Association and anecdotal evidence.
Blog time : it’s late August 2020 , peak summer holiday season so of course it’s raining steadily here which must be nice for all the
muppets, sorry ‘holidaymakers’ that totally blocked all of the roads around here with their camper vans and caravans last week. Personally i’m just waiting for the endless holiday that has been the lockdown- so called, all comes to an end, the kiddywinkles all go back to school….once the teachers actually decide to ‘grow a pair’ and just do their job : just imagine if we in the NHS had acted in the same cowardly and self protective way that the teachers are doing now….sorry we can’t have patients because they might be infected etc.
When the holidaymakers and kiddywinkes have all gone i’m going for a good long walk and hopefully with a new tent, assuming that is that ‘wild’ camping hasn’t become a capital offence. In the meantime i continue in my deep dive into the world of the science and bad science of diet and nutrition, this time with a first look at some of the strange claims around (non) Vitamin D.
Here’s a little story though before i begin.
Many years ago and once upon a time i was a kind-of senior nurse with a few odd additional skills ; i was for example allowed to say (and write in the notes) ‘this one is dead’ which i seemed to have to do a lot on night shifts around the medical wards. Anyway, the original idea of the actual job was to try and coordinate the mainly junior doctors at night and to help them out with technical support : a quick cannula here, an arterial blood gas or ECG there and a slightly slower and more thoughtful ‘this one is dead’ followed by the essential question “are they allowed to be”.
Long introduction shortened a little and one of those junior doctors was a bloke you may not have heard of….Dr Ben Goldacre : well, young Ben even then was a writer and if my memory serves then he had his first book – ‘Bad Science‘ out at the time and which was followed by ‘Bad Pharma’…..i seem to remember reading Bad Science bit by bit when hiding out in the hospital library on nights although i have to admit that i haven’t read Bad Pharma yet. I seem to remember our man Dr Goldacre as having a bit of an anti-thing about several things : nutritionists were one , homeopathy another, quackery and charlatanism in ‘science’ and journalism generally. Just recently i thought to check out Dr Goldacre’s blog ‘Bad science’ to see what if anything he had to say about LCHF and ketogenic diets- so far i can’t find anything, good or bad, which suggests that Ben either hasn’t taken an interest yet or hasn’t found anything particularly ‘bad science’ about either….yet !
Why i mention bad science is for several reasons : first that i was accused of quackery, fakery and pseudo-medical charlatanism when i mentioned my weight loss and anecdotes about reversing my hypertension and arthritic pain with the LCHF diet, IF and some exercise, secondly….that in a discussion with friends who were interested in what i had done i got talking about the science, the evidence and anecdotes from other people and one of my friends made a good comment/question about the ‘science’…..that of “but what can we believe….what is the ‘good’ and valid science and what is faked or manipulated ” ?.
Well, going back to Ben Goldacre and ‘Bad Science’ for a moment, one of his pet hates seems to be the manipulation of data and numbers to conveniently fit the desired result of the study and that seems to be the case often when ‘the study’ is sponsored by a large company that clearly has a dog in the game. Other writers and You-tube presenters such as science journalist Gary Taubes and internet physician Dr Ken Berry say the same kind of thing in different ways, that for example that there is a lot of ‘evidence’ that says that sugar isn’t a problem in a diet and strangely enough that studies that show that seem to have been sponsored by front organisations that lead back to the giant ‘food’ corporations….both Nestle and Pepsi are mentioned in that line.
Perhaps then one valid reason to distrust ‘the science’ is when it has been supported or sponsored by the food and pharma industry, both very big players in the corporate and political worlds : remember that neither the ‘food’ or pharma industry exist to make you healthy, rather they exist to make a legally mandated profit for their shareholders. It could even be said of the pharmaceutical industry that they have a negative interest in making you healthy, after all what they are after is the profit margin from medicines directly caused by their corporate colleagues in the ‘food’ (notably sugar) industry !
In his book Bad Science Dr Ben Goldacre has particular themes that seem to get him a bit ‘worked’ shall we say, things like poor study design and the sometimes fraudulent manipulation of data for example . Well, today for the main part of this post i want to talk about 2 things that get me up on my soap-box , one though is something i actually like and one which i think is one of the worst examples of simply bad or non-science : those 2 things being association and anecdotal evidence
Medicine and science both , don’t seem to like anecdotal evidence but being neither a scientist or a doctor i like anecdotal evidence especially when it’s someones honest observation of something that they have experienced, tried or seen happen themselves : in a way that’s actually the beginning of the true scientific process….see something happen or try something and then wonder why that thing happened. A slightly more advanced version of this we could call an experiment of N=1, where 1 is the number of experimental subjects observed, in this case the person making the self-experiment/observation. Sometimes, even often, we might make a hypothesis of what is going on and we might get it right or wrong but i think that whatever happens the original observation and intent is usually good…..add a whole load of N=1 together and we might be on to something both interesting , valid and repeatable.
Here’s a couple to chew on and spot the mistakes if you can.
Ten years ago i made my first attempt to get very fit and to learn everything i could about training with free weights after a minor accident that i had. One of the things that i did once i started training as a beginner was to join an online Forum (T Nation) that had both a beginners section and an over 35’s group….i kind-of bounced between the 2. I got to know a small group of members that i regularly conversed with online and one of those was an older guy like me….i was in my 50’s…..but who had been diagnosed with Type 2 diabetes. He had been diagnosed a couple of years before, was admittedly overweight when he started training and posting and already on medicines to control both his blood sugar and hypertension.
Anyway, the story is that a few months later he posted to say that he had completely cured or at least reversed his Type 2 Diabetes and his hypertension and he was in a big argument with his doctor about coming off medicines : i think eventually he took matters into his own hands and just stopped taking his meds. Now, things got really interesting on the blog…..a surprising number of other members called him out for being a fraud and even i thought that it was impossible to reverse or cure T2DM. In the end i PM’d him and asked him if he could post or prove his blood ‘numbers’ and he did just that…..i seem to remember that was the first time ever that i’d come across the HbAIC measurement as well as crude blood glucose measurement.
As a nurse i found that really interesting and one quiet night i asked the duty medical registrar what his opinion was and whether the physiology made any sense, i don’t remember much of that conversation although what i do remember is that the same med reg a couple of nights later left an abstract of a paper for me that strongly supported resistance exercise as being able to temporarily reverse insulin resistance so maybe that is what had happened with my online friend.
This year the situation was reversed in that it was me recounting an anecdotal experience, with measurements, and getting called out for it on a completely unrelated group forum. In January of this year it was me who was overweight and hypertensive, i hadn’t had any bloods taken so iv’e no idea what my blood chemistry was like : anyway i started my LCHF/keto experiment, started to lose weight rapidly and one day felt weirdly faint as i was working, and that turned out to be a ‘spontaneous drop in my blood pressure. I had a very pleasant conversation with a very happy GP and came off my medicines…..shortly after that i happened to report my experience of weight loss, reversal of hypertension and huge reduction of new arthritic pain : which was when i was called out on that forum for being a fraud and a charlatan !
Association ( as opposed to causation).
To me, it is when ‘association with ‘ seems to become rapidly conflated as ‘is the cause of’ that my very own personal soap box comes out and on which i have a good old rant.
What are we talking about here ?…..well, it is that lots of things can be seen to (maybe) happen in association with something else ; a measurable value of something in response (maybe) to some other occurrence. One of the earliest and most infamous of these ultimately slipped through the medical back door as what is now known as the diet-heart hypothesis of Dr Ancel Keys back in the 1950’s : to tell that very long story in very short form Keys held the opinion that high cholesterol was the cause of heart disease and heart attacks……even though his own data showed his hypothesis to be nulled !. Keys thought he saw an association between high cholesterol and heart disease and jumped to the conclusion that high cholesterol causes causes heart disease (and heart attacks) and for a while the ‘numbers’ in the USA seemed to support that, as more people turned on to his low fat and high carbohydrate diet. Today, and it seems much more likely that it was the large scale reduction in smoking that was the real reason behind the reduction in heart attacks.
Where that went !…..i was actually taught this by a dietitian .
One of the fun ones today is Vitamin D and was to have been the subject of this little missive today ; i originally wrote it as ‘Bad Science ‘D’ and only half way through realised that i needed to cover a couple of ‘A’ points before jumping to D.
There is a well known study of heart attacks among men in Finland that demonstrates a large degree of association between low levels of Vitamin D and high rates of heart attacks : from that many concluded that low levels of Vitamin D are a direct cause of heart attacks and of course suggested supplementing Vitamin D as a preventative measure. Yet….when Vitamin D is supplemented at clinically significant values guess how much change there was in the rate of heart attacks in the same population ?…..answers on a postcard please.
Don’t worry, i’ll spare you the pain and suspense ; the answer is a big fat zero….no change and therefore a null hypothesis.
So, what was happening in that study ?, glad you asked that : well there is a valid connection and association between the 2 observations , low levels of Vitamin D and high levels of heart attacks but…..more interestingly the rate of heart attacks goes down in the summer when there tends to be a normal rise in northern people’s level of Vitamin D. The mechanism is a simple one, higher levels of UVA sunlight directly causes Vitamin D synthesis in the deep layers of skin and interestingly , that cholesterol is a precursor to Vitamin D….which isn’t a Vitamin anyway so i should stop calling it that.
However, and it’s a big ‘but’ another actual physiological mechanism is taking place here right under our noses and it’s much less well known unless you happen to know the science or , like me, be an ex ICU nurse : that process being another side effect of sunlight on skin and that being the production of Nitric Oxide in the skin….Nitric Oxide (NO) causes dilation of all blood vessels including the coronary arteries. Dilation of blood vessels reduces resistance (SVR) and improves flow while reducing blood pressure….high blood pressure being positively associated with higher levels of heart disease, heart attacks ; and strokes by the way…..not a lot of people know that by the way.
Association is not causation (Dr Robert Cywes : https://www.youtube.com/watch?v=-XyGWshVRJI
Total bollocks by the way (below) in my opinion of course.