potential conflicts of interest which draws even more question marks about this study.
Dr. Johnston signed a disclosure form that he did not have any conflict of interest over the past three years — but as recently as December, 2016 he was the senior author on a similar study that tried to discredit international health guidelines advising people to eat less sugar. As per the NY Times:
Johnston claims that the money was from 2015, so it wasn’t within the past 3 years. This, while may be technically correct, violates the spirit of disclose — it’s a particularly relevant detail which should have been mentioned in the study, says Marion Nestle, a professor of nutrition, food studies and public health at New York University who studies conflicts of interest in nutrition research. This is particularly telling since researchers did include their personal eating habits in the study, listing it as a potential bias. Surely, such an association would warrant mentioning, even though the paper would have probably been published nevertheless.