How Ideas That Are Not True Become Facts - part 3 of 3


dr_malcolm_kendrick_m.d._134by Dr. Malcom Kendrick, M.D.
Author of The Great Cholesterol Con

When it comes to the creation of an untested auxiliary hypothesis, the French data is the veritable mother lode. This is because the French represent perhaps the greatest single threat to the cholesterol hypothesis.

They eat more saturated fat than any other nation in Europe, their cholesterol levels are slightly above average. They smoke more than most, their blood pressure is around average, they don't exercise much, and they have the same BMI ( body mass index ) as most other countries. Yet, despite all of this, they have the lowest rate of heart disease in Europe. One quarter the rate in the USA, and UK (women and men under 65).

What possible explanation, or explanations, can be there be for this fact? The first immunising tactic used was to call it a paradox. In fact it was described exactly this way in many prestigious journals, from The New England Journal of Medicine to The Lancet and BMJ. (And still is).

However, as mentioned before, you can't really keep on saying something is a paradox. Especially when you are talking about a country with sixty million people in it.

This is not just one paradox... mais non. It represents sixty million living, breathing paradoxes. Therefore, in the end you must try to explain what ‘protects' them. Here is a short selection of the ad-hoc hypothesis developed:

• The French incorrectly certify deaths from heart disease i.e. they don't really have a low rate of heart disease - proven wrong by the WHO MONICA study (still widely believed).

• Red wine protects against heart disease - no outcome evidence.

• Garlic protects against heart disease - no evidence at all.

• Lightly cooked vegetables, which retain anti-oxidants, protect against heart disease - no evidence.

• The French have not had a high intake of saturated fat for long enough for it to affect their heart disease rate, the ‘time-lag' ad-hoc hypothesis - not true, hypothesis proposed in 1998. French rates of heart disease are still falling, twelve years later, and forty years since their consumption of saturated fat was lower than the UK and USA.

I have looked into these and many, many, many more. In the final analysis, they all have four things in common.
One, they were developed with one purpose in mind - to immunize against refutations of the cholesterol hypothesis. Two, they very rapidly became very widely accepted throughout the scientific community. Three, they have no clinical outcome data to support them. Four.... They are not true.

In fact, now that I more fully understand what is actually going on, it has made it relatively simple for me to gauge the truth, or otherwise, of the new ‘risk factors' for heart disease that are announced almost every day.

Just to pick a recent one at random that you may have heard of. ‘Chemicals found within chocolate protect against heart disease.' I read this headline and I thought.... Well you don't really need to know what my first thought was. However, in order to decide if this was likely to be true I asked myself the question. Is this an ad-hoc hypothesis? If so, treat it as highly unlikely.

So I took a few steps backwards. According to the diet-heart/cholesterol hypothesis, chocolate is high in fat, and fat raises cholesterol levels. So, chocolate should cause heart disease - should it not? In fact, chocolate consumption has found to reduce the risk of heart disease. This clearly contradicts the cholesterol hypothesis, and so an ad hoc hypothesis was required.

Now, if I were trying to protect the cholesterol hypothesis, what would I do? I would find some chemical in chocolate - there are many. Then I would do a study to demonstrate that one or more of these chemicals had some physiological effect to explain chocolate's protective effects.

Lo and behold, researchers in Sweden ( 1) (and Swedish researchers always seem the most believable, somehow), have found that catechins and procyanidines, found in dark chocolate, inhibits the enzyme Angiotensin Converting Enzyme (ACE). When ACE is blocked, blood pressure drops. This is actually how ACE-inhibitors (blood pressure lowering drugs) work.

Things became even more promising when researchers found that healthy volunteers given dark chocolate demonstrated an 18% reduction in ACE activity. This is comparable to the effect seen with ACE-inhibitors. It was all looking good, and a story was being stitched together. Hold the front page. We have a biological plausible explanation, and real scientific data to support it.

But before you get too carried away, and start believing headlines such as ‘Why chocolate protects against Heart Disease (2),' you should step back and ask yourself. What is missing here? (Apart from a five year long double-blind, randomised controlled outcomes study, obviously).

What is missing here is a clinical effect. Something mentioned almost in passing by the authors was that there was no drop in blood pressure in those taking cocoa extract. They stated that this meant nothing. In fact the lack of impact on blood pressure was ‘exactly as expected', as it would require a longer time period than the eighteen days of the study to see any effect.

Does this sound reasonable? Of course it does, ad-hoc explanations piled on top of ad-hoc hypotheses always do. But is it likely to be true? Could it really take more than eighteen days for a de-facto ACE-inhibitor to lower blood pressure?

To find out, I decided to review the Cochrane collaboration meta-analysis of all ACE-inhibitor studies to find out how long it takes a pharmaceutical ACE-inhibitor to lower blood pressure. The answer is, between one and twelve hours. ‘ACE inhibitors reduced BP measured 1 to 12 hours after the dose' (3).

To be frank, I already had my doubts about the author's explanation as to why blood pressure did not drop. For the simple reason that one of the widely recognised problems of ACE-inhibitors is that they very rapidly lower blood pressure. So much so, indeed, that you have to start with a very low dose, and gradually increase it to ensure that hypotension does not occur. At one time ACE-inhibitors were only initiated in hospital to avoid this problem.

In short, it is widely known - and proven beyond any doubt - that ACE-inhibitors reduce blood pressure rapidly. Yet, when you eat cocoa extract, which allegedly reduces ACE activity by as much as an ACE-inhibitor, there is no drop in blood pressure - even after eighteen days. Something clearly does not add up here.

Once again, as with all ad-hoc hypotheses, we have an idea that sounds highly plausible....if you don't look too closely. But if you dig down deeper, there is a well camouflaged, but fatal flaw in the evidence. This is far from unusual. In fact, a vital missing link is a common feature of almost all ad-hoc hypotheses. They are like conjuring tricks for the mind. You think you have been told everything that provides an explanation, but the clinical evidence card has been cunningly hidden up the researcher's sleeve.

This does not mean that chocolate does not protect against heart disease through some chemical effect. It may do, it may not. But up to this point, the research has proven nothing. It may be that eating chocolate is so enjoyable that it releases chemicals in the brain that protect you from heart disease. Or it may be that enjoying yourself is good for your health (my preferred explanation).

Equally some ad-hoc hypotheses can be true. Omega 3 fatty acids do reduce the death rate from heart disease (they have an anti-arrhythmic effect, which prevents sudden death, and this has been proven in long-term controlled outcome studies). Also smoking, which was never part of the original diet-heart hypothesis does cause - or at least accelerate - heart disease. Just because a hypothesis is an ad-hoc hypothesis does not inevitably mean it is wrong.

But the vast majority of these ad-hoc hypotheses are wrong. Just to mention one final example. Red wine protects against heart disease. Most people believe this is true. Indeed, if you type the words ‘red wine heart disease' into PubMed you will be deluged with 'evidence' of its protective effects. You will also be inundated with chemical names, such as: resveratrol , and polyphenols and procyanidines, (you may recognise them from the cocoa story) and anti-oxidants. All of which have ‘proven' effects on heart disease.

It will all seems very reasonable, and plausible, and believable, and thousands of papers have been written on this exact topic. You can even buy bottles of wine based on their protective effect on heart disease, dependent on the concentration of various heart healthy chemicals. Thud....(that is the noise of my forehead hitting the desk).

But if you decide to look for the outcome study demonstrating that red wine reduces the risk of heart disease you will look long and hard. And I can guarantee that you find nothing. Yes, the French drink red wine and have a low rate of heart disease. This is not proof of anything. Rich middle class people in the US also drink red wine, and they too have lower rates of heart disease than the surrounding population....quelle surprise. (See under hormone replacement therapy).

But the main reason why I can be virtually certain that red wine does not protect against heart disease (any more than moderate consumption of any other form of alcohol) is because of why the idea was born. As an ad-hoc hypothesis designed to protect the cholesterol hypothesis. Pure and simple.

At this point I think I may have answered the question posed in the title of this article ‘How ideas that are not true become facts.' At least I think I have. Hopefully, after you have read this article it will be easier for you to decide if a fact is likely to be true, or a mere conjuring trick of the mind.

Dr. Malcolm Kendrick (MbChB MRCGP) M.D.
Dr. Kendrick has worked in family practice for twenty years.
He has specialized in heart disease and set up the online educational website for the European Society of Cardiology.
He is a peer-reviewer for the British Medical Journal.

How Ideas That Are Not True Become Facts part 1

1: Inrgid A-L person et al. ‘Effects of Cocoa extract and Dark Chocolate on Angiotensin-Converting Enzyme and Nitric Oxide in Human Endothelial Cells and Healthy Volunteers.' Journal of Cardiovascular Pharmacology, 2010.
2: www.sciencedaily.com/release/2010/11/10/1011014127.htm
3: http://www2.cochrane.org/reviews/en/ab003823.html

November 2011

Books From Amazon

The Dark Side of Statins
The Statin Damage Crisis
Cholesterol is Not the Culprit
Statin Drugs Side Effects
Lipitor, Thief of Memory


Over 12,000 reader posts:

spacedoc Forum