• We have updated the guidelines regarding posting political content: please see the stickied thread on Website Issues.

The Placebo Effect


Gone But Not Forgotten
Jan 17, 2002
Scientific trials testing both new medicines and alternative treatments such as homeopathy usually compare groups of people who actually take the drug with people who just think they are taking it but are really just taking chalk pills or drinking distilled water.

Many people taking the placebos actually do report a positive (or negative) effect. What is going on here? Such results are usually just dismissed as the placebo effect and more of a nusiance than a possibly important insight into the power of the mind.

I say again, what is going on here? Has any research been done into how the placebos work?

In the TV series on human senses last week, volunteers were tested for their reaction to pain. Then they were divided into 2 groups. One group was given pills that were painkillers, and sure enough, the volunteers could now tolerate more pain.

The other group were given pain enhancers, and sure enough, they were now more sensitive to pain.

But (you guessed it!) actually both sets of pills were placebos!

It just demonstrates the power of suggestion. The same prog showed the presenter being hypnotised against pain, and feeling nothing when several needles were stuck in the back of his hand.
rynner said:
It just demonstrates the power of suggestion.

That's exactly my point! How? Why? What could we learn from this oft-dismissed "well known fact"?

I think a lot of medecine works because we believe it will work. The mind is a powerful thing if we think we will get better as a lot of illness cane be psychosomatic anyway, especially stress related ones.
However, a friend of mine was part of a trial for a new painkiller when she had her wisdom teeth out and she had the placebo. She suffered terrible pain.
Moral of the story: It doesn't work every time!
Loads of research on placebo effect, especially in psychological literature. Just get yourself a good Psy book and you can read all about it.
mejane said:
That's exactly my point! How? Why? What could we learn from this oft-dismissed "well known fact"?


Jane, to answer your underlying question, I think it's because humans quite literally create their own reality to a fairly large extent. Therefore, the reality that you expect to occur, is more likely to occur. Literally.
For instance, if I genuinely believe I'm going to be a millionaire, and entertain no inner doubt that I will be a millionaire, OK, I won't be a millionaire tomorrow, but in five years it's quite possible that I will be, if I work hard at it. But if I very much doubt that I will ever be a millionaire, the chances of my being a millionaire are vanishingly small, unless I win the lottery. Is being a millionaire a literal thing? Yes, it can be confirmed by a bank statement.
What a placebo does is to give you a postulate of a new reality, whether good or bad. A literal reality. So if you get the pill which you think will cure the illness, bingo! You yourself create the reality in which you get better.
So the reason the power of suggestion works is because it helps a person create a literal reality. I got the impression this was the question you were asking, rather than the fact that a placebo can work etc. :)

Big Bill Robinson
Well, I used to have a link for this story, but it's long gone now;
when testing a particular drug to treat depression, the placebo worked better that the actual drug, and had no side effects;
The main reasons given for all this were the undesirable side effects of the real drug.
even though it did the job it was designed to do it couldn't compete with the placebo.
but (as far as I can remember) when the control group were told that they had recieved a placebo, they quickly became depressed again.
Works a treat with kids, give them any old stuff ,tell them its "strong medicine", and they fell much better,even my autistic kid, who dos'nt really speak will ask for a plaster for a "hurt".
The placebo effect is much more complicated that you'd think.

For example, it's commonly found in hospitals and hospices that a patient with control over their pain via a special pump will often use far less of the painkilling drug than expected. This is because they are confident of being able to stop the pain when necessary and so feel able to leave off pain relief until it really 'stings' (as we say) rather than asking for it in advance.

People can cope with quite a lot of pain if they feel in control. What really scares them is not the pain itself but the helplessness they might experience if it gets too much for them.

Women in labour are a good example of this- most can give birth quite calmly with the help of just Entonox, which is a fast-working self-administered gas anaesthetic.

Done this myself 4 times- piece of cake!
I remember feeling in total control of my pain and confident of leaving the gas until the last minute because I knew it would work when I wanted it to. ;)
On a related note, if people can "think themselves better" then can they also "think themselves ill"? Do hypocondriacs really get more/worse illnesses?

If that doesn't make sense, just ask Bill what my real question is :D

Interesting article in the Guardian:

What's wrong with the placebo effect?

Ben Goldacre
Thursday April 15, 2004
The Guardian

· For some strange reason I've never understood, pseudoscientists tend to get huffy when you suggest that their cash cow only works through the placebo effect; perhaps they were so distracted by their sea of flawed research into alternative therapies that they missed the excellent crop of good scientific studies on the placebo.
So we know that placebos can affect lots of things, especially stuff with a subjective component, like pain, or mood; and we know that two placebo sugar pills have a bigger effect than one, and that an intramuscular placebo injection is more effective than a placebo sugar tablet. But what grumpy alternative therapists miss is that placebo goes well beyond dishing out sugar pills: it's the ceremony, and the cultural meaning of the treatment.

Confidently waving an ultrasound machine around someone's face is effective for post-operative dental pain, regardless of whether the machine is switched on. Likewise, in the 1950s, we used to ligate the internal mammary artery to treat angina: but when someone did a placebo-controlled trial, going to theatre, making an incision, but only pretending to ligate the internal mammary, the sham operation was as effective as the real one. Like morons, instead of applauding the power of the placebo, we just stopped doing the procedure, assuming that it was "useless".

· It goes on: pinky red sugar pills are more effective stimulants than blue sugar pills, because colours have meanings. And a four-way comparison, with either sugar pills or aspirin, in either unbranded aspirin boxes or mock-up packaging of the Dispirin brand, showed that brand-name packaging, and the wealth of advertising and cultural background material that packaging plays on, had almost as big an impact on pain as whether the pills had any drug in them. So in some ways, it's not irrational to believe that costly Nurofen is more effective than cheap unbranded ibuprofen, even if they've both got the same active ingredient.

· Pseudoscientists, and alternative therapists, being expensive and long-winded, have more time to weave ceremony and cultural meaning, and maximise their placebo effect, than a rushed NHS GP. It's placebo, but that doesn't mean it's useless. If you wanted to maximise everyone's health, then doctors would confidently lie to their patients about effectiveness of treatments, the way they did before we began championing choice and informed consent over efficacy; and people like me would stop debunking placebo alternative therapies. No chance.

Please send your bad science to [email protected]


We really have to find some way to harness the placebo effect but that puts us in an interesting "Catch 22" situation because, as someone once said, 'You gotta have faith' ;) So an Institute of Placebo Medicine wouldn't work.

Its interesting what the article says about ritual and magic and while quite a few modern medicines work quite a lot of the small sclae interactions with our GPs seems little more than a visit to the local shaman or priest. What I'd really be interested to know is if it can be strong enough to actually perform miracles.

I was going to rattle on here about the complicated nature of the placebo effect but found that I'd already done that!

However, I'm interested in the 'shaman' effect.
Recently a woman in my area was expecting twins who were expected to die before birth because of a rare 'twin type' disorder (one twin was diverting all the nutrients to itself or something).

Someone came along and prayed with the family in the name of a saint and the babies were mysteriously healed and born normally and healthy.

I can produce a newspaper cutting if anyone's interested.

So, if prayer etc are the shaman effect at work, how can that process work on unborn children?
escargot said:
However, I'm interested in the 'shaman' effect.
Recently a woman in my area was expecting twins who were expected to die before birth because of a rare 'twin type' disorder (one twin was diverting all the nutrients to itself or something).

Someone came along and prayed with the family in the name of a saint and the babies were mysteriously healed and born normally and healthy.

I can produce a newspaper cutting if anyone's interested.

So, if prayer etc are the shaman effect at work, how can that process work on unborn children?

Very interesting and I would be interested in the article (it might give us a hook for finding any follow up studies).

It would depend on the nature of the condition but it does sound like some kind of parasitic twin (see the teratomas and included/parasitic twin thread for more info) which was shown in a recent documentary which is very serious and rather tricky to misdiagnose but that plus media spin are certainly factors that need to be considered. It also might be that in some cases the parasitic twin is viable and it is only tapping into the others blood stream (which would be interesting in itself) but if not then it would be a startling (even miraculous) recovery from what would be normally a situation involving at least one (and potentially three) deaths. So any more details would be much appreciated.

I have watched dental extractions performed under hypnosis, and not only was there no pain, there was hardly any bleeding and nothing but a slight 'soreness' afterwards.
When you hypnotise someone, they, in fact, hypnotise themselves, and having convinced themselves that it works go on to convince themselves that any further suggestions such as pain suppression will work too. Consequently, you can hypnotise yourself, and act on autosuggestions. Surely this is the same mechanism as the placebo effect?
And to carry it further, maybe that effect can influence cell growth and replacement, facilitating 'miracle' cures. Sadly, hypnosis will not work on everyone, just as faith type cures do not work on every induvidual. I'd would suggest that in cases where placebo effect , and miracle cures are most effective then the subjects are likety to be be 'good' hypnotic subjects.
brian ellwood: Its an interesting hypothesis - well worth testing (I do wonder if science tends to rather give up when they find the placebo effect rtaher than using it as a launchpad to something interesting).

Could this be extended to alternative medicine e.g. homeopathy or acupuncture? I saw a documentary where they unseamed a man from nave to chops and he was sitting there with a few needles in his upper body chatting away with just a screen between him and rather a bloody operation. It might explain why science has struggled to understand such things - I've always taken an 'if it works.....' approach and rather dodged the issue but.....

It is possible that placebo research hasn't been ambitious enough and that it might be a window into our own 'superhuman' abilities (which may also include things like those, mythical?, bursts of strength and certainly includes those extreme survival abilities like loosing a limb but still being able to fish it out of a threshing machine and wandering off to find help).

mejane said:
On a related note, if people can "think themselves better" then can they also "think themselves ill"? Do hypocondriacs really get more/worse illnesses?

If that doesn't make sense, just ask Bill what my real question is :D


People who are given a placebo in clinical trials will often report getting the side effects that may have occured had they been given the real pill. I've always found that rather amusing.:)
Yes, people can "think" themselves ill. Just look at cases of psychosomatic illnesses.
Placebo: Cracking the Code

I just watched "Placebo: Cracking the Code" on Discovery and it was very interesting.

Nick Humphrey from the Evolutionary Psychology department at the LSE brought otgether a team to try and crack the problem (or at least work out avenues for future study.

The other members of the team included

Anne Harrington - prof of History of Sceince at Harvard

Howard Fields - a neuroscientist from The University of california, San Francisco

Dan Moerman - an anthroplogist from The University of Michigan-Dearborn (who seemed like a lot of fun).

Fabrizio Benedetti - a doctor/researcher at the University of Turin

And they kicked around ideas and the documentary was hung off their work and discussion.

Interesting bits (which touched on a lot of areas of Fortean interest) include:

Charles II curing 100,000 people with a laying on of hands.

Irving Kirsch's use of the FIA to get his hands on anti-depression drug trials which showed the drugs were not much more effective than the placebo. Which in 2002 led to a lot of "The drugs don't work" stlye headlines although he is trying to point out that a better conclusion is that the mind is nearly a better tool for this kind of treatment.

Janice Schonfeld's case where she was part of a test of anti-depression medicine which also monitored brain activity and she was so rapidly cured she knew she wasn't on the placebo but of course she was. The studies showed increased activity in the mood latering part so the brain.

Michael Dixon a GP who has written a book on the Placebo Effect (' The Human Effect in Medicine') and uses a healer in his practice to try to harness the effect.

The nocebo effect - which can harm (a Birminghma study found 1/3 of people on chemo placebos suffered the expected side effects) and even, as in the case of Sam P. Londe, kill (his doctor was Clifton Neador).

Albert Mason who thought he had used hypnotism to cure a young man with warts but was informed it was incurable congenital ichthyosis. It made a big splash when it was publsihed in the BMJ in 1952 but he was never able to repeat the cure - he thinks it is because he now believes it is incurable.

Jean Piere Bely who was cured of MS 15 years ago at lourdes - he became the 66th officially recognised miracle sicne 1882.

Red placebos are better than blue, large ones are better than small ones and more are more effective than less.

Only a third of people seem repsonsive to it (or it only works a third of the time - I'm unsure if theyve done multiple testing on the same person). Fabrizio Benedetti reckons he can get that number up to 90% by 'training' the patient to expcet to be healed.

Anyway the Discovery Channel page on this:

Placebo: Cracking the Code

The placebo is one of the biggest mysteries of medicine. This documentary explores the truth behind 'the lie that heals' and asks whether placebo can do more harm than good.

Powdered mummy, crab's eyes, viper's flesh and pearls. These are some Medieval medicinal cures and few would question the fact that they worked as a placebo.

But how many drugs in modern science might also be ineffective? The US Office of Technology Assessment estimates that only one fifth of modern medical treatments in common use have been proved effective. Modern placebo effects include prescribing antibiotics for colds and flu, evidence that placebo use still has a place in modern medicine.

A study by Baylor College of Medicine in the States, found that surgery carried out on people with osteoarthritis of the knee produced no better results than in dummy operations. Patients who underwent placebo surgery were as likely to report pain relief as those who underwent the real thing.

Recently, the reputation of placebos as a deceitful fraud has undergone a dramatic change. To alternative medical practitioners, the placebo response represents the self-healing forces produced by the mind-body connection.

The placebo effect is the perceived improvement in health not attributable to treatment. A placebo is a medication believed by the administrator to be harmless.

The word placebo comes from the Latin, meaning 'I will please', and by the 19th century, it was a medicine given 'more to please than to benefit the patient', according to A.K. Shapiro, author of The Powerful Placebo: From Ancient Priest to Modern Physician.

Although it's not fully understood how placebos works, three major reasons may explain the effect. For many patients, taking a placebo causes the release of endorphins, opiate-like substances naturally produced by the brain in times of stress.

There's also the conditioning response to medical treatment, the belief that a drug administered by a doctor will do you good. Irving Kirsch, a psychologist at the University of Conneticut, believes the effectiveness of drugs such as Prozac may be attributed to the placebo effect. He analysed 19 clinical trials of antidepressants and concluded that the expectation of improvement accounted for 75% of the drugs' effectiveness.

And lastly, the expectancy that taking a drug will have a powerful effect on your body. In one study, subjects given sugar water were told that it was an emetic. Over 80% of patients in the study responded by vomiting. As Asbjorn Hrobartsson, author of the study the uncontrollable placebo effect puts it, "Any therapeutic meeting between a conscious patient and a doctor has the potential of initiating a placebo effect."

Many thousands of patients report relief from pain while taking placebos. Just because there is no medical explanation for this, is their recovery any less real?

The debate for and against placebos rages on. For many, it's a question of asking what difference does it make why something works, as long as it seems to work.

Dr Richard Tonkin of the Research Council for Complementary Medicine says, "The problem with the placebo effect is that it is regarded by most people as a nuisance or fake. But it isn't. It is a practical and positive effect that acts by catalysing the self-healing mechanisms within a patient." Watch the Placebo: Cracking the Code and see which placebo theory you swallow!

Arguments for

Mind over matter. The body has powerful, natural recuperative abilities and a placebo could help facilitate this.

Acts as a psychological boost, that a person's positive attitude may be important in recovery from illness.

The placebo effect may be a measurement of changed behaviour affected by a belief in the treatment.

Arguments against

Placebo-related changes could be over-estimated because:

With chronic pain conditions or mood disorders, patients may show spontaneous improvement.

Placebo effects can result from contact with doctors, perhaps a diagnosis or simple attention from a respected professional alleviates anxiety.

The patient, in order to please the doctor, might report benefit when no benefit has occurred. In other words, the 'politeness effect'.


An interesting Daily Telegraph article which also touches on pseudo surgery (refered to as 'sham surgery' in the Discovery doc):


I think an interesting point that came up was the image problem and Anne Harrington mentioned that it might be able to redefine it as part of a mind/body healing technique and I think this is avery interesting point. The placebo effect has been thought of as the ground level to measure how a drug would effect us but as we can actually make ourselves better it is no such thing - more a comparison of two different healing techniques. To get a properly base level to measure us from we'd need to have the drugs adminsitered without us realising and that isn't going to happen.

Perhaps it is time for a rebranding to start afresh with this kind of thing and as the US government have earmarked $3million for placebo research it might be interesting times ahead.

Perhaps it only works in a 3rd of people because that is the likely percentage of highly suggestible people.

The word "suggestible" always makes the person sound like a moron but it's not necessarily true. I'm very prone to suggestion which makes self-hypnosis very easy and useful. I didn't realize how often I used self-hypnosis until I began meditating recently and found that was the exact feeling I had in school while taking tests. I would answer all the easy questions first, then make myself relax until a certain feeling overcame me. In that state I could remember lectures word for word and pages from books. I also used the technique to get rid of head-aches if there were no aspirin around.
Perhaps it only works in a 3rd of people because that is the likely percentage of highly suggestible people.

Exactly,Tulip. From experience I've realised that around a third of subjects are in the 'sonambulist' category, that is, they can be deeply hypnotised very quickly, will respond to an amazing range of suggestions including anaesthesia, catalepsy etc. and will remember nothing when they awake.They will also follow post-hypnotic instructions which must be a nescessary condition for a placebo effect to work. I have found another third can be 'influenced' lightly and the depth of hypnosis improved by regular sessions, ie.training. They will usually remember everything that has happened during a session. They report a pleasant experience in which most say they felt they could 'wake up' if they wanted, but found it "far too much effort to be bothered". Many women report positive effects at childbirth when this is used as a technique at prenatal classes.
The remainder of the population do not respond to positively to hypnosis, but may still be mildly susceptible to suggestion.
Just look at charismatic religious and political leaders, advertising etc. to see how effective it all is!
Many years ago, before the dawn of mankind, I did a study on the placebo effect for school.

Ever since that day, I lost trust in doctors. And the only thing I think of when I see medicines are "placebo".

Two things have come out of it.

1. I don't fall sick very often like I used to. If you simply feel that you are alright, you will be alright.

2. Medicines don't work when I do fall sick. :(

Ugh. :mad:

I didn't go through the links, but there was an interesting story regarding placebos.

A patient was suffering from brain tumors. He heard that there was a new medicine called Krebiozen, which was effective on horses. He asked his doctor for it, but the doctor didn't have any krebiozen. The doctor instead, injected him with water.

Within a couple days, the tumors had melted away.

A couple days later, the guy read that Krebiozen was a failure.

Next day, the tumors were back.

:cross eye
Mystery pain 'all in the mind'

full story here

Mysterious pain, such as lower back pain, may originate in the brain rather than the body, according to a study.
Scientists from University College London and the University of Pittsburgh carried out tests on eight people.

Some were hypnotised and told they were in pain. Others were subjected to physical pain. Scans showed that both experienced similar brain activity.

The researchers said the findings, published in NeuroImage, suggested that pain can sometimes begin in the brain.

"The fact that hypnosis was able to induce a genuine painful experience suggests that some pain really can begin in our minds," said Dr David Oakley, director of the hypnosis unit at UCL.



some people realy do get money for old rope...:rolleyes:
I merged in the short thread above as it showed a kind of reverse placebo effect which may well be scientific evidence for things like curses, etc.
Israeli Doctors Often Give Fake Pills to Patients

Fri Sep 17, 2004 05:28 AM ET

NEW YORK (Reuters Health) - New research from a survey conducted in Israel suggests that doctors often give inactive "placebo" pills to their patients and, in most cases, tell them they're getting a real drug.

The deception involved in giving a placebo raises ethical concerns, the authors note.

According to the report in the British Medical Journal, 60 percent of healthcare providers surveyed reported giving placebos to their patients.

"Some have advocated banning the clinical use of placebos because of the deception involved in administration and the possible harm to the doctor-patient relationship," Dr. Pesach Lichtenberg and Uriel Nitzan, from Herzog Hospital in Jerusalem, note.

The new findings indicate that despite these concerns, the practice continues, they add.

Of the 110 physicians and nurses who were approached to participate in the study, 89 completed the survey: 31 hospital-based doctors, 31 head nurses, and 27 family practice doctors from community clinics.

Among the respondents who said they did use placebos, 68 percent misled their patient about the drug they were given.

Also, 62 percent described using placebos at least once a month, and 28 percent regarded placebos as a way to help reach a diagnosis -- in other words, to see if the patient had a "real" illness or not.

In addition, 94 percent of placebo users found the inactive agents to be generally or occasionally effective.

"Clearly, wider recognition of (placebo use), and debate about its implications, are needed," the investigators state. "Further investigations into the extent and nature of use should be conducted, particularly in a clinical context where the placebo's effects may differ from that found in randomized controlled trials."

SOURCE: British Medical Journal, September 16th online issue, 2004.

© Reuters 2004. All Rights Reserved.

Oregon scientists study impact of belief on getting well

2/6/2005, 11:37 p.m. PT
The Associated Press

PORTLAND, Ore. (AP) — Harvard anesthesiologist Dr. Henry Beecher coined the phrase the "placebo effect," after he discovered that injections of a saline solution helped wounded soldiers in World War II overcome pain once the morphine ran out.

Now, a group of Oregon scientists are following in Beecher's footsteps as they begin mapping what they call the "expectancy effect" — the impact on health of a person's expectations of getting well.

Oregon Health & Science University School of Medicine has been awarded a three-year, $2.4 million grant from the National Center for Complementary and Alternative Medicine to study the phenomenon.

Just as Beecher discovered that up to 35 percent of a therapeutic response to any medical treatment could be the result of belief, Oregon researchers hope to show that a patient's expectation of getting well can have a positive affect on their health outcome.

The study's goal is to develop models that can be used to study cognitive and physiological changes that contribute to the expectancy effect.

The models could range from "perceived self-efficacy" — the belief that a person can influence his or her own health — to hormonal activity and genetic changes that effect the brain's neurotransmitter systems, said Dr. Barry Oken, director of the Oregon Center for Complementary and Alternative Medicine in Neurological Disorders at OHSU.

The researchers also hope to improve the design of clinical trials by learning how individual differences contribute to changes in their responses to medical treatments.

The expectancy effect is related to the placebo effect, in which a patient reports a positive response to an inactive medical treatment, such as a sugar pill, as if it were an active medical treatment.

But the expectancy effect is broader than the placebo effect and includes all processes and influences that may affect the brain's anticipation of a response.

"We're not talking about patient-physician interaction, which, to some people, is considered part of placebo effect — the contact, the handholding, the bedside manners," Oken said. "We're really thinking about people's hope or expectation that they're going to get better."

For example, one recent study showed that Parkinson's disease patients who were administered a placebo experienced changes in brain chemistry similar to those caused by symptom-treating drugs levodopa and apomorphine.

Running on faith


trust this is the right place for this

Running on faith

Spiritual healing can appear to have a positive effect, but when placebo reigns over rationality, I'm wary

Edzard Ernst
Tuesday February 15, 2005
The Guardian

Around 15,000 spiritual healers practise in Britain, so it is important to examine whether their techniques do more good than harm. The term covers a great many approaches: reiki, johrei, therapeutic touch, intercessory prayer, faith healing and distant healing all fall under the umbrella of spiritual healing. The common denominator is that healers of all types claim to somehow channel "healing energy" into the patient's body with the intention of improving health. The concept is as simple as it is unproven - nobody, for instance, has so far been able to measure the "energy" that these healers are talking about.

Yet healers have no doubts about their methods, and there is no shortage of satisfied customers - including celebrities willing to pay handsome hourly rates for energy-channelling consultations.

So what reliable evidence is there? Dozens of clinical trials have been published, but those that appear methodogically sound don't usually show much. Take, for example, the recent trial with 799 patients of a US coronary care unit. Half of them received "intercessory prayer" for 26 weeks; basically that means a group of healers regularly prayed for these patients at remote locations. The other half received no such interventions in addition to their usual care.

At the end of this period, the number of deaths, heart attacks and other serious complications were similar in both groups.

Now take a headline-making study published in the US-based Journal of Reproductive Medicine which apparently showed that distant healing dramatically increased the success of in-vitro fertilisation. The expert named as "lead author" later distanced himself from the study, stating that he edited the manuscript but was not involved in the trial. The primary organiser, Daniel Wirth, who has published numerous other healing studies, has subsequently been jailed on several counts of fraud, unrelated to this study.

My research team conducted a trial of about 110 chronic pain patients. We were keen to accommodate the ideas of all five participating healers and wanted to find out whether spiritual healing is more than a placebo effect. Patients therefore received eight healing sessions either by five healers - or by five actors. The actors had been coached by the healers in how to pretend to be healers. In a separate experiment, healers were hidden behind a one-way mirror while, in the control group, there was no one behind the mirror.

The results were staggering. Improvements were so remarkable that several patients practically abandoned their wheelchairs during the study. But there were no differences between the groups. If anything, the control patients fared slightly better than those receiving "real" healing. Even the often-quoted tingling sensation and feeling of warmth during healing sessions were also experienced by patients who received no healing at all. Results such as these strongly suggest that spiritual healing is a powerful placebo, but not much more. This is also the conclusion drawn from combining the findings of rigorous trials conducted in this area.

Many people believe, however, that it is irrelevant whether the effects are "real" or due to the placebo effect. Professor Jonathan Waxman, an oncologist at Imperial College London, was recently quoted as sharing this view on healing: "I've seen my patients look and feel better as a result. Who cares how it works, as long as it does?" I don't subscribe to this line of thinking, for several reasons.

The placebo effect on which healing relies is one which most other medical treatments generate as well - so we don't need an ineffective therapy in order to profit from this effect. Placebo effects rely on factors such as empathy, time, understanding and the warmth of a therapeutic relationship. It would be a serious mistake if doctors delegated these core qualities of medicine to healers. Why pay £100 a session for placebo if it comes as a free bonus with most effective therapies?

There is some evidence that healing can harm certain patients; some mental conditions, for instance, can get worse. There is also the possibility that if patients feel encouraged to place credence in healers, some may give up vital treatments in favour of healing.

If we allow mystical conjecture to infiltrate our thinking, we are in danger of abandoning rationality in favour of superstition. It is easy to see how this could rebound on us.

I think the claims of healers should be taken with a pinch of salt. In the absence of convincing evidence, you might as well spend your money on something that demonstrably works.

· Edzard Ernst is professor of complementary medicine at the Peninsula medicine school at the universities of Exeter and Plymouth.