You don't need to be signed in to read BMJ Blogs, but you can register here to receive updates about other BMJ products and services via our site.

Edzard Ernst: Absence of evidence is not evidence of absence

19 Mar, 12 | by BMJ

“Absence of evidence is not evidence of absence.”

Amongst the many misleading arguments, which are frequently used to promote useless treatments, this one occupies a prominent place. When I first heard it, I was impressed: it is succinct and elegant. In fact, it is also entirely logical: the absence of evidence for extra-terrestrial life represents no evidence that such life does not exist; just because you have not seen someone being struck by lightning does not mean that lightening does not hit people; you may never have seen the Northern Lights, yet they do exist.

If the argument is correct, how can it be simultaneously misleading? The fallacy arises not from the argument itself, but from the way it is often used in the promotion of quackery. Prince Charles’ “Duchy Originals” are selling a herbal “Detox tincture” made from extracts of Dandelion and Artichoke. For neither ingredient is there good evidence that it eliminates toxins from the body. Clearly, therefore there is an “absence of evidence.” Ask any proponent of this type of detox and she will tell you that this “absence of evidence” is not “evidence of absence” of an effect. “Science has found no proof yet,” she might say, “but this does not mean that it does not detoxify my system.” Ask a chiropractor who advocates chiropractic care for otitis or bed-wetting; they might agree with you that there is no evidence of effectiveness—but they are bound to claim that this does not constitute evidence that chiropractic does not work. Ask a homeopath who treats autistic children why he prescribes homeopathic sugar pills despite the absence of evidence—the answer will be that the absence of evidence does not provide evidence of the absence of a benefit.

In alternative medicine, this argument is used to silence doubters and critics. As long as you cannot show that an unproven treatment definitely does not work, we are all supposed to give it “the benefit of doubt” because it might just work, because patients do like it, because it has been around for hundreds of years, because Prince Charles’ firm sells it, etc, etc.

This argument can appear so compelling that it is easy to forget why it is misleading. In real medicine, we rarely give “the benefit of doubt” to unproven treatments. In the interest of our patients, we usually consider any therapy ineffective until someone shows us sound evidence to the contrary. The onus to demonstrate effectiveness of a treatment rests squarely on the shoulders of those who promote it by claiming that it is useful.

Not so in alternative medicine! Here people tend to argue precisely the other way round: as long as we have not proven that a treatment is ineffective, we should happily use it. This implies that those who object to this line of thinking have to demonstrate that the treatment in question does not work. In other words, the burden of proof is reversed. This reversal is risky to the point of endangering the health of patients.

The “absence of evidence…” argument in itself is fine; in the hands of promoters of nonsense, however, it all too often turns into a prime example of apparently plausible yet fundamentally flawed logic which can mislead us all.

Edzard Ernst is a professor of complementary medicine at the Peninsula Medical School, Exeter.

By submitting your comment you agree to adhere to these terms and conditions
  • Rachael Forrest

    I often agree with prof Ernst, despite the fact that I work as an acupuncturist and my chosen line of practice is in the firing line of his attacks. I think we need rigorous trials and tests so that patients are not misled. In this particular column though, what he has forgotten is that what he calls “real” (I.e. Pharmaceutical) medicine requires such rigorous testing is because it can cause so much harm. I’m not a fan of homeopathy but I don’t think anyone has complained of vomiting, diahorrea, anxiety etc etc (refer to your nearest BNF) on taking it. Nor can acupuncture cause any serious side-effects except if used by an un-trained practitioner, and you can say that of “real” medine too.

  • Richard Lanigan

    As usual you take a very narrow view of evidence based medicine in a clinical setting. So let me quote David Sackett, in relation to the importance of “Clinical Expertise” and as a qualified chiropractor I would be happy to have my “clinical expertise” compared with any medically trained person.

    “The ability to use our clinical skills and past experience to rapidly identify each patients unique health state and diagnose their individual risks and benefits of potential interventions, and their personal values and expectations”.

    So fair to say there is a bit more to clinical science, than what Edzard Ernst can find on Pub Med

  • Ulrik Felding

    To Rachael Forrest…
    So you are saying that because there are limited side effects to homeopathy and acupuncture we should not test these as rigorously as pharmaceutical drugs?

  • Rich Moore Stinky

    I guess it depends whether you are testing for efficacy or safety….

  • notactualsize

    Fantastic. Some people who believe in God, of course, use this argument against those who don’t [believe in God]…

  • Tea And Blues

    I certainly consider absence of evidence as *evidence* of absence. Any evidence that dinosaurs built helicopters is entirely absent. ;) However, absence of evidence should never be mistaken as *proof* of absence. I think that’s the most common way this phrase is abused, with alt med supporters conflating the concept of “evidence”, which isn’t necessarily conclusive, with “proof”, which is conclusive.

  • http://www.chiropracticlive.com/ Richard Lanigan

    Noctualise, raises an interesting question which I have asked you about today  on twitter Edzard, you like to set an agenda but seem reluctant to answer questions about what you are proposing. 

    Do you think the NHS should provide priests or chaplains to patients who want to pray for better health. There is no evidence of the existence of a God however like homeopathy it provides comfort to many people. Most clinicians would encourage the priest to be at the patients bedside, was wondering what your data base advises? 

You can follow any responses to this entry through the RSS 2.0 feed.
BMJ blogs homepage

The BMJ

Helping doctors make better decisions. Visit site



Creative Comms logo

Latest from The BMJ

Latest from The BMJ

Latest from BMJ podcasts

Latest from BMJ podcasts

Blogs linking here

Blogs linking here