Should doctors prescribe placebos?

It’s the New Year so I’m in philosophical mood. No, not really. But my blog today has been triggered by a piece a I read yesterday about the use of placebos in medicine, which got me thinking in a somewhat philosophical vein.

In a study published in the Journal of General Internal Medicine, researchers at the University of Chicago polled 466 hospital doctors about their practice, of whom 231 responded. Of this 231, almost half (45 per cent, actually) said they had used a placebo during their clinical practice.

Of doctors who reported using a placebo at some point, about a third said they told the patient the substance was something that “may help and will not hurt.” Another third gave other information to patients including, “this may help you but I am not sure how it works.” 19 percent said it was a “medication,” and 9 percent called it “a medicine with no specific effect.” Only 4 percent of the doctors came right out and called it a placebo.”

My first reaction to reading a report of this piece of research was surprise: I, rather naively perhaps, didn’t even know that placebos were available for prescription. I honestly wouldn’t know where to get my hands on some myself. I have no idea, for instance, what I would write on the prescription.

This aside, the research, I think, raises some important ethical questions. First of all, if you’re going to give a placebo, then one might argue that being covert about it by not declaring it is a placebo is deceiving a patient. One could argue that if you have your patients’ best interests at heart, then this does not necessarily matter. But giving a placebo and not explicitly stating that is what is being given is a deception, whichever way you cut it.

A small percentage of doctors chose a more candid approach, it seems, and declared their chosen treatment to be a placebo�. Some would argue that while this is transparent, it is self-defeating in that for a placebo to ‘work’, individuals need to believe it might or will work.

So, one option available to doctors is to deceive their patients into believing that the placebo is something other than a placebo, or the other option is to jeopardise the value of the placebo by being transparent about it. I don’t think there are any easy answers here.

And it gets even more vexed when we consider other questions such as: should treatments that have been shown to be no more effective than placebos be advised by doctors and made available to patients? The knee-jerk response to this question may be ‘no’, but consider this hypothetical but not far from reality scenario for a moment.

Imagine a situation where a new drug is developed for a rather nasty and debilitating condition for which there is no existing treatment. Let’s imagine now that this drug is tested in individuals with this condition, in ‘double-blind placebo-controlled’ fashion. Now, lets imagine that half of the people treated with the active drug see significant and meaningful benefit with little in the way of side-effects. Some would see that as a good result, which suggests that the treatment has merit. But now imagine that the results of this experiment show that those taking the placebo did just as well. Suddenly, for many in the fields of science and medicine the drug looks worthless.

Now think about this from a patient’s perspective. Remember that this is a nasty and debilitating condition for which no other treatment has been found. Taking the drug gives you about a 50 per cent chance of significant relief, but this option is denied to you because the drug performed no better than placebo in trials. You could imagine that as a patient you might not mind about the fact that the drug performed no better than placebo, if you thought that you had a 50 per cent chance of benefiting significantly from that treatment.

I have this week experienced this conundrum from a ‘patient’s’ perspective in my own life. On 2nd January I woke up with a cough and cold. My girlfriend very kindly offered to make me a honey and lemon drink. I gratefully accepted her offer, drunk the drink, as well as a couple more throughout the morning. I believe I felt better for my honey and lemon ‘therapy’.

The reason that I mention this is that about a month ago I wrote about a study which showed that honey had soothed children’s coughing at night. The honey found to be about as effective as a cough-suppressant drug. However, in some other experiments, this drug has not been found to be any more effective than placebo. So some would argue (and indeed did) that this provides good evidence that honey is no better than placebo for childhood coughing.

However, this study did find that giving honey was significantly better than doing nothing. So, now imagine that you’re a parent, your child has a cough, and you know that a study has found that giving honey is quite likely to relieve it. But you also know that the honey does not perform any better than placebo. Bearing in mind the fact that toxic side-effects from honey are low, would you give it? I suspect most parents would. If I were a parent, I think I would. And, as I’ve already admitted, I would not hesitate to down a honey-containing drink myself, working on the assumption that it may well make me feel better.

Now, the decision about giving kids honey is not as charged, perhaps, as one about giving people drugs or treatments that might be expensive and potentially hazardous that have been shown to be no better than placebo. But, at their essence, these situations present us with the same conundrum. And, quite frankly, I don’t know what the answer to latter one is.

What I do know is that the way in which treatments are assessed is, in the case of double-blind placebo controlled studies, quite an academic pursuit. Nothing wrong with that, of course, except that in my experience, sick individuals are generally less concerned with science, and more concerned with getting well.

The reality is, that there really is very little evidence for most of medical practice, conventional or otherwise, anyway. As a practitioner, I am well aware that a lot the approaches I suggest in practice have no ‘evidence-base’. They may have some logic and reasoning to them, and I may even find them to be generally effective in the real world, but practically nothing of what I do has been subjected to systematic study. So, I have no idea whether when someone feels better or benefits in some other way for taking the advice I have given them whether this is as a result of the action they took, the placebo response, or both. Oh, and even if something has been studied and found to be better than placebo, I still have no idea whether the benefits derived by a patient have come from any real effect, or the placebo response, or both. Do I care? No, is the honest answer. And neither, generally speaking, do my patients.

The ‘placebo response’ has, for a long time, been a bit of a dirty term in scientific circles. My suspicion is that most academics would regard the concept of advising treatments that have not been shown to be better than placebo to be wrong. I don’t expect they’d be thrilled about the idea of doctors prescribing placebos either. Yet, the recent Journal of General Internal Medicine study seems to show that many doctors have prescribed them. Maybe, that’s because doctors are primarily concerned with helping people get better, and unlike many academics, care less exactly how they get better?

8 Responses to Should doctors prescribe placebos?

  1. Bill Cockerill 4 January 2008 at 3:54 pm #

    A refreshing view on the power of placebo John.

    I read this book a few years ago about Placebo that is very readable (i.e. science for the layman).

    As well as medicine (real or placebo) health might be improved by a caring consultation which has a placebo effect of its own. This is something Richard Dawkins suggested in his Channel 4 series Enemies of Reason when he looked at homoeopathy (which is of course totally unproven) but is likely to help patients from the placebo effect:

    “The fact that homeopathic doctors and patients do claim there is a benefit he [Dawkins] puts down to the human body’s power to restore itself when given the psychological boost of someone else’s concentrated concern and attention: the average half hour to an hour, rather than the typical eight-minute NHS GP consultation. “There was a time when old-fashioned family doctors used to hand out placebos but now they aren’t allowed to because it’s against medical ethics. Now it’s only the homeopaths who are allowed to benefit from the placebo effect.”

  2. Kevin eakins 4 January 2008 at 6:00 pm #

    Over the past 35 to 40 there has been some scientific research being undertaken by Professor William Tiller into the effects of human intention on animate and inanimate systems. His most recent book, Some Science Adventures with Real Magic, summarises his work and his theories offer a scientific persepctive for the placebo effect as well as other psycho-energetic phenomena. This may not yet be mainstream science but the results of his research seem undeniable. When we say there is no scientific basis for the placebo effect, what we mean is that the current, popularly accepted scientific beliefs (many of them implicity based on Newtonian physics) do not offer an explanantion. The word, placebo, itself is very reductionist. It reduces a very real and obviously repeatable interaction into some sort of con trick conjuring up images of poor patients being fooled into believing that they are getting better. A more useful approach is to expand our current scientific paradigm to include what does not fit our current picture rather than ignoring, reducing, and demonising it. Maybe then we could begin to actually use the “placebo” effect more efficiently in clinical practise and we would not practise “placebo” covertly but instead work openly and with confidence. I am sure we will be able to think of a better description as well.

  3. Bill Cockerill 5 January 2008 at 11:03 am #

    William Tiller believes in psychoenergetics. In Tiller’s book he discusses remote viewing, psychokinesis, auras, clairvoyance, telepathy, levitation and homoeopathy. None of these have been proven in controlled experimental conditions.
    James Randi offers a $1m prize to anyone who can demonstrate events such as remote viewing under proper observing conditions. It has not been claimed so far.

    The mechanisms through which placebo can work well may be related to inflammation as John suggests or through other chemical or hormonal changes. Real science is much more interesting than paranormal explanations. If Tiller really has discovered new phenomena then they should be repeatable, observable and he should claim a Nobel prize.

  4. rm 6 January 2008 at 12:45 am #

    What should we do with a person receiving a placebo treatment if they get worse? Blame the Doctor, the medicine or the patient? Perhaps we should find the patient guilty of improperly practicing medical placebo. If a patient expects a certain result and they get it, but suffer harm from the expected results what do we do? How could we investigate? How do you do research on placebo’s. Would any of them work better than a placebo? Will we accept “Low Expectation Medicine” as legitimate attempt at a cure.
    Sticky subject, yes.

  5. imh 7 January 2008 at 10:20 pm #

    Interesting article but I think your critique of the ‘double-blind placebo-controlled’ study is flawed.

    You suppose that “half of the people treated with the active drug see significant and meaningful benefit” and “that those taking the placebo did just as well”.
    Now assuming “just as well” implies that once again half of this group see benefit.
    Well these two premises together essentially mean that the actual drug has no effect at all, absolutley zero in fact as all positive effects can be explained simply by taking any of the pills be it placebo or the real one.

    Apologies if this is overly pedantic, but interpreting statistics can be non-intuitive

  6. Hilda 7 January 2008 at 10:29 pm #

    Hi, I think it is unethical to deceive or indeed treat the patient as a child. However, the psychological aspect of care is very important. One of my son’s friends is training to be a doctor and I said to him that I thought he would make a very ‘nice ‘doctor. He replied that he hopes he will be a ‘good’ doctor. I said that being nice is just as important as it is all very well to be scientific but patients respond to and are affected by kindness, friendliness and most of all reassurance. If you have a good experience in a consultation you can almost feel the therapeutic effect. THis is also a placebo but a good one. Hilda

  7. Dr John Briffa 7 January 2008 at 10:52 pm #

    The point here is that the giving and taking of a drug can have placebo effects that are REAL (even though they may not come from, say, the chemical itself), and that the benefits derived from this effect can be worthwhile (from a patient’s perspective particularly).

  8. J Dean 10 January 2009 at 10:41 pm #

    I am currently in a situation where my elderly father in law has been complaining about pains in his head for several years. The doctors have never been able to find anything wrong (MRI’s, Xrays, etc.) He has overdosed on sleeping pills several times when he had control of his own medicine. He is currently living with us now, and will constantly compain and try to get us to give him more tylenol even though he has taken the max suggested. Last night I noticed that right after I gave him a pill (around 30 seconds) and again today he said that he felt much better and the pain was gone. We are going to be taking him to see a Psychiatrist this week, but I am thinking a Placebo would be something very usefull here. Medicine cant work that fast so, now I am thinking it must be a mental thing. In this case the act of taking the pills seems to make a difference and I would think it is ethical if It helps until a cure could be found. Just my 2 cents.

Leave a Reply