The placebo effect is a very powerful tool, and an intriguing window into the power of the human brain. This blog offered an illuminating post in the past on the power of placebos. We also carried a post on how placebos can mimic the immunosuppressive effect of cyclosporine. Attempting to use placebos in clinical practice is clearly skating out into the middle of the pond on very thin ice. But understanding the power of placebos may help clinicians appreciate how and why some treatments work better or worse than others, when the physiology and the pharmacology do not explain what we see.

Adding to this intriguing body of knowledge is an article recently published on-line by the Annals of Internal Medicine. The authors presented a meta-analysis of 149 randomized studies involving close to 40,000 patients to assess the impact of different placebos for knee osteoarthritis. The meta-analysis compared standard pharmaceutical treatment for this diagnosis with four types of placebos – oral, intra-articular, topical, and oral-plus-topical.

The results were interesting. Notably, different routes of placebo administration had significantly different effects. The most powerful placebo was the intra-articular injection. It was significantly more effective than oral acetaminophen. Next came topical placebo, which was comparable to oral acetaminophen. Oral placebos showed little efficacy.

So, not all placebos are created equal. Why would this be the case? To understand this phenomenon, it appears we need to turn to the power of belief and the effect of the human brain on the other parts of the body. Getting an injection, even if what is injected is normal saline, is perceived by the brain as a very strong treatment. The fact it elicits a stronger therapeutic effect than other placebo delivery methods is not surprising. One may possibly invoke a physiologic phenomenon from the injection itself: the inflammatory substances in the synovial fluid may be diluted, or perhaps the injection leads to some positive immune activity which may counteract pain in some fashion.

But why would a topical placebo lead to greater pain reduction than an oral one? Once again, the brain appears to be the culprit. Pills are so commonly used for so many conditions that one’s expectations of them are well-established. But a nice strong cream, rubbed diligently onto the skin of a painful knee, is more likely to be helpful. Or so the brain appears to think.

To repeat, we are not encouraging the use of placebos in medical practice. But it is nonetheless fascinating to realize and appreciate the therapeutic impact of belief systems. In fact, much of what we as clinicians accomplish probably has as much to do with the placebo effect as it does with our pharmacologic and procedural interventions. This is not a cause for nihilism or cynicism. Rather, the amazing power of the placebo effect lends credibility to the fact that medicine, when practiced effectively, is both a science and an art.

Richard Fleming, MD