Reading Time: 8 minutes

Below is a discussion with Michelle Allison. Because I thought this was an interesting discussion from which I learned much, because I think Michelle is wrong, and because there are still some details to be worked out, I have challenged Michelle to a debate on my new podcast co-hosted with Avi Bitterman.

On this podcast, Avi and I will challenge individuals to debates who are prominent on social media in the nutrition science space. We will have a Challenge List that will provide for the reasons for these challenges, including posts documenting the reasons why we think the people that we are challenging are wrong, and why the challenge is important.

This is one of those posts.

To give context, Michelle Allison is a Health At Every Size advocate with 20K Twitter followers. She believes that trying to lose weight is psychologically harmful, for most people impossible, and does not provide any health benefit.

Others of her views can be found here:

These views are not supported by the scientific research, which I have documented here: https://nutritionalrevolution.org/2019/12/04/long-term-weight-loss-prevents-death/

This is the context within which I am arguing against Michelle.

Without further ado, here is the discussion.

This tweet was intended as a quote-retweet to someone who encouraged clinicians to abandon BMI charts because of the potential damage to patients because of stigma. Here is the original tweet:

Here is a comment on this tweet.

I am not certain about why my quote-retweet did not work. But what followed, I think, remains interesting and informative regardless of that detail, because it amounts, if I am right, to a general principle regarding what kind of advocacy is acceptable and what kind is not from the standpoint of modern medical ethics.

Without further ado…

(It should be noted that SDoH means “social determinants of health”, which is the phrase used in the tweet by Haley Goodrich, indicating that Michelle was aware of the tweet to which mine was referring. In any case…)

Here Michelle is referring to the principle of non-maleficence, a mainstay in modern medical ethics, seen here:

Here is a part of the abstract of this paper.

In particular, some of the “stigmatizing messages” were:

These are clearly not BMI charts. This research in fact has nothing to do with the argument that Haley Goodrich was making or Michelle Allison was defending.

The crucial arguments (to me) and the main purpose of this post are as follows:

  1. Some clinical practices are weakly supported by evidence, such as the use of information on the relationship between BMI and health, yet there is good reason to believe that they promote health;
  2. These should be challenged and tested scientifically to confirm that they do indeed promote health;
    • Carefully testing current, weakly supported practices helps to avoid violating the ethical principle of non-maleficence, by eliminating practices that may be harmful;
  3. Practitioners should not be dissuaded from engaging in these practices without evidence, as doing so might withhold benefit or withhold otherwise useful information from patients;
    • Withholding information that may be beneficial to patients violates the ethical principles of beneficence as well as autonomy.

For completeness’s sake, arguments that Michelle made were as follows:

(I decided not to answer Michelle’s question because the answer was already in the links that I sent: several individual RCTs were discussed as well as a meta-analysis. I wrote that blog post precisely to address the claims of folks like Michelle.)

Argument #2

This image has an empty alt attribute; its file name is Screen-Shot-2020-02-05-at-5.58.30-PM.png
This image has an empty alt attribute; its file name is Screen-Shot-2020-02-05-at-6.06.32-PM.png
This image has an empty alt attribute; its file name is Screen-Shot-2020-02-05-at-6.09.19-PM.png
This image has an empty alt attribute; its file name is Screen-Shot-2020-02-05-at-6.09.56-PM.png

Yet no response from Michelle to my counterargument. To be clear, Michelle is making the argument that the correlation between BMI and poor health outcomes is not actually from BMI itself but from other characteristics associated with high BMI. I made the point that many of those other characteristics (though not all) are caused by BMI itself, which is to some extent inherited.

Finally, after an initial bit of enthusiasm for the discussion, I grew a little tired of moving in circles, without any indication that Michelle was considering my arguments–instead, as one can see, the thread of discussion would just fall off and I was left to deal with new arguments. In debate, this is often called gish-gallop. There was a modest level of gish-gallop going on here, so I tweeted:

This image has an empty alt attribute; its file name is Screen-Shot-2020-02-05-at-6.18.23-PM.png

Because this discussion was in combination fascinating and frustrating, I decided that I wanted to formally issue a challenge Michelle according to the Challenge List system that I am using for my upcoming Diet Wars podcast. I wanted to do this in order to inform Michelle that she would be added to the challenge list.

This seemed to upset Michelle. My reasoning was that instead of jumping around from topic to topic and dropping arguments, we could get some kind of resolution over voice. She didn’t see things this way:

To which I responded:

This got Michelle going in ways that were still further regrettable:

It is important for me to emphasize that my motivation for pursuing this discussion was that I find the advocacy of people like Michelle likely harmful for the care of patients. I have no personal animus toward Michelle. It is her views that I find objectionable and harmful.

Michelle believes that weight loss is harmful, difficult to achieve, and/or that there is no evidence to support that any benefits accrue to those achieving weight loss; she promotes these views to tens of thousands of followers. Most or all of these views conflict with scientific consensus, and virtually nobody in the scientific mainstream believes them.

Michelle has a larger following than me, and her influence is much greater than mine. I believe that these beliefs (and her promotion thereof) are harmful and in many cases, such as when encouraging healthcare practitioners not to seek weight loss for their patients, unethical. Thus, I feel that it was important to push back and challenge what I believe is a harmful and unethical.

Furthermore, I believe that Michelle cannot support these beliefs with evidence or well-reasoned arguments. This post and the open challenge to Michelle for a discussion are intended to document the reasons for this belief on my part, in the hope that others can see how potentially problematic Michelle’s beliefs are from an ethical standpoint.

Other similar such challenges to individuals who reject science, as Michelle does, will be forthcoming in the coming months. It is not a “social dominance ritual”, except insofar as I believe that evidence and reasonable positions should be the socially dominant ones. I take some risk in posting this, and in particular openly posting my own mistakes in this exchange. I take this risk because I believe this is an important issue, and because I believe that it is important to highlight that Michelle Allison, one of the most prominent HAES advocates on Twitter, cannot seem to support her own positions when subjected to critical scrutiny, positions that if adopted by her followers, could lead to worse health and potentially result in an earlier death.

Regardless of the bad blood that this exchange engendered, I learned a great deal through the formal arguments exchanged, which apply to much that is argued by HAES advocates–in particular, these arguments impugn the claim that weight loss should not be sought by clinicians–but more broadly as well.

In particular, it seems clear to me that without good evidence to suggest otherwise, Internet gurus who tell patients to forego important medical treatments are violating the ethical principle of beneficence and might in theory be subject to penalties from their pertinent licensing boards. I think this is a very important argument that I have extracted from this.

To re-state, the crucial arguments (to me) are as follows:

  1. Some clinical practices are weakly supported by evidence, such as the use of information on the relationship between BMI and health, yet there is good reason to believe that they promote health;
  2. These should be challenged and tested scientifically to confirm that they do indeed promote health;
    • Carefully testing current, weakly supported practices helps to avoid violating the ethical principle of non-maleficence, by eliminating practices that may be harmful;
  3. Practitioners should not be dissuaded from engaging in these practices without evidence, as doing so might withhold benefit or otherwise useful information from patients;
    • Withholding information that may be beneficial to patients violates the ethical principles of beneficence as well as autonomy.

If you disagree with these arguments, I would love to know your opinion. Posts that are aggressive or do not engage with the arguments will be removed.

Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Producing high-quality scientific work like this takes many hours of work. This is made possible through the support of readers and viewers like you. Please consider donating below.

In just a few clicks, sign up to donate $5 monthly here:
Or $20 monthly here:

For one-time donations, or to become a patron on Patreon, click here.

Advertisements

Reading Time: < 1 minute

Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Producing high-quality scientific work like this takes many hours of work. This is made possible through the support of readers and viewers like you. Please consider donating below.

In just a few clicks, sign up to donate $5 monthly here:
Or $20 monthly here:

For one-time donations, or to become a patron on Patreon, click here.

Reading Time: < 1 minute

Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Producing high-quality scientific work like this takes many hours of work. This is made possible through the support of readers and viewers like you. Please consider donating below.

In just a few clicks, sign up to donate $5 monthly here:
Or $20 monthly here:

For one-time donations, or to become a patron on Patreon, click here.

Reading Time: < 1 minute

Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Producing high-quality scientific work like this takes many hours of work. This is made possible through the support of readers and viewers like you. Please consider donating below.

In just a few clicks, sign up to donate $5 monthly here:
Or $20 monthly here:

For one-time donations, or to become a patron on Patreon, click here.

Reading Time: < 1 minute

Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Producing high-quality scientific work like this takes many hours of work. This is made possible through the support of readers and viewers like you. Please consider donating below.

In just a few clicks, sign up to donate $5 monthly here:
Or $20 monthly here:

For one-time donations, or to become a patron on Patreon, click here.

Reading Time: < 1 minute

Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Producing high-quality scientific work like this takes many hours of work. This is made possible through the support of readers and viewers like you. Please consider donating below.

In just a few clicks, sign up to donate $5 monthly here:
Or $20 monthly here:

For one-time donations, or to become a patron on Patreon, click here.

Reading Time: < 1 minute

Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Producing high-quality scientific work like this takes many hours of work. This is made possible through the support of readers and viewers like you. Please consider donating below.

In just a few clicks, sign up to donate $5 monthly here:
Or $20 monthly here:

For one-time donations, or to become a patron on Patreon, click here.

Reading Time: < 1 minute

The “I have no words” series will consist of tweets in the online health space for which I have no words. This introduction will be the only part of the series for which I have words.

Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Producing high-quality scientific work like this takes many hours of work. This is made possible through the support of readers and viewers like you. Please consider donating below.

In just a few clicks, sign up to donate $5 monthly here:
Or $20 monthly here:

For one-time donations, or to become a patron on Patreon, click here.

Reading Time: 7 minutes

I received the following comment, I assume, from Dr. Garth Davis.

If this is Garth, great. If it is a troll, also OK, as this gives me the opportunity to fully discuss important issues about science communication and poor quality vegan diets.

For context, my disagreement with Garth started over the plant-based diet documentary film The Game Changers.

I have been very disappointed in the reception of The Game Changers by the plant-based community. I have a lot of respect for James Wilks–he is both well-meaning and clearly very intelligent–as well as many of the other people involved in the film.

However, there is very clearly a great deal of exaggeration in the film and of taking license with the facts to sell a story. This story gives the impression of benefits to meatless diets, especially with respect to athletic performance, that are not commensurate with a sober and impartial look at the evidence, and using means that misrepresent the facts. Layne Norton has provided a helpful overview of many of the important factual problems with the film.

However, these factual errors will not be the focus of this article. Instead, I will discuss what the role of factual errors is in scientific communication, as well as what I think the real problem with The Game Changers is: namely, that it poses a real risk of harm to the health of viewers who follow its advice by promoting a poor-quality diet.

Good Enough nutrition science communication

Some might excuse this problem with the film by arguing that switching to a healthy plant-based diet can and does have dramatic, positive health benefits for many or most people, even if the film does take some liberties with some of the facts.

According to this point of view, when talking to lay audiences, a precise representation of the facts is less important than a representation of the facts leading the audience in essentially the right practical direction.

In fact, precisely representing the facts might get in the way of representing the essential practical truth of the facts: being precise with the science might actually prevent one from telling a compelling and essentially true story that might be of benefit to audiences.

(The nuances of a strictly entirely true story, for instance, might be so complicated that they detract from the main point.)

This is a point of view that I struggle with all the time. And I think that there is some important truth to it. Ultimately I reject it as a starting point for science communication, though, for two reasons.

The harms of Good Enough science communication

First, powerful stories that involve misrepresentation of individual facts, even if that misrepresentation of individual facts leads to a “proper” representation of the big picture, eventually take on a life of their own and can be used for other purposes.

For instance, the insight that saturated fat intake is probably a dietary risk factor for cardiovascular disease can be conflated with fat phobia. This fat phobia, in turn, might be endorsed because it is “good enough” and leads to restriction of saturated fat.

This fat phobia, which is probably scientifically unjustified, then can be used by other interests to promote the idea that, so long as one does not eat fat, one will not get heart disease. Perhaps this is also “good enough” because it causes people to avoid saturated fat.

But, finally, this leads to endorsing high-sugar, low-fat products as “heart healthy”. Since they have little fat.

A story that was “good enough” over several permutations of “good enough” becomes counterproductive and harmful. What’s worse, this problem becomes entrenched in institutions of authority and a part of the orthodoxy and very difficult to change.

Exhibit A:

Honey Nut Cheerios Certified as “heart healthy” by American Heart Association

As I showed using a very basic risk calculator, while LDL levels, which are modified by fat and dietary cholesterol intake, are an important risk factor for cardiovascular disease, they are not the only risk factor modified by our food. If they were, then unhealthy plant-based diets, which do not contain dietary cholesterol or much saturated fat, would not increase risk of cardiovascular disease.

Yet, this breakfast cereal of clearly questionable nutritional value (nearly half of the carbohydrate content is from added sugar) is certified by the American Heart Association for the sole reason that it contains little saturated fat and dietary cholesterol.

Yet this is clearly at odds with current science, which shows that of all macronutrients analyzed, replacing saturated fat with carbohydrate and probably especially refined carbohydrate has a very modest impact on lowering LDL compared to other macronutrients. Substantial evidence suggests that refined carbohydrate might not be an improvement on saturated fat for lipids after all, and each might raise risk of coronary heart disease to an equivalent degree compared to other sources of energy.

Refined carbohydrates are associated with coronary heart disease to a similar degree as saturated fats

Yet because of the longstanding and myopic focus on saturated fat and dietary cholesterol–so long as it is low in each, it must be heart-healthy!–this distortion, this health-washing continues with the blessing of the American Heart Association. Dislodging this entrenched misinformation will be very challenging indeed.

This is why we must always insist on the facts and only the facts. Before we know it, our distortions have gotten away from us, and the story we invented is no longer the one that we control.

Good Enough science communication is a slippery slope that most certainly will be exploited by parties whose interests are not science. And there are many such parties. Science is fragile. Money is powerful.

Good Enough nutrition science communication may yield short-term benefits at the expect of long-term loss of trust

This leads to my second reason for rejecting the Good Enough point of view: it will produce short-term gains but for the reason I discussed above (and others), it will erode trust in science.

Not only will intelligent people see through the flimsy justifications for this or that Good Enough-based policy or recommendation or claim, but the unintended consequences will also lead to a loss of trust in science itself. Because while what was communicated was not itself science, it was portrayed as science; thereafter, science will be viewed with suspicion, for no fault of its own.

This will in turn impede scientific authorities’ future efforts in bringing about or guiding policy change that is consistent with science, which will be a great loss for everyone.

Scientists will come to be seen as having “agendas” and as selectively communicating the facts. This distrust will give rise to alternative sources of authority, who will themselves make the fanciful storytelling of the Good Enough storytellers look tame by comparison. Witness: the rise of the ketogenic diet and LCHF.

So I reject Good Enough and insist on strict adherence to reporting the facts. This is a substantial constraint in telling a good story, but not, I believe, an impossible one to overcome. So far, very few people in the nutrition science space have seemed to really try to do both. We seem to have so little faith in the truth.

The main problem with The Game Changers is that its message might harm substantial numbers of people

But even if I thought that the kind of science communication that distorts facts in service of communicating an overall “accurate” practical point of view was justified, would this apply to The Game Changers? I think not.

As I argued in this post, The Game Changers is a film that represents plant-based diets as healthier than omnivorous diets without respect to quality–in fact, mainly promoting poor-quality plant-based diets, presumably to appeal to the broad public.

Why The Game Changers risks harming the health of those who follow its advice

In doing so, the film not only does not use the means-ends reasoning that Good Enough science communication does, it actually risks harming viewers who follow its advice. This is why, from no point of view except, perhaps, a very ruthless kind of ethical or environmental veganism, can the information communicated in The Game Changers be justified.

Unless supplemented by explanations that whole foods are necessary to unlock the health benefits of plant-based diets, and that junk food plant-based diets are harmful to health (which would contradict the message of several key scenes in the film, as I document in my post), by no means can the film be justified or should it be endorsed by health professionals, whose professional obligation is to the health of human beings and not to the well-being of animals or the environment.

One debates opinions but discusses science

That aside, if this really is Garth, I would be glad to discuss this issue and others, but not to debate. The only positions that I have that are “set in stone” are principles. I think that we will broadly agree about both these and about what the science says.

Interpreting science is really just about reporting what the literature says. It is almost mechanical. While it is complicated in the details and while a priori philosophical commitments can affect how one “assigns weight” to different kinds of evidence… my belief is that reasonable people, if they invest the appropriate time and seriousness, will ultimately come to essentially agree on the details, strengths, limitations, etc. of what the current body of literature says. From this point of view, either a person is ideologically motivated and thus not a part of a scientific discussion and not worth seriously talking to, or they are not ideologically motivated and a discussion is possible.

Given the above, if a discussion is what you want, Garth, I am down for that. Let us arrange it.

Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Producing high-quality scientific work like this takes many hours of work. This is made possible through the support of readers and viewers like you. Please consider donating below.

In just a few clicks, sign up to donate $5 monthly here:
Or $20 monthly here:

For one-time donations, or to become a patron on Patreon, click here.

Reading Time: 6 minutes

As a former Paleo dieter (for 10 years!), the ideas that grains are nutritionally deficient, meat nutritionally rich, and carbohydrates inherently harmful were deeply engrained in my habits of thinking. So much so that my first forays into eating whole grain bread after learning about plant-based diets were marked by an anxiety that lasted for years–“am I eating carbohydrates instead of fats for an ideological reason, at the expense of my health? Surely these plant-based writers cannot be right?”

Well, certainly the plant-based writers are not right, as I understand now that they could not impartially analyze their way out of a wet paper bag. Still, those days of carb anxiety are long over, as I now understand carbohydrate and fatty acid metabolism to a sufficient degree as to now be inoculated from the Paleo/keto/low-carb/low-fat macronutrient fear-mongering. (When I say “understand”, I mean that I understand how little we know–and how, when popular nutrition writers inveigh against carbohydrates or fats in the diet, they are, to be generous, reading the scientific literature creatively.)

But what had lingered was the sense that, because grains are nutritionally deficient relative to meat, I needed to be especially careful to avoid the nutritional downsides of wheat bread, of which I ate copious quantities.

See for example the bloggings of the major Paleo writers.

Chris Kresser:

Robb Wolf:

So surely whole grains have a well-justified bad reputation?

It is worth noting that neither writer actually explains how the “nutrient density” algorithms determine the rankings that they report. My guess is that, as with much they write, they did not in fact critically analyze the primary sources that they used to make their recommendations but instead used the ready-made listings that agreed with their biases.

That is my guess, in any case, because when I actually looked at the raw data, I saw a different story, which is significantly more complex. This was purely accidental, just the result of playing with Cronometer. Yet what I found shocked me.

For instance, here is the nutritional profile for 2500 calories of whole wheat bagels:

10 whole wheat bagels, 2500 calories

Now here is 2500 calories of lean ribeye, the favored cut of the online so-called carnivore subculture.

36 ounces of ribeye, 2500 calories

All-ribeye diet: severely deficient in 9 nutrients, moderately deficient in 4. All-bagel diet: severely deficient in 6 nutrients, moderately deficient in 1.

In other words, if one could choose between only wheat or meat, and one was aiming at nutrient adequacy, the answer is clearly wheat.

Of course choosing between eating only meat or only wheat would be insane, but the former is, as I said, just what one online subculture has chosen to do. (I am just reporting the facts.) Not a coincidence, Robb Wolf, mentioned above, helped to promote the growth of this subculture.

What happens when one slightly diversifies the plant foods?

2500 calories, half lentils, half wheat bagels:

Half wheat bagels, half lentils

All-ribeye diet: deficient in 13 nutrients: severely in 9, moderately in 4.
Half-bagel, half lentil diet: deficient in 6 nutrients, severely in 6.

What about reducing wheat and lentils to 43% of total calories each (86% together), putting almonds at 9% of calories (a handful of almonds), and making up the final 5% of calories broccoli and carrots (a handful of each) and sardines (a single sardine, 1.6% of calories)?

I call it the Poverty Omnivore Diet, and it is very nearly my own diet (as an amateur athlete, I add supplemental protein for optimal recovery and muscle mass, and a lot of berries):

A diet of 86% legumes and grains, with very modest quantities of nuts and vegetables

Add sunlight, and you are meeting all of your vitamin, mineral, and protein needs in spades.

For reference:

Again, for comparison, here is an all-ribeye diet:

To provide a further contrast, it is easy to find claims by pseudoscience pushers online that animals are more “nutrient dense” than plants:

That said, a few points are in order.

First, it is sometimes claimed that meat nutrients are more bioavailable than plant nutrients. For a few minerals, this might be true, but I’m not aware of any evidence that this has a meaningful impact on nutrient sufficiency. If readers are aware of such evidence in an otherwise nutrient-rich diet such as the one I have posted, please leave a comment or contact me.

Second, the idea that grain consumption led to worsened health at the beginning of agriculture has led to much speculation that grains per se are suboptimal foods. A paper that reviewed all of the literature on the subject, published in 2011, concluded that stature certainly diminished in most (but not all) human populations during the transition to agriculture. Yet in its discussion, it focused on famine as the likely explanatory factor, not grain intake per se:

Furthermore, if grain consumption and agriculture was the cause of reduced height, then average height would not have increased dramatically over the past few centuries (the Western diet is still heavy in grains):

Finally, if grains replacing meat were the cause of reduced height during agriculture, then vegetarian children wouldn’t be of similar height as omnivorous children. Yet they are.

So why are Paleomyths about meat being more nutrient-rich than plants so widespread, despite the actual story being much more complicated (to say the least)?

My guess is that this is largely motivated by modernity anxiety. For all of human history, humans have idealized and romanticized simpler ways of life, as exemplified by our hunter-gatherer ancestors. We imagine our hunter-gatherer ancestors all ate large quantities of meat. It follows that we should eat meat. We therefore look for sciency-sounding reasons for this belief. Folks like Chris Kresser, Robb Wolf, and Mark Sisson all provide these sciency-sounding reasons. Because this is what we already believed anyway, and we are just looking for reasons to believe it, instead of critically evaluating what these bloggers write, immediately we think “ahhh so that’s why!” We are already primed to believe, not to question. And so we believe. Or at least, once upon a time, I did.

The reality is that the role of plant foods in our ancestors’ diets is substantially more complicated than these writers let on. For instance:

And the rabbit hole goes much deeper. But that is a subject for another post.

Suffice to say, every scientific field is more complicated and richer in controversy than we think at first. Often, we first access a scientific field via a popular writer who tells a good story and appeals to pre-existing beliefs to sell that story. Yet no matter if they are a New York Times bestseller or a famous columnist or respected by large popular audiences on the Internet, we should always be skeptical of the new things we learn, especially if they resonate with and make sense of what we already believe. Unless we are aware of and have critically assessed the relevant body of scientific literature, there is no way whether we are being sold a good yarn or something with strong basis in scientific fact. Until we know better, we need to proceed through life with the assumption that most of what we know is simply a “best guess” based upon “something that we heard that sounded credible”. Because that’s all it really is. My journey through nutrition science, especially with respect to grains, carbohydrates, meat, etc., has taught me that over and over again.

UPDATE

Some people have pointed out that ribeyes should not have carbohydrates. I do not know why the entry at the USDA has ribeyes with carbohydrates. Therefore, I have found a new ribeye entry:

Slightly better than the ribeye I included originally. Still not as good as grains.

Subscribe to Blog via Email

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Producing high-quality scientific work like this takes many hours of work. This is made possible through the support of readers and viewers like you. Please consider donating below.

In just a few clicks, sign up to donate $5 monthly here:
Or $20 monthly here:

For one-time donations, or to become a patron on Patreon, click here.