Placebo, The Lies That Heal
By Peter French
Medical Ethics and Placebo
Double Blind Trials
Holes In Some Trials
Basic Research Findings For Placebo
Placebo Delivery Mechanism
Colour Has An Effect
Effectiveness Of Placebo
Doctor Patient Relationship
Athletic Performance enhancement
Limitations Of Placebo
The Placebo Effect in the Brain
Where do we go from here?
a substance or procedure that has no therapeutic effect, generally used as a control in testing new drugs.
a beneficial effect produced by a placebo drug or treatment, which cannot be attributed to the properties of the placebo itself, and must therefore be due to the patient’s belief in that treatment coupled to the natural ability of the patient to heal themselves.
a detrimental effect on health produced by psychological or psychosomatic factors such as negative expectations of treatment or prognosis.
You have probably heard about the placebo effect.
Depending on your background and the amount of reading about the placebo effect that you have done, then you will either see it as a slur on logical approaches to healing or a boon that is available for free to enhance the treatment of many patients.
At one time patients were bled to heal them, had leaches applied for many ailments and received tonics that had no medical efficacy. Oddly the patients appeared to get better more often than not, the placebo effect went hand in hand with what patients expected might help them and the confidence that they had with their physician.
Such remedies were rarely effective against serious disease, but were effective against many maladies and in particular they were often effective in relieving symptoms.
The development of what we now see as real medicines changed many things.
Treatments such as:
- Using the use of cowpox (a non-lethal disease) to induce an immune response that would protect against (the lethal disease) smallpox.
- The development of a mould called penicillin that was found to spoil cultures of bacteria, in to a drug to combat bacterial infections.
- Many recently developed antibiotics.
These have all caused a change in attitude to medicine. These new drugs and treatments worked against the infection and not just the symptoms, they truly extinguished infection regardless of a patient’s confidence in the physician. Interestingly it brought about increased levels of confidence in physicians and has actually raised doctors to new status within communities.
This meant that medical treatments no longer relied on the rapport skills and encouragement from the physician, and for a long time it wasn’t noticed.
Some researchers like Irving Kirsch, Ted Kaptchuk and others have spent a large part of their careers researching the placebo effect. To some extent they have rediscovered a healing art that was lost in the face of the new era of highly efficacious drugs.
The primary directive of medical professionals is essentially “do no harm”.
In medical practice the use of placebo is a severe ethical problem because it involves offering a treatment that has no obvious efficacy to a patient presenting what has to be treated as a real issue.
In times not far past it was not uncommon for a doctor to prescribe a tonic, or vitamins which they knew had no issue of harm but equally they knew it would have no medical beneficial effect on the issue other than what we now call placebo. So there is precedent for the use of placebos in this role.
Along similar lines, many cases of pharyngitis (a sore or scratchy throat) drive many people to seek help from their doctor (ref 13, 14). The statistics show that most instances that present as pharyngitis are viral in nature, but often the patient is prescribed antibiotics which are only effective against bacterial based infection. The generally accepted maximum improvement in recovery time is 1 day. It can be argued in some cases that the antibiotics are part of a prophylactic approach but such cases are rare. This could be argued as a modern precedent for doctors offering treatment that will have little or no medical effect on an illness. In fact the over prescription of antibiotics can have a detrimental effect on the gut of the patient since commonly prescribed antibiotics generally target bacteria non-selectively. As a general approach it also contributes to the increasing antibiotic resistance problem.
In double blind surgical procedure trials where a placebo surgical procedure is required, the placebo involves administration of anaesthetic and creation of real wounds and suturing to simulate surgery, and this is generally not acceptable from an ethical point of view because of the harm that is being done without the inclusion of a potentially useful procedure. In the case of surgical placebo trials, great debate is usually required about the ethics before patients are subjected to placebo surgery as part of trials.
The placebo effect has been known about and has been recognised as a contributory factor in recovery from illness for some time.
The placebo effect has been demonised, mainly by drug companies because it was discovered that many people have the innate ability to heal themselves. This means that the drug companies now have to prove that any beneficial effect in a trial of a new treatment or drug is due to the efficacy of the drug under test alone.
When a new medicine is designed, developed and produced it is subjected to extensive double blind trials.
A double blind trial is a trial where people requiring a treatment volunteer to be part of the trial of the new treatment. The volunteers will be randomly selected to receive the new treatment or a placebo. Generally the trial is arranged so that the people delivering the treatment will not know whether they are supplying the medicine or the placebo to any patient that they deal with. That way they deal in the same way with each person that they interact with. Essentially that means that any rapport effects between physician and patient are the same for the real treatment and the placebo.
When the trial is completed the statistical analysis of the effect of the drug has to be shown to be significantly more efficacious than the placebo.
That at least is the spirit of the technique.
An article in Nature reported that it is getting harder to get painkillers through clinical trials. Analysis of trial data has shown that it is getting more difficult to prove a drug’s advantage over placebo. This apparent increase in placebo effectiveness appeared to be a USA development (ref 1).
In testing new drugs it appears that the placebo effect is actually getting stronger. Stronger placebo responses have been reported for trials of antidepressants and anti-psychotics (ref 2,3).
In 2008 when Dr Irving Kirsch started research in to the efficacy of anti-depressants he found a few interesting things about the true nature of the rules (In the USA).
He discovered that not all of the trial results had to be published as proof of efficacy to the authorities. In order to perform what was one of the first meta-studies of existing study results he needed access to all the drug trial results for several anti-depressant drugs. He got access to the results he needed using the freedom of information act and ran a meta-analysis on several drug trials.
A meta-analysis is where the results of several similar trials are brought together to statistically check the significance of certain findings.
Some short falls with the methods were highlighted.
The rules stated that only two sets of trial results had to be submitted to scrutiny, this meant that many actual trials could be run until they found two that were favourable. Ignoring the trials where the statistical significance of the results against placebo was insufficient meant that the statistical outcomes were skewed in favour of the drug under trial.
Another short fall in the statistics was the measurement of statistical significance. Where a trial shows only a small improvement over the performance of the placebo the statistical significance can be enhanced by having a bigger sample. For example comparing a trial of 50 people with the same level of performance over placebo as a trial of 5000, the significance is magnified for the bigger trial.
Another factor (particularly in the case of drugs with side effects) that skewed results were people who had taken a similar drug before as opposed to those who had never had similar medication. Placebos don’t have side effects, so ‘experienced’ patients can tell the difference and so can their physicians. And the knowledge of receiving the ‘real’ drug or the placebo has a measurable effect on the results.
The results from Kirsch’s work on anti-depressants were startling and caused controversy. The drugs being tested were shown to perform no better than placebo at treating depression. An interesting statistical outcome was that several complementary therapies including acupuncture actually performed better than the drugs being tested, for helping with depression.
In terms of depression, the ineffective nature of the drugs has had a knock on effect. The drugs were developed in support of the “brain imbalance” theory of depression. In brief, the brain imbalance theory states that depression is caused by an imbalance of chemicals in the brain. So the Kirsch results actually call in to question the underlying theory of imbalance in the depressed brain because the same effect was achieved by placebos. That alone has put more emphasis on other theories of depression.
The bottom line here is that there are areas of medicine where practitioners may have been unwittingly prescribing medication that had no obvious therapeutic value for some time. And in many cases they have recovered from their illness.
Remember that the placebo effect is a self-healing effect that is stimulated or triggered in some way.
We as humans tend to be predictable and the following has been found to be true:
- A capsule invokes a stronger placebo effect than a pill.
- A large capsule has more of a placebo effect than a small capsule.
- An injection has more effect than a large capsule.
- Placebo surgical procedure is the strongest of all.
Within the realm of capsules and tablets it has been found that colour has an effect (ref 4):
- Blue and green capsules and pills work best for calming or tranquilising effects like helping anxiety.
- Red, yellow and orange capsules and pills work best for stimulating effects like anti-depressants.
- Red pills are more effective at treating pain
- Green pills have been shown as more effective for phobia medication
Although the blue pills are more effective for treating anxiety in general, an exception was found with Italian males – blue is the shirt colour for their national football team and invokes a stimulant effect.
When it comes to drugs, colour was seen to be significant but not consistent, but drugs have active ingredients too.
The relationship between the practitioner and patient is important in furthering the effect of treatment (Ref 6).
The reliance on drugs that ‘just work’ has led to many practitioners who have a poor ‘bedside manner’ or poor rapport skills with patients.
Studies have shown that the ability of the practitioner to connect with and raise levels of enthusiasm in the patient is much more important than was previously thought. That ability to prime the patient for recovery combined with the respected position of the ‘expert’ can improve a patients’ confidence and acceptance of recovery a great deal.
It’s unusual to refer to placebo as being a treatment. But when used in trials with the associated belief that it is a real treatment then it really is a treatment.
Placebo has been shown to be an effective treatment compared to other medication many different illnesses including:
- Attention Deficit Disorder
- Parkinson’s disease
- Irritable Bowel Syndrome
Generally placebo is actually administered to patients deceptively. Patients believed they were taking a new highly effective drug regime and they received the placebo instead. It was therefore believed that the deception is part of the placebo response trigger.
A significant trial took place of the non-deceptive use of placebo to treat IBS (ref9). One woman who took part in the trial was interviewed on TV in a BBC Horizon program. She was a qualified medical assistant suffering from Irritable Bowel Syndrome (IBS), it was accepted that her medication was keeping the symptoms under control. She understood completely from her training what a placebo was, and she agreed to exchange her medication for placebos for the trial.
She described doing it in the sure knowledge that it wouldn’t and couldn’t work. Then after three days on the placebos the symptoms went away again much to her surprise.
The study reported that 62% of participants got adequate relief from the placebos.
When I started researching this article I was under the impression that double blind trials and the placebo effect was restricted to medical and psychological illness, and that double blind trials were never done for surgical procedures. I freely admit that I was wrong. There is a body of evidence, not as much as for non-surgical treatment, but important never the less, that indicates that the placebo response has been detected in double blind surgical procedures.
A double blind placebo surgery trial published in 2004 investigated the effectiveness of transplantation of human embryonic dopamine neurons into the brains of persons with advanced Parkinson’s disease.
Some patients were given the full procedure and others a sham surgery. Sham surgery is the term used to describe sufficient surgical intervention to make it impossible for a placebo patient to tell they had not received the ‘real’ procedure.
The quality of life determined after 12 months showed that the patients who believed they had the real surgery did better than the ones who believed they had the sham surgery. Belief and actuality in this case was not the same thing. Some patients believed they had received the full intervention but they had received the placebo procedure, others believed they had received placebo intervention when they had actually had the full procedure, the remainder were correct in their assumptions about treatment. The conclusions indicate that the placebo effect was significant in the surgical procedure.
Most of the evidence for the doctor patient relationship is based in therapy for mental and behavioural issues. Because such therapies and treatments are delivered on a personal basis then it is deemed to be a more important facet of treatment.
In all medical interactions between doctor and patient, and indeed therapeutic interactions between any therapist and patient there is a therapeutic alliance. The clinician must play their part and the patient must play their part too.
There are more and more examples coming to light where a ‘good doctor’ who provides encouragement and explains treatments in an accessible way will tend to get better results compared to a doctor who neglects the so called niceties. Even when administrating the same treatment regimen, the better patient relationship has been shown to pay dividends. (ref 6).
The therapeutic alliance provides examples of placebo and nocebo in action, where it is possible for the attitude and rapport skills of a doctor or therapist to actually to affect outcomes of an intervention. Their people skills can contribute to a negative outlook that leads on to poor results, or to a positive outlook that promotes better healing.
In my work at a hospice, I have seen the effect of a patient who was told by his doctor to get his affairs in order because he had weeks and not months left to live. The patient I worked with was severely anxious because he’d been essentially told that he was going to die and it would happen soon. That may have been a true and accurate prognosis from the doctor but the delivery of the information and the inferred lack of support in the future left the quality of life of the patient at a real low. The resulting anxiety reduced the effectiveness of the individual to deal rationally with every-day life. Continual anxiety has the same effect as long term stress and is generally detrimental to mental and physical health.
In this case, the support and ability of the hospice to provide a trusted environment, with credible health professionals on staff, at the same time as treating the patient as an individual was enough to help him regain balance. One of the key skills the hospice prides itself on is listening to the patient, it’s a key part of rapport and people skills.
Essentially it has been suggested from trials of performance enhancing drugs (ergogenic aids) that part of the effect was expectancy. This premise was tested (ref 16) with the intention of demonstrating to athletes that the belief or expectancy of an improvement was enough to get a legal performance enhancement. The trial was administered as a double blind trial and athletic performance was measured by running a 1000m timed trial.
Essentially they demonstrated that athletes who believed they had taken the performance enhancing drug ran faster than normal, and nearly as fast as would have been expected if they had taken the real drug. On the other hand they demonstrated that taking the placebo without believing it to be a performance enhancer had no performance enhancing effect.
BBC Horizon created a TV program that demonstrated the research.
It has been said anecdotally that complementary therapies are only placebos, and it is usually said in a derogatory fashion.
If we put that in to context then it means that many complementary therapies are effective for many issues for up to 60% of people. That seems pretty impressive when you think in terms of populations and issues that could be helped with therapies that are readily available and generally inexpensive to administer.
Irving Kirsch described hypnosis as a non-deceptive placebo. In other words, it’s a mind based therapy that should have no effect on physical symptoms, and yet it has been shown to be effective in many instances. There is however a considerable (and growing) body of evidence to support hypnosis as a therapy. But if it is a placebo then so be it.
If it’s a question of belief and expectancy as is commonly thought for placebo then it’s quite possible that triggering the placebo effect in different people may require different therapies. So one person may find mindfulness meditation effective for anxiety where another finds that a basic massage has the same calming effect.
Placebos appear to be effective at managing symptoms but they don’t affect underlying disease. For example examination of cancer trials involving placebo showed a very low (1 in 191) improvement in tumour size but did report improvements in symptoms and pain (ref 5).
Symptom management is an important facet of treatment. Symptoms are the effects of a disease that the patient experiences, so improving the quality of life depends on removal or control of symptoms like pain or nausea.
This is often referred to in life limiting illness where the medical prognosis is an early death from a disease. But the patient experience can be improved by properly applied palliative care, which involves removal of symptoms to facilitate a better life experience.
It is not just the realm of the terminal illness though. Any illness presents symptoms and they often cause frustration and other negative emotional reactions. If the symptoms are removed then the body can often heal better because of the removal of negative effects.
If we take the frustration as an example, it has an associated increase in stress hormones that are common to the fight or flight response. The presence of adrenaline causes fatigue because it leaches away the stored energy from muscles. Removal of that symptom would leave the body with more energy reserves to deal with disease and make a patient feel more energised.
With the advent of functional magnetic resonance imaging (FMRI) of the brain, a new approach to experiments in to placebo has been made possible.
An experiment in to the effect of placebo on the anticipation and experience of pain (Ref 7) was set up to investigate if placebos alter sensory pain transmission, pain affect, or simply produce compliance with the suggestions of investigators.
They found that placebo induced analgesia was related to decreased brain activity in pain-sensitive brain regions. Similar areas of the brain were affected in the anticipation of pain. They concluded that the experience of pain was actually changed by the placebo analgesia (pain killer). That was an important revelation.
Using FMRI has opened up new areas of drug testing too. Double blind placebo controlled trials of a range of drugs including pain killers and drugs to treat mental conditions are now able to compare the induced brain activity of the actual treatment with the effect of placebos. It’s another development that may lead to more effective drugs. It may also lead to a better understanding of the placebo effect.
There is evidence that the placebo effect is actually not all bad. It is still the bane of the drug companies and a major contributor to escalating costs of producing new drugs.
In fact it should be seen as good news that so many people can heal themselves with the right encouragement and stimulation.
The fact that patients can show reductions in pain and symptoms just because they believe that is what should happen is something that should be encouraged. And there is evidence of that starting to happen (ref 6).
The fact that mental health has been shown to improve with the use of drugs that performed no better than placebo, should be encouraging too. It means that for some mental health issues, people get better because they think they should.
The evidence is starting to emerge about what parts of the doctor patient relationship contributes to a higher rate of placebo and therefore improved therapeutic results.
There is evidence as to what constitutes good practice in dealing with patients, but it appears that it is mostly taken in to account in areas of mental health.
It may help if a systematic approach to determine validated best practice so that doctors treating patients suffering from medical illness as well as doctors and therapists treating mental illnesses can truly be part of the healing experience and not just a dispenser of pills and sick notes.
It may pay dividends in the long run to discover how to boost the placebo effect, rather than see it from a drug company point of view. That isn’t to say that drugs play no part in health care, they are incredibly important, but demonising the placebo effect (as in the past) is not the answer.
1. Tuttle, A. H. et al. Pain http://dx.doi.org/10.1097/j.pain.0000000000000333 (2015).
2. Dunlop, B. W. et al. Neuropsychopharmacology 37, 2830–2836 (2012).
3. Leucht, S. et al. Mol. Psychiat. 14, 429–447 (2009).
4. A. J. de Craen et al. BMJ. 1996 Dec 21; 313(7072): 1624–1626.
Effect of colour of drugs: systematic review of perceived effect of drugs and of their effectiveness.
5. Chvetzoff G1, Tannock IF. Placebo effects in oncology. (2003)
6. Ernst and Herxheimer. The power of placebo. BMJ. 1996 Dec 21; 313(7072): 1569–1570. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2359089/?page=1
7. TD Wager et al 20 Feb 2004:
Placebo-Induced Changes in fMRI in the Anticipation and Experience of Pain
8. Kelley JM et al. Psychosom Med. 2009 Sep;71(7):789-97. 2009 Aug 6.
Patient and practitioner influences on the placebo effect in irritable bowel syndrome.
9. Placebos without Deception: A Randomized Controlled Trial in Irritable Bowel Syndrome
Ted J. Kaptchuk et al. December 22, 2010
10. Wolf, Buckwalter
Randomized Surgical Trials and “Sham” Surgery: Relevance to Modern Orthopaedics and Minimally Invasive Surgery
11. Jonas WB, Crawford C, Colloca L, et al To what extent are surgery and invasive procedures effective beyond a placebo response? A systematic review with meta-analysis of randomised, sham controlled trials BMJ Open 2015;5:e009655. doi: 10.1136/bmjopen-2015-009655
12. McRae et al. April 2004
Effects of Perceived Treatment on Quality of Life and Medical Outcomes in a Double-blind Placebo Surgery Trial
13. Palla et al BMC Pulmonary Medicine May 2012
Over prescription of antibiotics for adult pharyngitis is prevalent in developing countries but can be reduced using McIsaac modification of Centor scores: a cross-sectional study
14. Herman Goossens Matus Ferech Samuel Coenen Peter StephensEuropean Surveillance of Antimicrobial Consumption Project Group April 2007
Comparison of Outpatient Systemic Antibacterial Use in 2004 in the United States and 27 European Countries.
15. J. Christopher Muran, Jacques P. Barber 2016
Book. The Therapeutic Alliance, An evidence based guide to practice.
16. M McClung, D Collins June 2007
“Because I know It will!”: Placebo Effects of an Ergogenic Aid on Athletic Performance