Sunday, 2 August 2009

The Pseudo-Scientific Nature of Health Blogging

I read a lot of blogs both in favour of and against a high protein, low carbohydrate lifestyle. I would like to think that I am fairly objective and open minded and that I strive to obtain all the facts from a variety of viewpoints and biases before forming my own opinion. Having written and published several of my own research papers (on memory and cognition), I am acutely aware of the need to be fully reflexive in any writing and acknowledge my own biases, preconceptions and those of any researcher or study I am reading to ensure that fact and opinion are kept as separate as possible.

Or maybe in this context I just like to see what the other 'side' is up to. Either way, as well as subscribing to many primal and low-carbohydrate blogs, I also have some more, erm, odd choices on my RSS feed. Take Disease Proof, for example, written by some dude on behalf of Dr Fuhrman, writer of the Eat to Live diet book and proponent of a largely vegan high-carb, zero/low animal protein lifestyle.

Firstly, let me state that in the grand scheme of 'diets' available, Eat to Live is probably one of the least offensive out there. It aims for natural foods, few grain products and lots of vegetables. I have no problem with that. What I have a problem with is its inherent biases against high-protein meat and animal product based diet to the extent that many of its blog posts are full of half-truths or even downright falsehoods. As some guy is attempting to sell a few books of the back of this blog, I think it's abhorrent to misrepresent information in such a way for pecuniary gain.

The problem with this blog is it is a prime example of how not to report mainstream research findings, time and again just regurgitating sensationalist headlines to support its lifestyle prescriptions, without fully considering the actual conclusions and applicability of the findings. Usually studies have sampling problems, data collection issues or correlational or statistical assumptions which mean that at best some vague relationship may be inferred and, at worst, the findings are more or less meaningless to a generalised population of people. Or else the data is massaged (ala Ancel Keyes) to enable a statistically significant effect to be observed.

It's quite plausible that perhaps the blogger doesn't know how to critique a scientific study? Who knows. Because many of the studies which are linked to as 'proof' that the Eat to Live dogma is 'true' have been criticised and pulled apart by those with a shred of understanding of sampling errors, correlational analyses and, well, basic common sense. Take the China Study, for example (as Disease Proof so often does): does it 'prove' the vegan hypothesis, or is it just a misrepresentation of the facts?

Which brings me to the second major issue with this guy. In the world of Disease Proof, questioning the gospel of Dr Fuhrman's Eat to Live is unthinkable. And if you present a rational scientifically-based argument against the Eat to Live principles, well, be prepared to be subjected to some childish insults (read the comments, they're golden!). Yeah, name calling, that ought to show them who is right!

Dude, learn how to systematically deconstruct quantitative research because it's sort of embarrassing to read!

So to summarise: if a blog/blogger misrepresents research, fails to reflexively acknowledge his/her position and biases, is inflexible in the face of conflicting evidence, resorts to abuse and insults if challenged, well, I would take their 'opinions' with a pinch of salt and, most importantly, stay far away from any endorsed products!

No comments:

Post a Comment