Be skeptical of research on diet and nutrition

First of all, my thanks to all those who have replied to my first ever blog post either on this website or via Facebook. When starting a blog, you’re never quite sure if anyone is going to be remotely interested in what you have to say, so it’s great to have some encouragement.

I wanted to blog today about some of the science and research that underpins lots of our views on nutrition and health. I’ve listed the books and authors I’ve developed my own thinking from in the “Useful books and resources” section – John Briffa and Mark Sisson are particularly worth reading if you want more detail. If you only buy one book, buy “Escape The Diet Trap” by John Briffa – as a result of my own weight loss, loads of my friends have bought it (or at least his earlier book, “Waist Disposal”) and have also been able to shift the pounds too.

The scariest thing I’ve come across during this journey has been that we know less about the impact of what we put in our mouths has on our bodies than we do about the origins of the Universe. Yes, this is mentioned in the QI Book of the Dead, so it must be true! Seriously, that’s a scary thought given how much “guidance” we are given by the media, the government, health authorities etc about what we should and shouldn’t eat. My advice to you reading this is take every supposed bit of research on nutrition and health with a pinch of salt and a lot of scepticism, and there are a number of reasons for this.

Firstly, the main problem here is unlike other scientific experiments, where you have a control group and you have a test group where we identify a set of variables and amend one at a time in a controlled environment, the vast majority of research into diet involves people keeping a food diary of what they eat. And people tend to not be completely truthful (either consciously or not) – we tend to underplay how much we’ve eaten. So it’s hard to completely rely upon the actual base data of evidence if you like – it’s hardly a controlled environment.

Secondly, it’s very difficult to actually work on changing one variable at a time over a suitable long time period for it to have any real effect, simply because there are so many variables to consider. But my biggest issue with most research into this space is what I consider to a really basic flaw in the conclusions. If you’ve ever studied quantitative analysis, you’ll know when trying to understand the relationship between two variables you need to understand the difference between correlation and causation – it’s basic scientific practice to know this, yet so much of the diet research I’ve read seems to get this wrong. If by increasing the amount of variable (a) in our diet, we see that variable (b) also increases, it doesn’t necessarily mean that (a) causes (b). Just because (b) goes up when (a) goes up, it doesn’t mean (a) causes (b) to go up.  It could that (a) has a relationship with (b), but that it is the presence of an unidentified third variable (c) which drives both up. This can lead to wrong conclusions. For example, if I go on holiday to Turkey, and I eat ice cream every day, I become tanned. Just because my eating ice cream is accompanied by my getting a tan, doesn’t mean the ice cream eating causes the tan. It’s the fact that I’m on holiday in a sunny country that causes me both to eat ice cream and get a tan. So there is a correlation between ice cream eating and getting a tan, but one certainly doesn’t cause the other. Nutritional research is littered with this sort of stuff. Dr Briffa has countless examples of this on his website (www.drbriffa.com).

Anyway, the point of all this was to get you to move away from thinking about what the science of nutrition says, and to use as my father in law likes to say – YOUR LOFTER! Next post will be about how actually if you apply a bit of logic and think about evolution, actually what we should be eating to stay healthy is just common sense.

Leave a comment