Noakes, #LCHF and the Professoriate of the Twitterati

LCHF SummitIn the conclusion of my previous post on Prof. Noakes’ Big Issue column, I mentioned that he had spoken of addressing the malignant influence of scientists via the power of crowdsourcing research findings on Twitter.

Today, I’ll conclude with my description of how much that Big Issue column gets wrong, and also provide a fresh example of shoddy thinking from the LCHF crowd, in the form of a debate topic that is being proposed for the upcoming “First International Low Carb High Fat Summit“.

(I don’t think it’s relevant here, but in case you think it is, I’ve been invited – and have accepted – to attend the LCHF summit as a guest of the organiser, Karen Thomson. So if you see me becoming all soft on LCHF in late February, you’ll know that I’ve either been bought off, lost my senses, or “seen the light”.)

Let’s first return to “The Digital Doctor” (the title of the Noakes column). One of the things he’s right about in the column is that increased availability of information via the Internet has undoubtedly given consumers and patients more power.

You’re able to shop around, whether through using Dr. Google to find out what your fellow sufferers have tried, or for crowdsourcing information about who seems to do good work in a certain area, for what price and so forth.

Patients can enter the consulting room armed with some understanding of what ails them, and I’m sure that can help (and also sometimes hinder, no doubt) the process of diagnosis and treatment. We’re no longer victims of as large an information asymmetry as we once were.

But this doesn’t mean that truth becomes something that is resolved via democratic process. The majority can be wrong, and this is especially the case when we’re dealing with a majority drawn from a select population, and where the audience does further filtering of what they think worth listening to and what worth ignoring.

In short, this is again the problem of the filter bubble, writ large.

To quote Noakes:

But the growth of social media and the internet has changed that reality irrevocably.

Today as a result of these very modern developments, patients now have access to the experiences of hundreds of millions of others producing what has been termed The Wisdom of the Crowds.

Exposing millions of people with common issues to a multitude of different interventions soon determines the relative efficacy of treatments more effectively, cheaply and swiftly than any other testing method yet invented. (Indeed, these crowd-based, uncontrolled experiments will add a new model of “scientific” research that will overcome the weakness of traditional laboratory-based research, which by its very nature is “unnatural”.)

We have access to the millions of anecdotes shared by the population that self-selects to be on Twitter, yes. But we have no way of verifying the (literal) truth of those anecdotes, and we also know that because they are uncontrolled experiments, there is no way of verifying their truth in the sense of having confidence that the right cause for the observed effect has been identified at all.

This is precisely why we have and value the scientific method: to rigorously test hypotheses, in a way that minimises errors that might result from selection bias and so forth. We’d should take care to compensate for – and not celebrate – the fact that all of our case reports come from a demographic (because they are on Twitter) that was perhaps on average wealthier, and with more leisure time, than the typical person.

Why? Because that characteristic might be associated with different dietary choices, or different levels of physical activity, both of which would be relevant to how we interpret the data. So the anecdotes can be useful, in pointing to what we should research, but they are not themselves the research we should treat as conclusive.

(Sidebar before moving to the next quote: I’m leaving the “unnatural” towards the end of that quote alone, but just to briefly note that it’s another example of the paranoid and conspiratorial talk Noakes seems partial to. Sure, lab-based research has flaws, but its lack of being “natural” – whatever that might mean – isn’t one of them.

One can say we can’t replicate some real experiences and effects in a lab, and there it becomes a potential problem, but describing this in terms of “natural” sets up an advantage for the Romantic Paleo argument, and poisons the well against anything positioned as non-natural, which might include GMOs, vaccines, etc.)

Then, we’d want to compensate for our own biases also. Later on in the column, Noakes says:

I soon learnt that Twitter is unquestionably the best way to acquire the most up-to-date information on my particular areas of scientific interest. By following a group of scientists who use Twitter to disseminate information they find interesting, I now have access to new knowledge within minutes of its first appearance in the scientific literature. The result is that acquiring new information is absolutely effortless, and dependent only on my choice of whom I follow on Twitter.

So, if I listen to the people who tell me things I like to hear, I get to hear a lot of things I like to hear. And, just to make certain that things I don’t want to hear don’t intrude on this, I’ll block people who post contrary research from my Twitter feed (he doesn’t say the latter above, but it is something he does, with me being one example of someone whose links to relevant research don’t get through).

The problem with the Big Issue article is that it takes a couple of sound points, and explodes them into such a grand narrative that they lose any sense they had.

Noakes tells us that “now it is only the advice that works that has long-term credibility”, and that is true, up to a point. You’ll get caught out on social media if you spread misinformation. But that doesn’t mean information gains credibility through being widely disseminated (on channels you’ve hand-picked) on social media. Those are separate issues.

It’s also true that professionals – in various spheres – were able to exploit the ignorance of the consumer to peddle quackery or defective goods, and that they are now less able to do so than in the past. But our easier access to information doesn’t mean that (proper) experts don’t often know better than we do. Those are also separate issues.

There’s a tendency in the LCHF narrative, at least how it’s playing out in South Africa, to continually hyperbolise the consequences of choice, and the lack of middle-ground options between one or another extreme.

To conclude, here’s a recent example of that elimination of the middle-ground (in logical terms, a false dichotomy), that comes from the Facebook page of the LCHF summit I spoke of at the top of this post. On that page, Karen Thomson says:

I am desperately trying to find medical professionals willing to debate: ‘Is the low carbohydrate diet the cause of, or the cure for the global epidemic of chronic ill-health?’ with our LCHF team.

Any thoughts? None of my invitations have been met with any success.

Here’s a thought: if a debate topic is framed so that only the pro-LCHF side have any chance of winning it, you’re unlikely to garner any interest from potential opposing speakers. The topic is itself a pithy example of the shoddy reasoning that I’ve written about so often here, in that:

  • The topic sets up a false dichotomy, because the truth might lie somewhere in the middle with LCHF being neither the cure nor the cause.
  • Second, the topic sets the non-LCHF people up for failure, in that it would be impossible to prove that LCHF is the cause of “the global epidemic of ill health”. This is true even on the simple grounds of chronology, where (leaving aside the contentious issue of what pre-social humans ate) modern humans haven’t been eating LCHF for long enough – or in large enough numbers – for LCHF to be identified as the cause of much at all except book sales, never mind a “global epidemic”. You might therefore think the topic indicates bad faith, but even if you don’t want to be uncharitable, the pro-LCHF folks have rigged the game here, by ensuring that they can amass at least some evidence for their position, while the opposition cannot.
  • Third, one can challenge the presumption of the topic that there is a “global epidemic” at all, in that despite diabetes, heart disease and the like, we somehow keep living for longer. So, simply accepting the premise of an epidemic accedes to one of the key claims made by the LCHF folk, namely that humans in the 21st Century are (in general) sick and dying, despite any appearances to the contrary.


3 Pingbacks/Trackbacks

  • Leon

    All twitter does is to describe in 144 characters that hey, I found this or that! There is no way to evaluate statements like these without having access to the experimental protocol or the data obtained nor the statistical manipulations as can be done when reading a paper in an academic journal. IOW: anecdotes. The plural of anecdote is not data.

    • City Bowl Nomad

      I’m sure they post links in Twitter to their research

  • Francois du Toit

    I think one of the points you are missing is that the LCHF people are not trying to say we are dying more per say, but lets be honest and have a look at the average weight of ppl these days compared to before high carb intake started. You have to agree that there is more obese and overweight ppl these days than ever before! Surely this must be questioned? And surley the diet is mostly to blame for this? If not the diet, what then?

    • Sure, that must be questioned. But doing so – and questions about diet in general – have almost nothing to do with the topic of this post.

  • Tom

    Why don’t you suggest a different title to the debate such as “Evidence Based Medicine informs us that LCHF is the optimal diet for healthy living” or something like that. You could take them to the cleaners on that one.

  • Pingback: #NoakesHearing, day 1: Reflections on futility()

  • Pingback: Populism, more than prejudice, is the problem with Trump – Synapses()

  • Pingback: Populism, more than prejudice, is the problem with Trump – Africa Blogging()