The [Grow] Network is pleased to publish Colin Austin’s 10 part series, How to Grow (or Buy) Healthy Food. This article is Chapter 2 of 10. You can read the other chapters here:
• Chapter 1 – Diet and Health, a Personal Experience
• Chapter 2 – Statistics and the Diet Controversy
• Chapter 3 – Eat Right, Not Less
• Chapter 4 – Finding a Diet by Self-Experimentation
• Chapter 5 – Essential Nutrients for Good Health
• Chapter 6 – The What and Where of Minerals
• Chapter 7 – The Rhizosphere
• Chapter 8 – Transferring Nutrients and Biology to Growing Beds
• Chapter 9 – From Garden to Kitchen
• Chapter 10 – Community Action
Chapter 2 – Statistics and the Diet Controversy
To recap the story so far, in the first episode we focused on Why you should seek out healthy food. I made the point that you have a choice. You can either be fit and healthy, and feel younger for the rest of your life; or you can risk going blind, losing limbs to diabetes, ending up in a wheel chair from a stroke, or at least – the mercifully short option – dropping dead from a heart attack.
I talked about some of the experiences Xiulan and I had in facing her diabetes, which was brought on by a poor diet. And I talked about how we received dietary advice which aggravated, rather than helped, her to control her blood sugars.
What’s to Come in Chapter 2
Now I know it sounds dead boring, but I want to talk about the science of statistics. Don’t turn your brain off just yet! The reality is that our knowledge of diet and health is largely based on the statistical analysis of large trials involving thousands of people. If all studies led to the same conclusion then we would not have to worry about the statistical analysis. We could simply say, “That is what the experts said,” and we could get on with growing healthy food. After all, that is what this series is really all about.
However, all studies do not come to the same conclusion. The world of dietary information is full of strongly opposed viewpoints. The world of dietary information also overlooks a dramatically important conclusion – that the minerals and nutrients that are present in the soil in which our plants grow have a major impact on our hormones, which control our bodies. The impact of soil minerals and nutrients can either make us healthy, or make us fat and prone to disease.
Before I start with the statistics, I want to clarify some points. Because of our health issues, Xiulan and I have put a great deal of effort into researching diet. Xiulan is a doctor, and my expertise is in science and innovation, so I feel like we have good credentials for acquiring and sharing significant scientific knowledge about diet.
However, our prime motivation is to develop a diet for our own health in our own situation. We are very happy to share our findings, and if they are helpful to you that is great! But we are not offering dietary advice. We have also used our own targets to define the goals of a healthy diet; we accept the reality that our DNA can only reproduce a finite number of times, so our life span has an upper limit. Our focus has been on staying fit, active, and healthy while we are alive; rather than aiming to live forever.
Food is not just about health. Food is also about pleasure, and sharing meals with friends is a source of pleasure for us. We want to enjoy our food.
We like to travel, and often it is just impossible to get the sort of food we would prefer to eat. At times we accept the reality of having to eat food which is not as healthy as we would like it to be. These decisions are personal. We are not trying to tell other people how they should eat.
Why do we eat?
We first asked ourselves what may sound like a silly question, “why do we eat?”
We eat for two very different reasons: 1) to get energy, and 2) to regenerate our bodies. A third but non-essential reason is that we eat simply because we like to eat. Again, food should be a pleasurable and social experience.
Modern food has an excess of energy, but it is lacking in key nutrients that are important to regenerating our bodies. Hence the saying, “overfed and undernourished.” For Xiulan and I, the most important factor was to regenerate our bodies.
Obviously we have to watch our energy levels, but having a diet which regenerates our body parts so we can remain fit and active is crucial for us. But it is not so easy. There are so many options and so much contradictory information. As an example, consider all of the low and high fat diets that are available today – Atkins, Biggest Loser, DASH, gluten-free, Jenny Craig, Mayo Clinic, Mediterranean, Nutrisystem, paleo, South Beach, TLC, Weight Watchers, and so on.
Where should you even start?
If you ask Mr. Google (or Youtube) using key words such as diet and health, diet and diabetes, metabolic syndrome, diet and hormones, and hunger, you will find dedicated and competent doctors arguing their own viewpoint. Drs. Michael Gregor, Tel Oren, Caldwell Esslstyne, Robert Lustig, Hoanna McMiller, Colin Campbell, Jay Wortman, Douglas Lisle, Sarah Hallberg, Joel Fuhrman, Gary Taubes, and Nina Teicholz are just a few. But their recommendations are in dramatic conflict, particularly on fat, meat, eggs, and dairy.
By chance it happened that in my search I watched a YouTube video by Dr. Jay Wortman followed immediately by Dr. Dean Ornish. When I watched Dr. Wortman’s presentation it was obvious that he had a really good understanding of the issues, and he presented much valuable information based on sound experimental trials. I was totally convinced that this guy had the formula. We needed to incorporate more fat into our diet.
Next, I watched Dr. Dean Ornish. Again, he was clearly on top of his job. He presented the results of many well conducted trials, but he came to the absolute opposite conclusion. We needed to eliminate fat from our diet.
How can two highly competent doctors armed with data from well conducted trials, who agreed on virtually all of the basic principles of how the body works, come up with diametrically opposed views about consuming fat? This is what my writing is all about – sorting out the meaning from the available mass of information and contradictions. Sorting out the meaning. As I studied the information, it became clear that we have been using models which are unrealistically simple. One oversimplification is thinking that people are all the same and one diet will fit everyone.
I believe that both doctors are correct.
If a patient’s arteries are clogging up and death is imminent, then an extremely fat free diet may keep him alive. But this may only be necessary for a small proportion of the population.
The bulk of the population is suffering from health problems stemming from being overweight. It is common to have a diet which is too rapidly digested, giving sugar spikes that cause the body to overreact by pumping insulin into the blood. This in turn leads to excessive drops in sugar and a craving to eat again. Fat slows these cycles and may benefit many people.
The mechanical use of statistics, without human judgement, can result in the oversimplification of complex issues.
Another oversimplification is the misuse of thermodynamics to say that we become overweight because we eat too many calories. The body has a complex control system which can send fat either to tums and bums, or to poop and pee. The solution is to manage our body’s control system by managing what we eat, rather than simply restricting calorie intake. Eat right, not less.
Many years ago I studied both statistics and thermodynamics at University. I found them to be difficult, and far less entertaining than a James Bond movie – and not much has changed since that time. However, to sort out the meaning of the dietary statistics I was reading, I needed to build my appreciation of these disciplines. And there is no way I can avoid talking about them here. I have tried to make the statistics more interesting than my old lecturers did, by telling horror stories and jokes. But my key message is simple. Eat right, not less.
Take 1 doctor, add 1 engineer. Stir well, and let them fight it out.
How can I make sense of this, and what qualification do I have to conduct this analysis? I am not even a doctor, I am an engineer. But Xiulan is a doctor and she takes her wifely responsibility seriously – always letting her husband know when he has it wrong. She takes the message from the old Tammy Wynette song “Stand By Your Man” to heart, and she won’t give up until I stop getting it wrong.
So, how can an engineer make any contribution to the debate on diet? I am always surprised by the positive image that engineers seem to have. True, we do build airplanes, computers, and dishwashers that seem to work. But while we may appear competent to the outside world, the truth is mostly that we have no idea what we are doing. We somehow manage to disguise our incompetence to the rest of the world, and we struggle through. If we want to keep our jobs and keep our company solvent, then we have to make products that actually work – on time and on budget. There is too much to know and too little time to learn it, so the greatest skill of an engineer is managing ignorance.
We often say we are in the applied science business – we study science, then we go out and apply the findings. A few lucky scientists will come up with some world shattering breakthrough, but generally they make small advances and they leave some issues still unresolved. Scientists can quite happily admit that further research is needed, and then go off and apply for another grant.
Engineers have to somehow deliver a product, so we have to take what information is available from our scientist colleagues at the moment, and apply it as best we can. Scientists can study in their particular area of specialization, but we engineers have to study many disciplines. If we are designing a car, we have to know how engines, gearboxes, and transmissions work. We need to learn about the dynamics of road holding, and about the forming of metals and plastics.
It is simply impossible to master everything that we really should know – so how do we struggle through? It is by the process of extracting a simple mechanism, a principle, or a general rule from the vast amount of data. Scientists conduct tests and measure things, collecting vast amounts of information. With improvements in instrumentation and recording technology, this data is generally correct. The weak link is the step of going from good data obtained by reductionist science to a general principle which can be widely applied.
This seems to be at the heart of the problem with regards to dietary understanding. There is good data but it fails to lead to sound general principles. The problem is in going from the specific to the general, in trying to unravel the data.
I love the internet. It opens up global information to me, while I sit on my veranda and and watch kangaroos. This is where I get most of my information on diet, plus of course my friends in Bundaberg library. But there are two great traps.
In the internet world, people often use the term “pancake knowledge.” There is so much information that it is impossible to know anything in depth. So we end up knowing a little about a great many things. But to understand the basic principles you need depth of thinking.
The second major problem is Google bias. I like red wine and chocolate. I could Google ‘health benefits of red wine and chocolate’ and I would get a whole bunch of scholarly articles praising the benefits of red wine and chocolate. I could feel very smug and go gorge myself. But if I want to be honest, I need to Google ‘health hazards of red wine and chocolate’ as well so that I can see the counter arguments (which I may just decide to deliberately ignore – I am human too). In this way, the internet tends to reinforce existing prejudices and paradigms.
Which is Bigger – 5 or 0.15?
Smokers are 5 times more like to get lung cancer than non-smokers. This is a result of statistical studies, and 5 times is so big that no one is going to question it. Simply stop smoking – clear as a bell!
But look at the data on diet and health. We don’t see big numbers like 5, but instead we see very small numbers like 0.15. When we see numbers like this, we tend to get suspicious and question whether the drawn conclusion is really true, or whether there is something else going on.
I have done this, and it has led me to some dramatic changes in the way I think about diet and health. But to do this, I have to talk a bit about statistics, which I know most people find just plain boring. But I actually want you to read this stuff, so I am going to try and lighten it up with a horror story and a joke.
A Horror Story
Imagine that there is this deep gorge you want to cross. Hundreds of meters below, in the bottom of the gorge, are white water rapids flowing over jagged rocks. Swimming in the water are flesh eating piranhas and crocodiles. This gorge is also home to the world’s largest snake, which can swallow people whole. There are also the giant carnivorous birds which peck your eyes out first before attacking your stomach. And, of course, there are killer hornets that can sting you to death in large swarms. Even if you manage to escape all of these, there are mites that live in the slime covering the rocks. These mites penetrate your skin, and get into your blood stream to eat your organs from the inside, leaving only your skin behind.
The only good news is that you are almost certainly going to be killed by the fall onto the rocks, so you won’t have to suffer as your body is devoured. (Why do I bother to write about food? It is much more fun to write horror stories! Move over Sigourney Weaver and Aliens.)
Then you notice me standing nearby. You say to me, “Colin, you are an engineer. Can you build us a bridge over the gorge?”
Being a decent sort of fellow, I build a bridge out of the local vines and say, “Go for it.”
You look at the swinging vines and ask the obvious question, “Is this bridge absolutely safe?”
Here, I start to lecture you about probability. In this world, there is nothing that is absolutely safe. Everything is a question of probability. The chance of the bridge collapsing could be one in one hundred million, or it could be one in ten. So, naturally, you want to know what the probability of failure is and you ask me for my assessment.
“Well,” I say, “there is a 15% probability that you will reach the other side. The bridge is 20 meters wide, so after taking three steps the odds are that the bridge will collapse.”
You may not be too impressed by this so, being a nice guy, I rework the bridge with some really strong steel beams. I explain that this is still not absolutely safe – all I have done is to improve the probabilities.
So, why tell this silly story? Just to make the point that there is a lot we do not now for sure about diet and health. And the correlations are really very low. Compare this with the data for smokers who are 5 times more likely to get lung cancer. This is a serious correlation which we can virtually take as a fact. In comparison our knowledge of diet and health is much shakier. There is simply a lot that we do no know. But, we still need to make decisions.
Not Safety Factors… Ignorance Factors
I was trained as an engineer and the job of an engineer is to make something work in a state of ignorance. The first lecture at university is about safety factors. In reality these are not safety factors at all but ignorance factors – how to make sure things don’t fall to pieces even when you don’t have all the facts.
The ABCs of Statistics
Engineers don’t really like statistics, but we are such an ignorant bunch that we have to use them. Give us a nice simple formula like stress = load/area and we are like kids at the beach. But give us statistics and we become grim faced and suspicious. “Give us a mechanism,” we cry in unison.
Statistics uses the term ‘significant,’ which like many technical terms has a specific meaning which is different than the term’s general usage. Nowadays all statistics are managed by software. The software executes a highly sophisticated but mechanical process to look at all the data and calculate out the natural variability, and if some factors seem a bit unusual, or unlikely to happen by chance, it is classified by the software as ‘significant’ to impress us – it even gives us a number to measure just how significant it is.
This is very different to what we – as thinking humans – may say is significant. I will soon show the importance of this.
Let’s take a couple of examples of correlation between A and B. The security of results is measured by sample size and the length of time the results have been measured. The correlation between obesity and mobile phone use is very high, far higher than most of the publicized correlation data for diet and health. We can collect data for hundreds of millions of people going back over thirty years – good data.
Our first thought is to find the cause and effect.
One possibility is that people get fat because they use mobile phones. This seems feasible as time spent on mobile phones involves little activity, so it is quite reasonable that people who use mobile phones would get fat. Another possibility is that people who are fat find it difficult to be active and so they use their mobile phones a lot. Which one is right? Well perhaps neither A nor B is correct, but factor C is the cause – and factor C was never fed into the computer.
Over the last thirty years there has been a dramatic change in technology, making mobile phones much more sophisticated. There has been a similar dramatic change in the food processing industry, with the proliferation of products like high fructose corn syrup which now dominates our food supply. But if no one tells the computer about this, it will not be analyzed.
My favorite A, B, C scenario is the fact that bright red cars get more speeding tickets than dark blue cars. This is true. Now believe me as an engineer – painting a car red does not make it go faster. In this case factor C is that bright red cars tend to be driven by adolescent males whose brain is saturated by sex hormones, so they drive bright red cars to attract the young ladies, and they also happen to drive faster.
These examples are just a bit of fun, but let us look at one that is for real. One of the most important studies on diet and health was the Chinese study which came to the uncontested conclusion that a diet high in vegetables leads to better health. Now this was a well conducted proper scientific study and I am not suggesting that the results are in doubt. But there is one interesting anomaly that the computer analysis completely missed.
It has a profound effect on how we think about diet, a theme I will pick up in the next chapter.
Coming Up in Chapter 3
The next chapter will lead to a dramatic conclusion – the minerals and nutrients in the soil in which our plants grow has a major impact on the hormones which control our bodies – making us fat and prone to disease, or alternatively making us fit and healthy.
• Chapter 1 – Diet and Health, a Personal Experience
• Chapter 2 – Statistics and the Diet Controversy
• Chapter 3 – Eat Right, Not Less
• Chapter 4 – Finding a Diet by Self-Experimentation
• Chapter 5 – Essential Nutrients for Good Health
• Chapter 6 – The What and Where of Minerals
• Chapter 7 – The Rhizosphere
• Chapter 8 – Transferring Nutrients and Biology to Growing Beds
• Chapter 9 – From Garden to Kitchen
• Chapter 10 – Community Action
© 28 July 2015 Colin Austin – Creative Commons – This document may be reproduced but the source should be acknowledged. Information may be used for private use but commercial use requires a license.
COMMENTS(0)
I love the way you write. You make it a very interesting story on top of it! Thank you for doing this!