When someone provides yet another whacky alternative therapy claiming anecdotal evidence of its success, we think of the placebo effect.
What's a placebo?
A placebo is a fake medicine used in double-blind trials so that the person treated doesn't know if they've been given real medicine or not. It's double-blind because even the doctor/therapist doesn't know if they've handed over the real medicine.
This sort of trial is carefully designed to eliminate bias. Neither doctor nor patient knows if the medicine is real and it won't be revealed until the results are in. That rigorously enforced impartiality is crucial to establishing scientific results and it means that anyone can reproduce the same trial with the same degree of objectivity - the results can be checked by anyone. Of course that's infinitely more sound that accepting probably biased anecdotal evidence: anecdotal evidence usually comes from people who have paid not inconsiderable sums for the cure and therefore they have a psychological predisposition to agree that it was effective. Double-blind trials eliminate this bias, anecdotal claims rely on it.
Of course, if people are suffering from an ailment and some alternative remedy provides relief based on the placebo effect, then surely that's better than not having it? Certainly for those individuals who react to the suggestion of treatment in this way, they are better off but it is not so clear cut.
The placebo effect is unpredictable
Unfortunately only around 30% of us are susceptible to a placebo effect and it is not possible to identify in advance which of us are. Not only that but at different times those who are susceptible might be a different 30%. And since placebos typically don't take effect immediately, it's rather difficult at any one time to determine if there is an effect or not.
Placebo effects do not persist and sometimes tail off even during the course of a trial.
Ethically dodgy?
There are ethical and legal problems as well with any suggestion of using placebos for therapy or treatment. One obvious issue is that where there is no clear mechanism for treatment, a patient may justifiably be suspicious of fraud - if someone is charging for a cure and the cure turns out to be auto-suggestion, they are likely to be disatisfied.
But there's also the potential for the supression of genuine clinical symptoms with patients claiming cure when there is no real cure at all.
With the diminution of placebo effect over time, unscrupulous or simply ill-informed amateur therapists may exploit the situation to argue for repeat custom.
What constitutes treatment and cure?
Medical authorities generally require a treatment to show an improvement greater than that attributable to placebo alone. In other words, it has to be shown that there is a measurable effect over and above that which is shown in the case of placebo.
So even if anecdotally some patients getting alternative medicine therapies report that they feel better, that cannot be used as evidence of its efficacy nor can it be accepted that the individual is actually getting real treatment even in terms of the placebo effect. It doesn't mean either that the placebo effect will continue.
The placebo effect is really the psychological noise which masks the identification and treatment of real medical symptoms. It has to be removed before it is possible to relate treatment to identifiable causes and without that, we have no basis for expecting the treatment to be successful. So using the placebo effect as treatment is a non-starter.
Unfortunately we can't treat people with placebos because it doesn't treat anything, it simply masks the symptoms. That's a shame because it would be very cheap to treat everyone with placebos for the many ailments they present with. Understanding how the placebo is used in double-blind trials shows why evidence is necessary as the basis for relating the treatment to the diagnosis and not simply heresay.
