Epidemiology uncovered

At the root of many contemporary health scares is the tendency to confuse sequence with consequence, and association with causation.

Dr Michael Fitzpatrick

Topics Politics

Post hoc, ergo propter hoc

Viewers of the US TV series The West Wing, currently running on Channel 4 in the UK, will recall the exposition of the meaning of this Latin aphorism by President Bartlett (Martin Sheen) to his White House staff.

‘After this, therefore on account of this’ – the aphorism ridicules the tendency to confuse sequence with consequence, association with causation. It was an error as familiar in Ancient Rome as it is in modern Washington, or London.

This confusion lies at the root of many contemporary health scares – such as those arising from the fear of deep vein thrombosis (DVT) among air passengers, the alleged link between the MMR immunisation and autism, and the supposed risk of childhood leukaemia from the electromagnetic fields (EMF) arising from overhead electricity cables. All three scares have been the subject of intense controversy in recent months and have generated widespread popular anxieties.

The science of epidemiology has emerged in part through the recognition of associations between environmental factors and the development of diseases. Its most famous historical moment was the link noticed by John Snow in 1854 between patients who contracted cholera and the pump from which they drew water in Broad Street in London’s Soho. A scarcely less celebrated discovery came a century later, when Austin Bradford Hill and Richard Doll observed the link between smoking and lung cancer.

The subsequent identification of numerous associations between risk factors and diseases – some 247 in relation to coronary heart disease alone, according to one survey – has led to attempts to clarify the distinction between association and causation.

In the 1960s, Bradford Hill himself proposed a set of criteria which would allow epidemiologists to judge whether an association was likely to be causal (1).

The fact that the association between cigarette smoking and lung cancer meets each of these criteria provides powerful evidence that indeed smoking causes cancer:

the association is strong: the risk of a smoker dying of lung cancer is 25 times that of a non-smoker;
the association is graded: the more you smoke, the greater the risk of cancer;
the association stands independent of confounding variables, such as class, gender, race, occupation;
the association is consistent: it has been observed in different types of study, in different study populations;
the association is reversible: if you stop smoking, your risk of cancer declines;
the association is plausible: cigarette smoke is known to contain substances that cause cancer (carcinogens).

‘Here then are the different viewpoints, all of which we should study before we cry “causation”’, insisted Bradford Hill. As James LeFanu comments, ‘in the following decade Bradford Hill’s cautious and logical views were ignored as epidemiologists claimed to have found that virtually every aspect of people’s lives was implicated in some illness or other, generating both anxiety and confusion in the public mind’ (2).

He might have added that in subsequent decades, this process has, with the support of politicians, lawyers and the media, gathered momentum to become a major influence in the life of the society, which is now permanently in the grip of one health scare or another.

Now, here is an interesting exercise. If you take the three current health scares noted above (MMR/autism, air travel/DVT, EMF/leukaemia) and apply the Bradford Hill standards to them, you will find that, in each case, the associations observed meet none of the criteria for causation.

It is rather sad to find Richard Doll, now the grand old man of British epidemiology, lending his authority to the latest claims for the association between EMF and leukaemia, which fall far short of satisfying the standards set out by his former mentor.

It is even sadder that, half a century after the revelation that smoking causes lung cancer, so few other environmental factors have been conclusively shown to cause cancer or other serious diseases. Hence the quest for strategies of prevention has led medical authorities – increasingly backed by political authorities – to focus on risk factors for which the evidence falls far short of demonstrating a causative role in disease.

The result is continuing controversy over the status of risk factors and the efficacy of prevention, and growing public anxiety about diverse aspects of everyday life, from diet to sunbathing, immunisation to air travel.

Given the damage to public health in its widest sense that now results from epidemiological propaganda that, in its confusion of association and causation, falls so far short of the standards of its own discipline, it is surely time to blow the whistle.

As LeFanu concludes, ‘The simple expedient of closing down most university departments of epidemiology could both extinguish this endlessly fertile source of anxiety-mongering while simultaneously releasing funds for serious research’ (3).

Dr Michael Fitzpatrick is the author of MMR and Autism, Routledge, 2004 (buy this book from Amazon (UK) or Amazon (USA)); and The Tyranny of Health: Doctors and the Regulation of Lifestyle, Routledge, 2000 (buy this book from Amazon UK or Amazon USA). He is also a contributor to Alternative Medicine: Should We Swallow It? Hodder Murray, 2002 (buy this book from Amazon (UK) or Amazon (USA)).

Read on:
Power cables – what risk? by John Brignell
No smoking issue
(1) For an excellent account of the history of modern epidemiology, see James LeFanu, The Rise and Fall of Modern Medicine, Little Brown 1999, p46-59. Buy this book from Amazon (UK) or Amazon (USA)
(2) James LeFanu, The Rise and Fall of Modern Medicine, Little Brown 1999, p59
(3) James LeFanu, The Rise and Fall of Modern Medicine, Little Brown 1999, p404

To enquire about republishing spiked’s content, a right to reply or to request a correction, please contact the managing editor, Viv Regan.

Topics Politics


Want to join the conversation?

Only spiked supporters and patrons, who donate regularly to us, can comment on our articles.

Join today