The Placebo Effect and how we might heal ourselves.
First written as an under-graduate psychology student on 12/02/2003. Blogged here without correction or updates.
Abstract: The nature of the Placebo effect and its plausible explanations are examined – specifically the classical conditioning and endorphin based theories. The methods and administration of placebos is considered along with the contexts in which placebo effects occur. The roles of the patient and doctors are assessed with particular reference to the influence of expectancy. The implications of an internal mind-based healing system and its potential applications are discussed.
Healing and the Placebo Effect
The deliberate augmentation and preservation of individual health and wellbeing is estimated to have begun no less than 60,000 years ago when humans first started to use plants to sooth and clean wounds and mend broken bones with splints. However the natural, internalised systems of healing have been brewing and growing stronger in the evolutionary cooking pot over tens of millions of years; refining and enhancing the human system through natural selection. As societies and cultures have also evolved across the Earth, building social hierarchies and seeking knowledge about the world around us, so too has the faith and reliance on individual sources of healing.
From sacred medicine-men and Holy shrines; to the strange charms, rituals and magic potions of the intellectually and socially peculiar groups who tend to dress in white. That is both of the past, and today where our high status members of academia are stereotyped in lab-coats and have a very complex language, understood only by the rightfully initiated. The cross-over of paradigms in healing, old and new, that co-exist today are often at times methodologically and theoretically conflicting; from modem conventional prescriptive medicines and psychological treatments to ancient “complementary therapies” such as acupuncture and reflexology.
There is also a large devotion to various forms of a more spiritual healing such as faith healing, crystal healing and prayer. Although many of the old, bizarre and potentially damaging methods such as ingestion of frogs-sperm, monkey dung, and ground bones; or cutting, bleeding and blistering ; or shocking/freezing and even bludgeoning; are no longer used, many old and new methods are still sought and relied upon. The one thing that links them in common with one another is that they are all regularly associated with illness and are (or were) at one point believed to heal. The one thing that has been guaranteed consistent over time; endured all the scrutiny of every new medical fad and idea; is that by some means, people get better.
The number of remedies long forgotten and those still used today that have relied solely on the placebo effect is astounding and equally frightening; the US office of Technology have estimated that of the modem medicine readily available, only 1/5 has been proven to have any pharmacological effect; in fact to some degree all 5/5 have a reliance on the placebo effect! The terms “placebo” and ” placebo effect” were first introduced into the language of physicians in 1811 via publication of a new medical dictionary where the placebo effect was believed to be a treatment given to the patients to please them rather than medicate them; to “cheer the spirits”. The mysterious, latent power behind the placebo lay unrecognised and ignored by most for over a century.
Does the Placebo Effect exist and what is it?
The term “placebo effect” has been scrutinised ever since first suggested and the events that we term the placebo effect are just as difficult to generalise and “pin-down”. The exact definition of the Placebo effect is probably as elusive as our grasping an understanding of it, but is most commonly known to be a positive effect relating to the treatment of an illness with a chemically inert substance. Shapiro and Shapiro (1997) have detailed an entire work on the definition of the placebo and they come to a generally accept view:
??any therapy prescribed for its therapeutic effect on a symptom or disease, but which is actually ineffective or not specifically effective for the symptom or disorder being treated.?
Further more, the authors describe the placebo effect:
“the non-specific psychological or psychophysiological therapeutic effect produced by a placebo, or the effect of spontaneous improvement attributed to the placebo.?
The term “placebo” is Latin for “I will please” and has been a universal medical mystery for as long as cures for sickness have been sought. Clearly many a charlatan has relied upon the placebo effect for sales, perhaps even innocently so.
Of course, the placebo effect does have a harmful alter-ego, the Nocebo Effect. This is the polar opposite of such a healing mechanism and translates as “I will harm”, such recent interest in what we now term psychosomatic illness and areas of health psychology where the immunosuppressive effects of grief are studied; demonstrate this universal self-harming phenomenon and have resulted in the beginning of a paradigm shift in the western medical thinking.
The elusive nature of the placebo effect often casts doubt on its existence and it is often the case that the placebo effect is likened to a chance observation that is circumstantial e.g. many patients tend to exaggerate symptoms to their doctor to qualify why they are seeking help and then obligingly report relief that may well be exaggerated to please their doctor. Although this is sure to occur, it cannot account for the entire phenomenon as placebo effects found in clinical trials which are very different circumstances from a doctor?s office and strict controls are made to account for such biases.
There are three types of healing that occur in response to physical injury:
- Autonomous responses are processes that are self- regulating systems such as the immune response, i.e. where the body heals itself.
- Specific responses are those which occur by facilitating the use of medicines ie antibiotic properties of certain fungi or salicylates of particular plants
- Thirdly there is the Placebo effect (or meaning response) where the healing occurs after some interaction with a healing context i.e. in response to the latest “branded” pain killers or the “power” behind an impressive medical technology.
Being given a placebo treatment is not akin to being given no treatment. Patients who receive placebos reliably report improvements in their health condition whereas control patients who are given no treatment what so ever, do not show this trend of improvement. It is perhaps surprising to note that most clinical trials conducted to test the effectiveness of a new treatment rarely employ a no-treatment condition. Clinical trials take the general form of a Randomised and controlled trial. That is, during the randomised controlled trial there is a group of patients who receive the new drug and a group that receive placebo. It is double blind meaning that neither the experimenters or the patients know which is given to who and so the experiment will be believed to indicate the effectiveness of the new drug by comparison. However, it is difficult to see exactly how effective the placebo has been without a comparison to a no-treatment group. This confusion has been noted and studied by Ernst and Resch (1995) who differentiate between the true placebo and the perceived placebo. The perceived placebo is the response that can be observed in the patients receiving the placebo in a randomised controlled trial; where as any true placebo effect can only be discovered by considering a number of factors- such as regression towards the means and the natural history of illness- that affect the duration and nature of the illness.
Regression towards the mean: All illness and disease is subject to fluctuations in severity of symptoms over time that is, no disease is without its cyclic decline and improvement whether or not the overall, long term trend is towards recovery or death. If measurements (i.e. of enzymes in a blood sample) are taken at different points during a clinical drug trial, statistical probabilities must be carefully considered. The enzyme count may be at a severe low when initially taken since treatments are normally given to more extreme cases; and due to the natural variability of enzyme production, when the next measurement is taken, it will be likely that the count will be higher. It is more likely that the fluctuation will increase the count rather than decrease it further and so the sample taken, will probably reflect a value closer to the mean score just by allowing time to pass and opportunity for natural fluctuations to occur as the body seeks homeostasis rather than reflecting a direct response to any specific drug treatment or placebo.
The natural history of the disease: Although some diseases do worsen without treatment, many get better due to the body’s immune response and are said to be ” self-limiting”.
The idea of spontaneous improvement in fact subscribes to the natural history of disease in this sense e.g. most colds clear up within about two weeks or at least show improvement in symptoms but it would be incorrect to infer a placebo effect has occurred, Keinle and Keine (1996). A placebo may well simultaneously relieve symptoms but we cannot be sure it has. Even the definition (Shapiro and Shapiro) accepted, makes way for confusion of the real and perceived effect. As Moerman (2002) points out:
“…left to itself, a cold will last about a week and a half, but when treated with the armamentarium of modern medicine, will last only about ten days.”
The rationale of most clinical experiments, testing drugs involve double-blind randomly controlled trials. Usually this takes the format of comparing a drug with other known and formalised drugs and of course with a placebo group – a group of patients who unknowingly receive an inert substance disguised as the drug being tested. Often this results in the placebo group showing improvement along with the new and active drugs, all be it perhaps not as much as the active drug. However it is surprisingly rare that clinical trials ever compare the new drug with a no treatment group. In the few studies that do make this comparison it is generally found that the placebo group make a significant improvement over the no treatment group.
The Placebo effect is not merely a mental disease and not an organic one; although there are often placebo effects reported and diagnosed as relief of some psychosomatic symptom; there remains a vast number of cases where the placebo effect has relieved physical organic disorders that were nothing other than “real” (Wall,1999). Also in the same study, Wall shows there to be no such thing as a placebo-prone personality type and all attempts to find such a person have ended in a mass of contradictions and prototypes suggested being an example of the Barnum effect.
It appears that any number of patients can respond to a placebo; depending on the circumstances of the trials, Wall estimated anything from 0%-100% whereas Beecher (1955) claimed that a fixed rate of 35.2% of patients would respond, however there has been much controversy over Beech ‘ s findings; Keinle and Keine (1996) point out that many of the studies on which Beecher’s paper was based are methodologically weak and even involve statistical errors.
According to Skrabanek and McCormick (1986) the placebo effect is only found in subjective feelings and no placebo acts as an agent that results in measurable physiological improvements. Support for this comes from many anecdotal cases where patients report feeling “better” whereas in reality their condition is worsening. Buckman and Sabbagh (1993) describe a prime example – the case of Evelyn White (a progressive breast cancer patient with a large tumour under her arm) who felt that chemotherapy was ineffective and making her sick, and so opted to try a complimentary therapy known as Cancell. She visited her usual doctor regularly and each time reported feeling less pain and believed her arm to be less swollen. The doctor kept routine measurements of the swollen arm (measured at the same location each time) and found that the arm was continuously increasing in girth. Mrs White gradually died believing that the Cancell had given her much relief. The doctor never told her that the arm measurements had been increasing and that the Cancell had no apparent healing effects; and so perhaps the expectation or “faith” in the treatment allowed Mrs White to suffer less, but there was no evidence of any placebo effect relieving the physical symptoms – despite her belief.
There are many accounts similar to this made by doctors over the history of medicine but they are anecdotal and other researchers purport that physical changes can be and have been observed after the administration of a placebo. Ernst and Abbot have noted that certain common conditions seem respond best to placebos such as migraine, sleep disorders and depression; the placebo seems to affect both the pathology and the symptoms i.e. the disease itself and the subjective illness experienced by the patient. Ernst and Abbot ( 1999) propose that objective measures of health are placebo-prone along with the subjective feelings of the patient i.e. blood-test results and tissue swelling. Similarly, Ho, et. al (1988) observed the reduction in post ? operative swelling in patients having their wisdom teeth removed Keinle and Keine, however, point out that although the placebo group responded to their inert treatment, in all instances of the Ho experiment, a cooling cream was applied and this may have had an effect on the swelling itself.
The nature of the Placebo Effect
It is important to note that the placebo effect, which is one of healing or relief from symptoms, is not caused, per se. by the administered placebo. It cannot be. A placebo is inert by definition and so it seems that it is in the action of taking the placebo that results in the effect. The explanations behind the placebo effect are thought to apply to the nocebo effect too, as both phenomena are thought to be two ends of the same spectrum. The placebo effect is, unfortunately rarely studied in it’ s own right. It tends to be treated as a nuisance variable in clinical trials and attempts are generally made to control for it.
However, since the publication of a medical paper entitled “The Powerful Placebo” in the fifties (Beecher, 1955), there has been a steady increase of interest in this healing phenomenon and although this paper has does much to forward the research of placebos , it has inadvertently spread many medical myths as a result of misinterpretations of results due to poor methodology in some studies.
There has been a recent suggestion by Moerman (2002) that what we usually refer to as a placebo effect should be re-named, the meaning response; by which it is to be understood as the psychological and physiological effects of the meaning in the treatment of illness. Moerman therefore includes such meaningful factors as form and method; and even applies to the power of such characteristics of active treatments as well as inert (placebo) treatments. The debate is hot and the understanding of terms invented to define an observation that might never actually have been observed is tricky; so for simplicity I will continuously use the familiar “Placebo Effect” to describe a physiological or psychological response to a chemically inert substance that has no specific properties to treat the condition being observed.
Although current ethical restrictions prevent “Sham” operations from being carried out, they were done in the past, some within the past fifty years. Angina patients who were candidates for artery-tying procedure were randomly assigned to either the proposed operation or a sham operation where they were anaesthetised, and an incision made in to their chest but closed up without the artery-tying. Similar placebo responses were found in that the patients who opted for heart surgery expected that the surgery would help (after being encouraged to do so by their doctors) and the majority of both surgery and sham surgery patients reported the same relief of symptoms, post operation (Cobb et. al., 1959 Dimond et al., 1960).
Clearly the ethical implications of sham operations override the practice of these tests, as interesting as they are, however these effects of positive expectancy in surgery are also seen in patients who receive coronary-artery bypass grafting where diseased sections of arteries are bypassed with veins grafted from their own leg veins. The operations were estimated to yield a high success rate of 90% of patients feeling much improved health and 75% performing better on tests of physical exercise. However, out of the 90% who felt better, only 20% had any improvement in their heart function, 60% showed no improvement and 20% were actually worse. It is interesting that the subjective feelings of well-being experienced (or at least reported) are contradicted by the physical state of health. Perhaps the placebo effect in these conditions is merely perceived as we do not know how these 90% of patients were questioned on their subjective well-being; we would need to ensure the questioning of patients is standardised to remove biases such as those of surgeons, eager to have performed a successful operation.
Explanations of the Placebo Effect
Many have purported a Pavlovian style explanation for the placebo effect; that the healing associated with a placebo is in fact a conditioned response to certain situations. In the classic example of Pavlov’s dog, the dog would salivate (unconditioned response) in the presence of food (unconditioned stimulus) and each time that food was presented a bell would ring (conditioned stimulus). After repeated food/bell pairings, the ringing of the bell was enough to cause the dog to salivate (now a conditioned response).
In much this same fashion as Pavlov’s dog many believe that when people are sick and seek a cure, they tend to visit a doctor and then ingest the suggested treatment which results in alleviation of symptoms. The proposal is that this pattern of seeking and receiving treatment that results in the patient feeling better is repeated so often in our individual lives that the visit to the doctor and / or the intake of medicines is a conditioned stimulus that will produce the conditioned response of healing or subjectively ” feeling better” . With this in mind I suspect that it would have to be the action of seeking and ingesting that is the conditioned stimulus and not he drug itself as it would be unlikely that exactly the same drugs and dosage are given for every doctors visit. Perhaps this theory would be more applicable to a recurrent long-term illness where in fact the same treatments are given continuously; interestingly it is a common belief that the more we use specific drugs on an individual basis, the less effective they will be.
It is said that our bodies develop a tolerance to them after time; and perhaps more research in to the truth and biological mechanism behind this waxing and waning of individual medical effectiveness could lead to more understanding of the placebo effect. Furthermore, if the conditioned response is one of healing rather than merely perceived healing, then this cure-by? association would indeed be a fascinating and priceless technique.
Support for classical conditioning in the Placebo effect was gained from certain animal studies, Ader (1997). The animals who were regularly injected with certain stimulants over time indeed did demonstrate some kind of placebo response when injected with inert saline. They tended to show the same behavioural and physiological effects of the drug in response to the placebo. However some animals actually showed a compensatory response to the placebo injection and this biological and behavioural counter- action resulted in the effects of the previous injections of active drug being neutralised. This makes the classical conditioning theory rather more complicated to suitably explain the placebo effect.
Classical conditioning does seem to propose a logical answer in part to the effect of a placebo; and although both situations result in the administration of a placebo resulting in a measurable response, the bi-directionality of the response through some doubt over this assertion. It should also be considered that there is no guarantee that it was specifically the action of injecting or the presence of a foreign substance (active or inert) in the blood stream that cause the conditioned response – it may have been due to a combination of other factors i.e. the setting and environment of the lab, the time of day, the experimenters, almost ritualised procedure. Perhaps they would have done well to prepare a group of animals that never did receive a placebo injection at all.
In human studies, even more complexities arise. Voudouris and colleagues attempted to demonstrate the role of classical conditioning in the placebo effect in people- specifically the analgesic placebo response. The subjects in this experiment were trained to respond to report the intensity of pain from an electric shock, the power of which they were made aware. One group were given an “analgesic cream” applied to the site of the shocks, and one group was not. The cream was completely inert. The outcome was, as expected, that the group with the cream could tolerate higher intensities of painful shock more than the group without the cream. On subsequent trials with the cream group, the intensity of shock was turned down (unbeknown to the patients) and the cream applied. When testing these subjects with the higher original shock levels, they were able to tolerate even more pain than they had previously done so. The explanation given by the experimenters is that the pairing of the cream with the low shock produced a conditioned response of high pain tolerance (Voudouris et al., 1989, 1990).
As plausible as this seems, there are other ways of interpreting this result. In 1997, the Voudouris experiment was repeated by Montgomery and Kirsch where the same trends appeared in their data, however, a third group was tested and here -in lies the conflict. The third group were treated much like the second group but were told that the shock level was being turned down and that the cream was inert. These subjects should, by classical conditioning yield the same resulting placebo effect , but they did not. It appeared that no association between cream and pain reduction took place. Consequently, the role of expectancy revealed itself to have more potential influence on the occurrence of a placebo effect. Health psychology has been investigating more and more deaths of people from drug over-dose; it is well documented that many individuals who have died from a lethal dose that they had a personal tolerance to, but was administered in different environment form their usual habit. The conditioned stimuli perhaps of their bathroom and environmental settings act as a cue and trigger a conditioned response where their body prepares to receive the otherwise fatal drug. If this cue is absent then the drug could be too much for their bodies; not primed and expecting to receive and tolerate it.
Another more promising approach to explaining the analgesic placebo effect has come from Levine et. al (1978) and their neuro-chemical research. It has been proposed that under times of physiological stress, i.e. pain, the brain releases endorphins; Levine and colleagues have shown that these endorphins when released at times of stress, counter-act the experience of pain. Injecting a substance called noloxene would block the effects of the natural opiate-like endorphins and prevent an analgesic placebo effect. The testing of this theory has proved successful and although the original experimenters have been criticized on their methodology, subsequent more methodologically strict tests have provided the same results. Although this seems to provide an adequate explanation of the analgesic placebo effect, the endorphin release in itself needs explaining. How can an inert substance cause the brain to perform such a task? Also, the endorphin theory relates well to analgesia-placebo but the evidence says nothing for other instances of placebo effect.
Roles of Patient and Doctor
Despite all the theorising and testing, the mechanisms behind the placebo effect is still eluding researchers at every tum; what we do know, however, is that the role of expectancy is very important. As there is no particular placebo-prone personality, the psychological make-up of the individual patient has next to no effect on the outcome pf a placebo response. In contrast however, the attitude, behaviour and contextual knowledge of the medical practitioner seems to be very important in determining a positive response in a patient?s health.
The common thread of thought in the patient and doctor is the idea of certainty and how likely the treatment is to work. Gracely and colleagues performed a pioneering study in the sixties that demonstrated this very thing. On observing the clinicians rather than the patients in clinical dental trials, it was found that the knowledge that the individual dentists had about various active and inert analgesic treatments, greatly affected how the patients responded to their medication. On this study, the dentists received information and mis-information so that they would expect some drugs and placebos to work well and other drugs and other placebos to not work. Gracely believes the successful relief from pain after the administration of a placebo was due to the administration of the dentist’s enthusiasm for its “powerful pain-killing” properties.
Similarly, Schizophrenic patients have been known to respond positively to the treatments given by specific doctors and fail to respond when given the same treatment by others. Whitehorn and Betz (1960) found that the difference in the success rates of the doctor/patient pairings was due to differences in the doctors ‘ attitudes and questioning techniques. As W.R. Houston said some sixty years ago;
?The Doctor Himself as a Therapeutic Agent, [to] be refined and polished to make of himself a more potent agent.”
It seems that the interest of the doctors in their patients, their knowledge and their enthusiasm seems to have a large effect on how well a patient will respond to the treatment they prescribe – if in fact the medication /treatment is in fact necessary to cause a positive response in well-being.
It seems that the event of receiving treatment is more of a predictor of a placebo effect than the content of the treatment itself. Faith in the treatment seems to be influential in the occurrence of a placebo effect; Shapiro and Shapiro (1997) have suggested a link between belief and outcome from their examination of homeopathic medicine v traditional medicine. It was found that New-Age believers tend to show a stronger placebo response to the homeopathic remedy than those reliant on Western medicine.
The link between patient outcome and diagnostic procedures has been studied at length by Brown (1998) providing numerous case-studies and evidence for this placebo effect phenomenon. Many a medical practitioner has witnessed the diagnostic procedure result in relief of symptoms such as an X-ray taken of a patient?s limb that results in the patient feeling “much better” despite a broken bone yet to be treated! Or an electro cardiogram resulting in the alleviation of chest pain (Ernst and Abbot, 1999).
The expectations of the patients are often largely determined by the behaviour and attitude of the medic or therapist advising them or the information received such as that conveyed via the advertising campaign of a product. With this in mind, it is indeed interesting to examine exactly what people believe and understand of the drugs they take in relation to their effectiveness, and perhaps why they come to expect certain effects. Studies have shown that “branded” placebos often result in a greater placebo effect than “non-branded” placebos, the same trend is parallel to the branded and non branded active pain killers (Braithwaite and Cooper, 1981). The physical characteristics and method of administration also seem to affect the outcome of a placebo effect; the colour and number seems to influence the power of the response. Two placebo pills are better than one (de Craen et al, 1999) even though twice the inert value of zero, is zero; learned associations seem to influence what the patient will expect e.g. that the brand of Asprin tablets that they have always relied on and always buy are highly effective. An interesting example is that placebo injections seem to have more potency than pills; since people tend to dislike injections, perhaps there is some cognitive dissonance to overcome and so they attribute more power and meaning to it. Perhaps there is a belief that injections will work immediately whereas pills will take time to affect. Furthermore, the form (colour and size etc) of pills affect the response; context appears to be important in that big blue pills are better sedatives than yellow ones and small yellow pills are more effective as stimulants and relieve depression, inert or not!
Why does the release of such endorphins happen in the first place – what triggers this ” self-help” activity, unbeknown to the patient? Although endorphin release can easily account for reduction in perceived pain (not necessarily reduction of the cause of pain), how can endorphins result in the remission of pathological illness and other measurable effects?
The implications of the placebo effect stretch to every comer of medicine and health care. The idea that sad people prefer yellow pills to blue ones; that branded tablets work better than unfamiliar ones but two unbranded placebos work better than one; that some therapies “work better” in some cultures over others and that when the doctor says “take two paracetamol and call me in the morning..” a strange and complex formulae to well?being emerges.
Although the administration of a placebo implies it to be a chemically inert substance, it seems that the form i.e. colour, size and shape or that the attitude and belief of doctors and patients are far from inert. This ” power” of suggestion certainly give rise to endless possibilities of how our individual and societal well-being may well be influenced and affected by our culture, lifestyles and technology; not to mention the terrifying implications of the power of advertising and media. Just how much do we rely on the healing suggestion of outside influence? With all the money and power to be gained through drug consumerism, just how safe is our well-being in the hands of medical science?
It is unfortunate that a kind of academic snobbery and a dominance of profit-making from pharmaceutical companies allows the existence of a placebo effect to be denied and ignored; the apparent miraculous ability for people to heal themselves- quite unaware they are doing so -has both wonderful and devastating implications. A world where medical attention could be concentrated on the seriously ill and helpless while others harness their natural abilities to heal minor and common complaints sounds like it would result in more productive day-to-day lives for everyone with less sick days taken and less hospital beds occupied. What are the limitations to self-healing and the physical influence of the mind? Are there any? What of our mortality? Perhaps the average natural life span of humans could be increased and there would be good reason to determine the applications of helping the disabled overcome their handicaps- either through prevention or intervention.
However, there may well be a real danger in relying on faith and wish fulfilment to bring about change to extremes where obvious opportunities for medical help can be grossly and maybe fatally over-looked. Also, the many drug companies who spend billions on advertising to make a profit from everyday ailments such as headaches and stomach complaints would surely loose out to self-help markets. Perhaps the manufacturers of “tic-tacs” could employ a whole new marketing strategy and sell yellow coloured, “pick? me-up” breath fresheners that keep both your breath and brain fresh at the mere cost of a mint.
Since the current psychological and medical paradigms offer no particular answers to this healing riddle-demonstrating it to be a universal phenomenon, independent of personality types, but seemingly dependant on beliefs and expectancy; a parsimonious explanation of the placebo effect is still absent. Perhaps we should tum to a more theoretical and philosophical approach; the placebo effect is- as all who chase it- know, elusive and the mechanisms by which it occurs may well be best understood in terms of meta-physics. Research in consciousness studies would also prove interesting to examine what apparently innate healing mechanisms can be employed to heal ourselves or others by the powers of suggestion, faith and expectancy. The physical seat of this function we can only guess at, and it would probably be wise to compare and contrast placebo effects to the results of distant mental healing experiments; where techniques such as prayer and visualisation are used to heal the sickness of others. Furthermore , if we are to uncover some of the mechanisms behind self-healing, then perhaps we could further understand why we get sick in the first place and dare to prevent the development of tumours and disease, after all, prevention is better than a cure.
Ader, Robert. 1997. The Role of Conditioning in Pharmacotherapy. In The Placebo Effect: An Interdisciplinary Exploration. Ed. Anne Harrington, 138-65. Cambridge, MA: Harvard University Press.
Braithwaite, A., and Cooper, P. Analgesic effects of Branding in Treatment of Headaches. British Medical Journal (Clinical Research Ed.) 282, no. 6276: 1576-8.
Cobb, L., Thomas, G.I., Dillard, D. H., Merendino, K.A. and Bruce, R.A. An Evaluation of an Internal-Mammary Artery Ligation by a Double-Blind Technic. New England Journal of Medicine 260, no. 22: 1115 -1 8.
Craen, A.J., Roos, P. J., Leonard de Vries, A. and Kleijnen, J. 1996 Effect of colour of drugs and their Effectiveness. British Medical Journal 31 3, no. 7072: 1624-6.
Dimond, E.G., Kittle, C. F., and Crockett, J.E. 1960. Comparisons of Internal Mammary Ligation and Sham Operation for Angina Pectoris. American Journal of Cardiology 5: 483-6.
Kienle, G. S. and Keine, H. 1996. Placebo effect and Placebo concept: a critical methodological and conceptual analysis of reports on the magnitude of the placebo effect. Alternative Therapies in Health and Medicine 2, 39-54.
Kirsch, I., Weixel, L.J. 1988. Double-Blind versus Deceptive Administration of a Placebo. Behavioral Neuroscience 102, no. 2: 319-23.
Mishlove, J. 1993. The Roots of Consciousness, The Classic Encycloppedia of consciousness studies revised and expanded. Revised Edition. Marlowe and Company, New York
Moerman , D. 2002 . Meaning, medicine and the “placebo ef fec t”. Cambridge University press
Omstein,R, and Sobel, D. 1988. The Healing Brain, A radical new approach to health care. The institute for the study of Human Knowledge. MacMillan London Limited
Radin, D. 1997. The Conscious Universe . The scientific truth of psychic phenomena.
Roberts, R., and Groome, D. 2001. Parapsychology. The science of unusual experience.
Arnold Publishers, London
Wall, P.D. 1999. The Placebo and the Placebo Response. In Wall, P.D. and Melzack, R.(eds),Textbook of pain, 4th edn. Edinburgh: Churchill Livingstone, 1419-30