A Collection of Strange Beliefs, Amusing Deceptions, and Dangerous Delusions

From Abracadabra to Zombies | View All

Cochrane Collaboration

The Cochrane Collaboration is a large group of academics who publish systematic reviews of medical research papers regarding the effectiveness of various health care interventions, conventional and unconventional. The Cochrane Collaboration was founded in 1993 under the leadership of Iain Chalmers. The reviews are published in the Cochrane Library.

Experts from around the world examine clinical trials and observational studies in an effort to determine what treatments work and what treatments are ineffective or work no better than placebos. According to Wikipedia, in October 2010 the Cochrane Collaboration had "over 27,000 volunteers in more than 90 countries." Ernst and Singh (2008) write that the Cochrane Collaboration has twelve centers around the world and over 10,000 "health expert volunteers from over 90 countries."

As Ernst and Singh note: it would be "impractical and nonsensical" for doctors to read the hundreds of studies that come out each year supporting or undermining medical treatments. The task is daunting. Studies often contradict one another. Some are very well designed and meticulously conducted. Others are small and sloppy. Many are a mixed bag of good and bad design and execution.

The Cochrane Collaboration is named after Archie Cochrane (1909-1988) who lamented that the medical profession to which he belonged had not "organized  a critical summary, by specialty or subspecialty, adapted periodically, of all relevant randomized controlled trials." Thus was born the idea of a systematic review. Studies on a particular treatment would be given different values depending on how they were designed, how many subjects were in them, etc. A systematic review does not assume all studies are of equal value. The goal is to determine in an unbiased way what the best evidence is for any particular treatment.

Ernst and Singh write:

As well as providing judgments on the effectiveness of pharmaceutical drugs, these systematic reviews evaluate all sorts of other treatments, as well as preventative measures, the value of screening, and the impact of lifestyle and diet on health.

Fortunately, many Cochrane Collaboration reports are available online. So, if a medical practitioner or a potential patient wants to know, for example, what is the best evidence for using acupuncture to treat migraines, she can consult the Cochrane Collaboration to find an unbiased, independent, rigorous, and relatively up-to-date report. Just google "Cochrane Collaboration acupuncture migraine" and you will find what you're looking for.

Unfortunately, the practitioner or potential patient may still have to do some thinking even after consulting the Cochrane Collaboration. If you followed the link on acupuncture for migraine you found:

Collectively, the studies suggest that migraine patients benefit from acupuncture, although the correct placement of needles seems to be less relevant than is usually thought by acupuncturists.

The traditional practitioner should be disconcerted by the fact that it doesn't seem to matter where the needles are placed to get the result. This conflicts with traditional teaching that the needles must be placed in specific points along alleged meridians that carry chi. This fact implies that there is something wrong with the traditional basis for acupuncture. On the other hand, the fact that migraine patients benefit from both true and sham acupuncture indicates that sticking needles into migraine patients is an effective treatment. However, if one is unfamiliar with the fact that there have been recent developments in the testing of acupuncture that involve a sham procedure that does not involve sticking a needle into the patient at all (a retractable needle is used), one might jump to the conclusion that sticking patients with needles triggers some sort of mechanism that blocks migraine pain. One might, for example, infer that the effectiveness of acupuncture can be explained by the gateway theory of pain. But there are problems with that interpretation.

The latest results contradict some of the more positive conclusions from Cochrane reviews. If these results are repeated in other trials, then it is probable that the Cochrane Collaboration will revise its conclusions and make them less positive. (Ernst and Singh 2008)

One must also consider that a number of studies have found no significant difference in effectiveness  between true and sham acupuncture. Many of these studies do not involve measuring pain, so if there is a mechanism that explains the effectiveness of acupuncture for, say, nausea, it must be something other than the gateway theory mechanism. Many critics of alternative therapies are betting that the effectiveness of acupuncture can be explained by placebo effects; they refer to acupuncture (and other alternative therapies such as homeopathy) as placebo medicine. What these critics mean is that the effectiveness of acupuncture is not due to sticking needles into the skin, but is due rather to a host of other factors, such as the expectations of the patients, the conditioning of the medical ritual, the relaxation effect that comes from being treated by a professional in a professional setting, the Hawthorne effect, etc. Also affecting the outcome of acupuncture trials are such things as spontaneous improvement, fluctuation of symptoms, regression to the mean, and a desire to please the healer. Further compounding the problem of establishing the effectiveness of acupuncture above a placebo effect is the seeming impossibility of devising a double-blind test. The retractable needle may solve the problem of blinding the patient to the procedure, but how does one blind the acupuncturist?

This might seem unimportant, but there is a risk that the practitioner will unconsciously communicate to the patient that a placebo is being administered, perhaps because of the practitioner's body language or tone of voice. It could be that the marginally positive results for acupuncture for pain relief and nausea apparent in some trials are merely due to the slight remaining biases that occur with a single blinding. (Ernst and Singh 2008)

While the Cochrane Collaboration is an excellent place to begin one's research on a health-care treatment, it is not an infallible guide and should be considered within the context of all the available evidence regarding a treatment.

See also control study and meta-analysis.

further reading

Ernst, Edzard and Simon Singh. 2008. Trick or Treatment: The Undeniable Facts about Alternative Medicine. W. W. Norton.


The Cochrane Collaboration - Top 50 Reviews (On 26 October 2010, the number one abstract accessed was on drinking green tea to prevent cancer. The bottom line: "There is insufficient and conflicting evidence to give any firm recommendations regarding green tea consumption for cancer prevention."

The Cochrane Library - Wikipedia

Last updated 15-Dec-2013

© Copyright 1994-2013 Robert T. Carroll * This page was designed by Cristian Popa.