The Raw Food Diet, Correlation and Causation.

On TV this morning there was a report on the ‘raw food revolution.’ It featured a man who’d lost a large amount of weight and ‘cured’ himself of cancer. That sounds pretty amazing and is definitely a good promotion for the raw food diet. But, there’s a big BUT… Hang with me until the end of the article…

I loved the statistics subject at university (there’s every possibility I am the ONLY student to have ever said that). It taught me how to carefully analyse, scrutinise and understand research studies. It taught me that correlation doesn’t equal causation. I’m lucky I had the opportunity at university to learn these skills.

For much of the public though, reading research studies can, understandably, be quite confusing. It’s through the media that most people are exposed to results of research. But day to day we often carry out our own little studies. It might be something as simple as trying different laundry washing powders, changing pillows until we find one that’s comfortable, flicking through radio channels until we find one we like the most. We experiment, we decide what works and what doesn’t, and based on this we make a decision.

The results of our own little studies or that of high-grade research will either show no relationship or evidence of some sort of relationship between the variables. Upon concluding a relationship does indeed exist, we consider this to be a correlation. If we can prove that a change in one variable is solely and directly due to a change in the other variable, without any other contributing factors, then we can declare causation.


For example, I stayed at a hotel in Sydney a few weeks ago. The bed wasn’t very comfortable (not compared to my bed at home) and I had an awful sleep. Based on that, I made the assumption that my bad sleep was due to the bed being uncomfortable. Therefore it’s all the bed’s fault. Right? Not quite. Sure, there might have been an interaction (correlation) between the uncomfortable bed and my quality of sleep; but my sleep might have also been affected by the light in the room, the noise of the traffic, what I’d eaten for dinner that night, the stress of travelling, the room temperature etc etc. Proving that it was indeed the bed’s fault (causation) becomes a little problematic when there are all these confounding variables. Unfortunately, we often misconstrue correlation to equal causation.

Let’s go back to the apparent magic of the raw food diet. Whilst I would agree that there is a correlation between the man’s results and the raw food diet he undertook; this most definitely does not prove causation. Think about the other interactions: what was he eating before? How much activity was he doing? He could potentially have achieved those results because he increased his overall fruit and vegetable consumption or because he began exercising. Maybe he would have achieved those same results by reducing his saturated fat intake and eating more (cooked!) protein and whole grains. Maybe he began eating more mindfully and was therefore honouring his hunger and fullness signals. Did he ‘cure’ his cancer through eating raw foods, or was it because he lost weight? Maybe it was due to the medical treatment he received? Or maybe something else happened in his life to reduce his stress levels. Maybe he moved to Colorado. Maybe it wasn’t raw foods at all but the inclusion of chickpeas in his diet. MAYBE CHICKPEAS ARE THE CURE FOR EVERYTHING?! Uh, maybe not.

7EHhlkOA similar relationship exists between obesity/overweight and the increased risk of disease. Yes, we acknowledge there is a correlation between extreme levels of body size (at both ends of the scale) and risk of disease/reduced life span. Yet, we don’t have the evidence to support this as causation. If it was indeed causation, then ALL people who don’t fit into a set weight range would be experiencing those health issues; we know this isn’t the case. There’s the potential that it’s not weight itself that causes the development of disease, rather a multitude of other factors should be considered. Weight stigma? Exercise levels? Stress? Dietary intake? Environment? Genetics? These factors also show correlation with disease, hence why those who are a ‘healthy’ weight are still at risk of so-called ‘obesity related’ diseases. Further, in some instances being overweight can actually be protective and increase life span (check out the obesity paradox here and here for more).

Now before anyone gets their knickers in a knot, I’m not definitively saying that the raw food diet isn’t solely responsible for the man’s weight loss and cancer cure, maybe it is. But unless we know for sure, it would be incredibly incorrect and irresponsible to make the assumption.

It’s great to research and to experiment with different things. Have a play and find what does and doesn’t work for you. We are all individuals and will experience things differently. But remember to not mistake correlation for causation. When you see media reports or hear friends reporting on how great the latest fad diet, don’t be afraid to scrutinise it. Consider the other possibilities. Think about not only the physical effects, but also the emotional, mental and social. Take what works for you, and leave the rest behind.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s