Reviews

"There's none so blind as the double blind." Discuss

I confess to a deep unease about therapeutic correctness and the guideline industry. While it is worthwhile knowing that a drug on average does more good than harm when tried on many thousands of patients, ultimately it is its effect on my patient that matters.

Archie Cochrane was extremely sceptical of established practice and questioned everything. The first thing he ever said to me was that he believed lateral chest radiographs were unnecessary, a point that at the time few would have agreed with. When pressed, he admitted that he couldn't interpret them, but that only a trial would convince him of their value. Fortunately I had some evidence. I like to think of him in Heaven, whisky in hand, smiling a sceptical smile as bigger and bigger studies, funded by increasingly anxious drug companies, are required to show smaller and smaller benefits. But I'm sure he is pleased at the scrutiny this initiative has brought to common intractable problems of patient management such as incontinence.

All progress depends on the sceptic

In 1906, Sir Clifford Albutt of Leeds, who introduced the clinical thermometer to medicine, quoted Lloyd Morgan thus: "The prime condition of progress is to think the unthinkable." Indeed, all progress depends on the sceptic, the questioner, the person who does not wholly conform. The greatest threat to the progress of medicine is thus enforced conformity. Most doctors (like everyone else) naturally feel comfortable in conformity, and our organisational structures depend on it. But just think of the effect that Galen's followers had on medical progress.

Which brings me on to guidelines. The principle is good, to indicate to the busy doctor what evidence base there is for the treatment of ill health. But have you noticed how they are now straying from this worthy objective, and how the imprimatur of mention of a drug is being used in promotion? As president of the British Thoracic Society (BTS) some years ago, I discovered that not only were BTS guidelines sponsored by the pharmaceutical industry but that also all members of the guideline committee had received some benefits from that industry.

Since then declaration of interests has been obligatory (and is revealing). Production of guidelines is now a time consuming and expensive business, and difficult to achieve without sponsorship. Many doctors will argue that their receipt of benefits from the industry does not influence their prescribing habits and their advice to colleagues. Nevertheless, there must be the possibility of at least influence in guideline production and more especially in the production of the data on which they are based. It is worth looking at the level of evidence on which the recommendations in guidelines are basedin a surprising number of cases, usually the large majority, it is low or non-existent. This explains the differences between different guidelines on the same disease. Are they much better than textbooks? Which guidelines should we follow?

As a newly appointed chest physician in the early 1970s I found that there was almost no evidence base for the treatment of acute severe asthma. My registrars (mostly) and I set out to find one, and were subsequently able to base our management on our own published double blind studies. Aminophylline came out rather well and pretty well everyone still uses it, but it is not in the recent BTS guidelines. One must ask whether this is because there is no strong commercial interest pushing it. We noticed that chronic "steroid resistant" asthma was a problem and found in a carefully designed, unsponsored study that intramuscular triamcinalone had a useful place in these patients. I still use it, but it is not in the guidelines. And, on the other hand, mite eradication and removal of carpets get a mention in spite of the absence of evidence of their efficacy, and some evidence of inefficacy, in preventing or ameliorating asthma. One may ask, why?

Our conformist nature leads us to rely on guidelines for advice on management and to disregard the strength of the evidence behind their recommendations. I prefer to see them as a challenge. Look at them for evidence of how little we know about patient management and there you will find the opportunities for worthwhile research, especially in non-pharmacological management. Where do you think they are plain wrong? And if your patient tells you she gets benefit from a treatment that isn't in the guidelines, who do you think is right? When I was a student in 1960, an elderly physician said to me "There's none so blind as the double blind." I was a little shocked at his unorthodoxy. I am now that elderly physician.

Footnotes

AS has received no benefits or gifts other than plastic pens from the pharmaceutical industry for 30 years. He has a friend who is manager of a carpet factory and who worries about making workers redundant as a consequence of advice given to patients that their child's asthma will improve if their carpets are removed and replaced by (usually Swedish) wooden floors. He asked AS what scientific basis there was for this advice.

Aberdeen
a.seaton{at}abdn.ac.uk