Researchers Rush to Criticize Paper for Claiming Association Between Meat and Type-2 Diabetes

Read Time:7 Minute, 20 Second

One might say the ideological battle over the optimal human diet has devolved into something like a cafeteria food fight.

Factions promoting the inclusion of red meat, animal products, and saturated-fat-containing foods have formed something of a coalition in response to the same by those favoring a plant-based diet. Each side cites their coalition partners as references, and every time a scientific paper is published, the others rush to blogs, newspapers, and academic journal editors’ email inboxes to try and issue a riposte.

So it was last week in response to a paper published in a scientific journal that has become a hive of plant-based diet advocacy—The Lancet. Its title is “Meat consumption and incident [of] type-2 diabetes” and was a meta-analysis of 1.97 million adults and 100,000 cases of type-2 diabetes from 31 study cohorts in 20 countries.

Fundamentally it delivered a very slight association between meat consumption and type-2 diabetes, but has been criticized for being of low quality, failing to account for numerous possible corruptions in the data, and most seriously, issuing recommendations for public health policy on the findings, and providing senseless hypotheses to explain the potential association.

Critics say that it features numerous amateurish anomalies like incorrect arithmetic, inaccurate definitions of “meat,” miscommunication of relative vs absolute risk, and other such issues.

No scientist under normal circumstances would be bewitched into thinking that the data presented in the study could be enough to base policy on, or really to conclude with anything other than “more research is needed,” but these are not normal circumstances. They are an ideological battleground in which billions of dollars in profits (or losses) for food, pharmaceutical, and agricultural companies hang in the balance.

The true depth of this failure of modern science stretches back to the 1950s with the Diet-Heart Hypothesis of Ancel Keyes, but the evolution of it is well-summarized in this 2021 article from Men’s Health.

PICTURED: Blaming the meat for what the bread did. PC: Eaters Collective

Blaming meat for what the bread did

The study discovered an associative hazard ratio (relative risk) for 100 grams of unprocessed red meat of 1.10, or 10%, while 50 grams of processed meat yielded 1.15, or 15%. Relative risk is used often in epidemiology in circumstances when absolute risk is difficult to ascertain, such as the case with this paper in which 31 studies of different kinds were pooled together. However, the consensus in good epidemiology is that a relative risk factor has to be close to 100% for there to be any consideration that the observed effect may be caused by the controlled-for factor, in this case, meat.

So this study was nearly four times below the point at which an associative hazard ratio should be considered plausible. This didn’t stop the authors from writing in their abstract: “The consumption of meat, particularly processed meat and unprocessed red meat, is a risk factor for developing type 2 diabetes across populations. These findings highlight the importance of reducing meat consumption for public health and should inform dietary guidelines”.

“We woke on Wednesday 21st August to the news that ‘Eating just two slices of ham per day could raise diabetes risk’ alongside a picture of a ham sandwich,” writes Zoe Harcombe, Ph.D. and front-line thinker on this ideological battleground. “I tweeted ‘blaming the ham for what the bread did'”.

Though she makes a good point. The researchers in their paper didn’t define meat, and included sandwiches and lasagna as processed and unprocessed meats respectively—another red flag in a study that would typically be identified by the authors themselves if this was an honest scientific discussion.

Another red flag was the Healthy User Bias. Harcombe explains.

“The highest red meat consumers have higher BMIs, they are less physically active, they are more likely to be current smokers and they are less likely to take multivitamins etc. There is always a healthy person confounder. The burger/hot dog consumer is less healthy than the quinoa/kumquat consumer in many ways,” she wrote on her blog as part of a takedown of the meat and type-2 diabetes paper.

Whether you have a Ph.D. or not, there is something that jumps out right in front of you when looking at the paper—that there’s no glucose in red meat; how can it therefore lead to diabetes which is a glucose-handling disease?

Another major issue identified by Harcombe, is how much actual meat consumption was used to estimate risk.

“I likened what the researchers had done here to—if, across 31 studies, people averaged fewer than 50 cigarettes a week, the researchers decided they were going to estimate the risk of smoking 100 cigarettes a week,” she wrote.

She furthermore added that the calculative value of determining how different each paper in the study was from all others likewise included was 65%—a remarkable difference, and indicative of including the findings of a study of 141 Asians eating pork belly and wok stir-fried vegetables with 141,000 Americans eating burgers with fries, rhetorically speaking.

“The researchers failed to propose a plausible mechanism for the absurdity that something that contains no glucose (meat) can somehow be associated with a glucose-handling condition (T2D),” Harcombe concludes. “The plausible mechanism is that meat is eaten with carbs—e.g. a ham sandwich or a burger, bun & fries—and this paper is condemning the meat for what the carbs did”.

Making a mountain out of a molehill

The conclusion of the authors of the paper is striking, that their findings highlight the importance of reducing meat consumption for public health and “should inform dietary guidelines”. The last four words are particularly compelling, that national dietary guidelines should be set according to such a petty scientific exercise.

In 2019, a monstrous blow was struck by those who might agree with Dr. Harcombe—that researchers too often blame meat for what the bread did—in a series of 6 papers that were presented as a rigorous process to create recommendations, rather than simply conducting science.

Science was indeed conducted in them, however, including a review of randomized trials involving 54,000 people that found low- to very low-certainty evidence that diets lower in unprocessed red meat have any effect on major metabolic health markers (including type-2 diabetes), and a dose-response meta-analysis from 23 cohort studies with 1.4 million participants that found just the same.

“The panel suggests that adults continue current unprocessed red meat consumption (weak recommendation, low-certainty evidence). Similarly, the panel suggests adults continue current processed meat consumption (weak recommendation, low-certainty evidence),” the review paper, entitled Unprocessed Red Meat and Processed Meat Consumption: Dietary Guideline Recommendations From the Nutritional Recommendations Consortium, concluded in its abstract.

In the piece from Men’s Health referred to above, the writers spoke with David L. Katz, M.D., M.P.H., Ph.D., founding director of plant-based advocacy group True Health Initiative, and general plant-based advocate, who described the moment he and colleagues on his side of the ideological battlefield saw and read the series of papers as something akin to “Defcon 1” according to the interviewer.

“We started calling one another and saying in effect… Holy shit, this is not for print,” Dr. Katz said in the piece from 2021. “We thought [it was] going to hurt a lot of people”. Hurt a lot of people, or convince a lot of people?

In any case, Dr. Katz wrote to the journal that published the 6 papers, Annals of Internal Medicine, an extremely prestigious journal, and requested the papers be held prior to their publication for an independent and rigorous review. Dr. Katz felt that the authors were not seeking to perform rigorous science, but merely to leverage the journal’s reputation to create “alternative ‘guidelines”.

Katz described it as a provocation. “If they’d just published the [data] and not the guidelines, it would have been a yawn from us. But to devise guidelines directly at odds with your own findings and pretend like that’s business as usual… this is a provocation”.

If they were at odds with their own findings, it’s nothing that the type-2 diabetes study taken apart by Dr. Harcombe wasn’t attempting to do. Both sets of papers received significant and similar criticisms from their detractors on the other side, and bear witness to the difficulty today in attempting to posit a pattern of human eating behavior as healthy. WaL