May 24, 2011


Think yourself better: Alternative medical treatments rarely work. But the placebo effect they induce sometimes does (The Economist, May 19th 2011)

Over the years Dr Ernst and his group have run clinical trials and published over 160 meta-analyses of other studies. (Meta-analysis is a statistical technique for extracting information from lots of small trials that are not, by themselves, statistically reliable.) His findings are stark. According to his “Guide to Complementary and Alternative Medicine”, around 95% of the treatments he and his colleagues examined—in fields as diverse as acupuncture, herbal medicine, homeopathy and reflexology—are statistically indistinguishable from placebo treatments. In only 5% of cases was there either a clear benefit above and beyond a placebo (there is, for instance, evidence suggesting that St John’s Wort, a herbal remedy, can help with mild depression), or even just a hint that something interesting was happening to suggest that further research might be warranted. [...]

A placebo is a sham medical treatment—a pharmacologically inert sugar pill, perhaps, or a piece of pretend surgery. Its main scientific use at the moment is in clinical trials as a baseline for comparison with another treatment. But just because the medicine is not real does not mean it doesn’t work. That is precisely the point of using it in trials: researchers have known for years that comparing treatment against no treatment at all will give a misleading result.

Giving pretend painkillers, for instance, can reduce the amount of pain a patient experiences. A study carried out in 2002 suggested that fake surgery for arthritis in the knee provides similar benefits to the real thing. And the effects can be harmful as well as helpful. Patients taking fake opiates after having been prescribed the real thing may experience the shallow breathing that is a side-effect of the real drugs.

Besides being benchmarks, placebos are a topic of research in their own right. On May 16th the Royal Society, the world’s oldest scientific academy, published a volume of its Philosophical Transactions devoted to the field.

One conclusion emerging from the research, says Irving Kirsch, a professor at Harvard Medical School who wrote the preface to the volume, is that the effect is strongest for those disorders that are predominantly mental and subjective, a conclusion backed by a meta-analysis of placebo studies that was carried out in 2010 by researchers at the Cochrane Collaboration, an organisation that reviews evidence for medical treatments. In the case of depression, says Dr Kirsch, giving patients placebo pills can produce very nearly the same effect as dosing them with the latest antidepressant medicines.

Pain is another nerve-related symptom susceptible to treatment by placebo. Here, patients’ expectations influence the potency of the effect. Telling someone that you are giving him morphine provides more pain relief than saying you are dosing him with aspirin—even when both pills actually contain nothing more than sugar. Neuro-imaging shows that this deception stimulates the production of naturally occurring painkilling chemicals in the brain. A paper in Philosophical Transactions by Karin Meissner of Ludwig-Maximilians University in Munich concludes that placebo treatments are also able to affect the autonomic nervous system, which controls unconscious functions such as heartbeat, blood pressure, digestion and the like. Drama is important, too. Placebo injections are more effective than placebo pills, and neither is as potent as sham surgery. And the more positive a doctor is when telling a patient about the placebo he is prescribing, the more likely it is to do that patient good.

One wonders if one of the reasons that people in countries with socialized medicine are "more healthy" is just the psychological factor of knowing they have access to care (at least in theory).

Posted by at May 24, 2011 5:39 AM

blog comments powered by Disqus