Sunday, 9 September 2012

Book review: Effectiveness and Efficiency, by A Cochrane

Effectiveness and Efficiency: Random Reflections on Health Services (1972)
Author: Archibald Cochrane
Publisher: The Royal Society of Medicine Press.

Cochrane was a physician and epidemiologist whose request for an organised summary of all randomised clinical trails in medicine was answered in the form of the Cochrane Collaboration ( This classic book, written as an invited lecture, is essential reading for anyone interested in of the evolution evidence based medicine, and anyone sceptical about modern medicine. Archie Cochrane pushed for evidence based medicine before the phrase was coined.  He questioned the effectiveness of much of the (then) current medical practice (such as prolonged bed rest for heart attack patients, oral therapy for diabetes, iron for anaemia, ergotamine for migraine, and my favourite, antidepressants) and was proved right.
Part of the book tells his story, which explains how he became sceptical, and the other part is a comment on the National Health Service in Britain – less relevant now, and even less relevant for those not affected by that system.

The development of his scepticism came about in extreme circumstances. As a prisoner of war in Germany during World War II, Cochrane was the medical officer overseeing 20,000 prisoners, all of whom had diarrhoea, with frequent epidemics of typhoid, diphtheria and other infections. With no medicine (except for aspirin and antacids) he expected hundreds to die. In his six months at the camp, only four prisoners died, three as a result of being shot by their captors. On requesting more doctors from a German officer, he was told “Nein! Artze sind uberflussig”. My favourite quote from the book, which means ‘Doctors are Superfluous’. In the end, Cochrane had to agree with him.

His second experience as a prisoner of war was possibly more enlightening. Here, he was well equipped, with several treatment options at his disposal (including surgery) for the many prisoners with tuberculosis. He realised at the time that there was no good evidence for many of the treatments he was administering, and that he probably “shortened the lives of some of my friends by unnecessary intervention”.

Cochrane expressed frustration when pushing for research into the effectiveness of medicine. The university professors were not supportive, with one stating “The best research is utterly useless”. An interesting comment, as it has some merit because it stands in contrast to the current way of thinking at academic institutions, that research should be utterly profitable; “patent or perish” is the new mantra. It reminds us that many major research discoveries (like penicillin and X-rays) came about by chance, in the pursuit of something unrelated, often with no tangible benefit apparent at the time.

Cochrane also showed further frustration with the funding of research at the time, directed solely to “pure” laboratory research rather than clinical trials of effectiveness. Also (as I have raised before) he gives examples of untested therapies becoming standard practice, such that a randomised trial to properly test it would be considered unethical, only to later find that the treatment was ineffective. Apart from attacking overtreatment, another theme from the book that is recognisable today is that of the overuse ofdiagnostic tests. I like that he casts his net wide, covering internal medicine, surgery, obstetrics and gynaecology, psychiatry and even counselling. It is very hard to do that these days, with ever-increasing subspecialisation.

The book is small and well written, so it is an easy and enjoyable read. The trouble is, it is hard to find. I got mine through Amazon years ago, but I notice that it is no longer available. I think it gets reprinted every now and then, so keep looking, or ask around; you would be surprised how many people keep a copy.

The book is full of good advice, for example: “One should … always assume that a treatment is ineffective unless there is evidence to the contrary”. He emphasises the importance of randomised trials to properly test hypotheses (by minimising bias), but not by dismissing the importance of strong observational evidence.  Examples of treatments validated by strong observational evidence included immunisation for diphtheria and insulin for juvenile onset diabetes.

Cochrane clearly had a good handle on the problems with clinical medicine, and he understood not only the desire to treat, but the desire to be treated. His insights, so early in the development of modern medicine, and his influence since, should be lauded. He stood up to current thinking and remained objective, and he was right.

The best quote is left for the final paragraph, from T.S. Eliot, which Cochrane used as a warning to doctors who desire to treat patients:

Not for the good that it will do
But that nothing may be left undone
On the margin of the impossible

He suggests that we abandon the pursuit of the ‘margin of the impossible’ and settle for ‘reasonable probability’. I agree.


  1. “One should … always assume that a treatment is ineffective unless there is evidence to the contrary”.

    That is actually the null hypothesis, a cornerstone of any scientific research.

    1. Thanks Moriakos, I agree: this should be the way we approach any scientific enquiry.

  2. A good review.
    The book's easy to get:

    I especially like your comment on 'counselling', a particular bĂȘte noire of mine. My favourite Australian example - a school bus caught fire, so all the students were offered 'counselling'.

    1. Thanks Bellthorpe,

      Thanks for the link to get the book. Must admit I was surprised I couldn't get it on Amazon - that is where I by most of my books.

      Yes, counselling is something that is simply assumed to work and to be necessary. I get annoyed when an institution is accused of not providing enough counselling. Clearly, evolution has not taught us humans to grieve properly, so we need instruction. Sounds like I need to review the evidence on that one. Stay tuned.

  3. Off topic now, but is my first port of call to buy any book. The results list is sorted by ascending price, and Amazon is often not the cheapest. More recently, nearly all my books have come from AbeBooks.