On Caterpillars, Bias and the Null Hypothesis


Last week I picked two dead caterpillars out of a floret of organic broccoli that my daughter had refused to eat. I had earlier picked a small slug off the head before washing, chopping and briefly boiling it. I had also spent some time carefully washing aphids out from between the florets of the previous week’s vegbox offering.

The caterpillars gave me pause for thought. Apparently the diets of primates such as gorillas, that appear to only eat plants, do in fact consist of about 1% insects. They do not spray their food with insecticide, or wash or boil it. They simply eat it, bugs and all. This is where they are thought to get their vitamin B12, and other important nutrients may also be supplemented in this way.

One of the speakers at last weekend’s VegFest was Neil Robinson, an ex-footballer from Liverpool and a longtime vegan campaigner. I checked out his website, where he has written his thoughts about the use of the term plant-based diet. Neil is concerned that a number of authors using this term include small amounts of animal products in their recipes, and suggests that vegans, who reject all animal products on ethical grounds, adopt the term plant-exclusive instead.

Plant-based and vegan are terms that are increasingly used interchangeably, but Neil is right that there is a difference between them and I agree, from a slightly different perspective, that it is an important one. In the discussion of nutrition I prefer to use the term plant-based, because while there is now plenty of evidence that eating fruit and vegetables is good for our health, and nobody is really disputing that we should all be eating more of them, I am not aware of any scientific evidence that indicates that a diet consisting of 100% plants is preferable to one that consists of 99% plants and 1% caterpillars; or for that matter, 95% plants and an occasional egg or piece of oily fish.

In fact, it is perfectly reasonable to think that as primates ourselves, not ruminants with four stomachs that can produce vitamin B12 internally, we do need a small amount of animal matter in our diets. This is why those choosing to wash their vegetables and otherwise exclude all animal products, for whatever reason, do need to take supplements.

I consider myself to be a vegan, because I choose to exclude all animal products from my diet, and as far as possible I also try to avoid wearing leather, wool and silk (well, I was never really into wearing silk). I consider this to be a compassionate lifestyle choice, and I value the fact that eating a vegan diet is also good for the environment in terms of reducing my carbon footprint. But if I’m honest, it was health considerations that finally caused me to make the change, and I don’t think I’m alone in this.

As a healthcare professional, I’m interested in how much influence our dietary choices have on our health, and what the best diet for human health might be. And as a scientist, I’m acutely aware of the potential here for something called bias.

Bias is what happens when someone who wants to prove something they believe, designs and carries out a scientific study to demonstrate this fact to the world. It is what happens when, if they fail to show what they hoped, they do not then publish the results. Perhaps the most important source of bias is what happens when someone does not fund, design or carry out said study in the first place, because they believe it will show something they do not like or do not want the world to know. Bias is generally subconscious (unlike editing your data to make it show what you want it to, which is sadly more common than we might like to think), and may be built into the design of the study, for example in the wording of questionnaires: leading questions that invite a particular answer.

Bias is everywhere in science and nowhere is its influence more problematic than in the science of nutrition. And this leads me to a very important point. I have worried about my own bias, and my own ability to be objective… but EVERYONE is biased as regards nutrition. If you eat meat, and like it, then you will be biased towards the conclusion that a meat-based diet is the best. If you are vegetarian or vegan then you will likely be biased in the opposite direction.

The most important way to avoid potential bias, and remain objective, is to acknowledge that bias, and to work with integrity according to proper scientific method. Proper scientific method begins with something called the null hypothesis. The null hypothesis is the hypothesis of no difference. In the example of diet, it states that there is no difference in health outcomes between meat-based or plant-based diets. Or, put another way, diet has no effect on health. The onus is then on the scientists to prove or disprove (in one or other direction) the null hypothesis.

There is a general tendency when people are discussing nutrition to consider the western wheat, meat and dairy-based diet as the gold standard against which all other options must prove themselves. This is a cultural and economic bias, which exerts itself most strongly in current government health recommendations. Telling people to more or less continue to eat what they have always eaten, and devising recommendations that more or less reflect what people are currently eating (perhaps with a little more emphasis on fruits and vegetables), has no basis in good scientific practice. If anything, we should begin by acknowledging how little we really know, and opening our minds to many possibilities.

So for me, the term vegan is an individual identity badge. It describes an ethical and lifestyle choice, and it’s quite useful when it appears on food labels or when talking to waiters in restaurants. Plant-based nutrition on the other hand is about considering that something other than the Standard UK Diet (SUKD) may offer real hope for health improvement in the population as a whole, and that is something that warrants considerable further scientific investigation.

Leave a comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.