In Defense of Nothing

The inestimable Shaun Holt has released a survey on placebo usage in general practice. Showing that 3/4 of doctors have used placebos on patients and 1 in 7 do this more than 10 times a year. The Dominion Post has the particularly hysterical headline Doctors prescribe drugs that don’t work. This is probably true in a way – many medicines we use are certainly not as effective as drug companies make them out to be –  but that is not what the placebo effect is about. The placebo effect is a significant healing effect produced by simple autosuggestion, usually in the form of a pill or potion that is otherwise inert or, at least, harmless. Many areas of alternative medicine seem to derive their power entirely from this effect and, despite the caustic remarks from some parts of the medical profession, that healing is not invalid. The body has enormous recuperative powers, and if you can harness that power by a little quackery, I don’t have any serious objection, provided one is not usurping a proper evidenced-based therapy to use a placebo therapy. For instance, using echinacea to treat a cold is perfectly acceptable, but using crystals to treat a pneumonia is not. Shaun has a highly amusing video on his blog entitled If Homeopaths ran A&E . I particularly like the part where they put three drops of highly dilute Ford Mondeo on the crash victims tongue…

So, used appropriately, placebos can be an effective form of treatment. Yet the use of placebos is subject to silly myths such as these:

“However, given the deception involved, it has been suggested that placebo use could harm the doctor-patient relationship. The American Medical Association warns that placebos are unethical and could expose doctors to malpractice suits.”

The first sentence shows a deep misunderstanding of the placebo effect. The reason why it works is because a good doctor-patient relationship has already been established. Indeed, this is a pre-requisite for a placebo to work. The deception is entirely harmless and in the best interests of the patient. In fact, the whole process works best when the therapist thinks that the therapy has value as well, which is why the placebo effect works so well in alternative medicine. Even in the uncommon event that the doctor is “found out”, the patient is usually quite understanding, even appreciative. Doctor-patient relationships that are damaged by the use of placebos are, invariably, fairly poor relationships already, with the doctor prescribing the placebo in an arrogant, contemptuous way. Note that the patients most likely to find a placebo offensive are the ones least likely to be given a placebo.

The second sentence, about the American Medical Association, says far more about the absurd state of frivolous litigation in America than it does about the ethicality and legality of placebo use.

Shaun does bring up one legitimate concern, and that is the use of antibiotics in viral upper respiratory infections. Antibiotics are of no use in viral infections, although they can be used to treat super-added bacterial infections such as strep throat and pneumonia. My feeling here is that this use is not merely to take advantage of the placebo effect, but is a consequence of a real diagnostic dilemma.

Take a sore throat, for instance. Most of these are viral and don’t need an antibiotic. But you can’t tell that just by looking at a red throat, you need to take a throat swab, send it away and culture it. This takes 3-4 days. Leaving your patient with an untreated strep throat for this long will not make you any friends. Consequently you prescribe an antibiotic up front. But stopping an antibiotic before the course is finished is the prime way to generate resistance. This means that even if the throat swab is negative, the patient should still finish the course – hence the swab is redundant and the antibiotic will be a placebo 9 times out of 10 (or, perhaps 5 times out of ten, if you are a really good diagnostician).

The same principle applies to bronchitis and viral chest infections (although the diagnostic evidence is a little firmer).

It is for this sort of reason that I don’t consider antibiotic use in viral infections to be a true use of placebo. I also am dubious about some of the tales of antibiotic resistance, as resistance seems to plateau after a few years. We have been using Amoxycillin for decades but it still seems roughly as effective for strep throat as it did 30 years ago. That’s just my impression. Feel free to correct me, you pharmacologists and microbiologists out there.

Although I don’t use placebos often, I would be most reluctant to see the Medical Council trying to legislate around the issue. Placebos being used in an abusive manner (as in “you are mad, take these sugar pills”) are a rarity and the Council would have no difficultly disciplining a doctor around such an issue, under current guidelines. Attempts to limit the use of antibiotics prescribed for viral illness are likely to be either overkill or doomed to failure. Bear in mind that it is the GP who has the patient in front of him, not the Medical Council. It is therefore the GP who is best able to decide what is most appropriate for his patient. Guidelines that are too prescriptive become dangerous. If there is one thing that I have learnt in 30 years of medicine is that patients do not follow pathways.  And neither should doctors.

Related posts: