The placebo effect

Is it better than placibo? a simple question?  Not really!  To test a new drug for to treat some disease, scientists usually use a type of experiment called a randomized controlled trial (RCT).  They recruit as many volunteers as possible and divide them at random into groups that are homogeneous in terms of age, gender, etc.  Then, let's say, one group will receive the drug, another will receive an innocuous pill and the last group will receive nothing.  Most of the time the trial is set up as a double-blind test, which means that neither the volunteers nor the clinicians that hand out the pills know who is receiving the drug or who is receiving the innocuous pill.  The sham pill is a placebo.  Placebo means “I will please” in Latin.  There is a curious story about how it came to be called placebo, that involves St. Jerome's mistranslation of Psalm 116 and hired mourners who chanted the word repeatedly (see Finniss for the story).  The whole point of the RCT is to find out if the drug is effective against the disease.  Unfortunately, very often the sham pill produces a positive effect from moderate to large, and this is the placebo effect.


    The first recorded use of a RCT was in 1784 when Louis XVI appointed Benjamin Franklin and Antoine Lavoisier, among others, to investigate the curative methods that Frank Anton Mesmer, a German physician, developed from exorcism rituals.  Mesmer claimed to have discovered “new forces” that could be manipulated to create “mesmerized” objects.  The healer or mesmerizer would instruct the “patients”, usually women, to touch a mesmerized object; this touch would induce a “crisis” consisting of a display of strange gestures, sounds and behavior that, after exhaustion, would leave the patients with a sensation of well-being.  Franklin and Lavoisier designed experiments that included patients touching mesmerized objects, other touching non-mesmerized objects but believing that they were, patients touching mesmerized objects but informed that they were not mesmerized and finally those touching non-mesmerized objects knowing that they were not.  Although 1784 is the first time that scientists used sham or placebo treatments, they were used before by Catholic and Protestant church members in “trick trials” to limit the popularity of exorcism (See Kaptchuck for more details on the early history of placebo).

What is known

    Today the placebo effect is recognized as real and important, a “genuine psychobiological phenomenon” (Finniss).  Scientific studies of the effect demonstrated that it is complex with many independent components: expectations, conditioning, meaning (memory and learning), emotion (desire, anxiety, reward) and others.  Most of what is known about the placebo effect comes from studies involving pain control.  From results of functional magnetic resonance imaging (fMRI) or positron emission tomography (PET), we know that placebos induce portions of the brain to produce natural opioids, called endorphins, which reduce the sensation of pain.  However, there are other mechanisms that produce the placebo effect.  One involves the family of peptide hormones called cholecystokinins (CCK) which, in addition to regulating digestion and pancreas function, induce the sensation of satiety and, curiously, anxiety.  Other mechanisms influence the respiration and cardiac rates, but they are not well-defined yet.
    The complexity of mechanisms that produce the placebo effect reflects in the wide variety of factors that affect the response.  Having expectation of results is clear. If the patient thinks that he/she is receiving a drug, the placebo effect is larger than if the patient is not sure what he/she is receiving and it is even lower if he/she knows that the treatment is a placebo.  Conditioning, as when the treatment involves a ritual, such as going to a hospital, talking to people garbed in white lab coats and taking a pill, also increases the placebo response.  Seeing other persons taking the pill or submitting to the treatment will also increase the response.  The number, size and color of the pills also have a positive effect as well as the environment in where the treatment is applied; kind and attentive personnel will elicit a further increase in the effect.
    There is also a negative side to the placebo effect and it is call the nocibo effect.  If a person receives indications that a bad or more painful outcome is the result of the treatment, then he/she will feel it.  Similarly, if the person receives a true pain-killer, but under the impression that he/she is receiving a sham pill, the analgesic effect of the pain-killer is much lower than if the person knows that he/she is receiving the pain-killer.  In other words, a portion of the effect of the pain-killer is the placebo effect!  The nocibo effect is produced by different mechanisms than those involved in the placebo effect because the nocibo effect activates different areas of the brain than the placebo effect.

Acupuncture and placebo effect.

    Most RCTs testing the effect of acupuncture yielded no difference between acupuncture, sham acupuncture and placebo acupuncture, but they were all better than no treatment.  The sham acupuncture consisted of using needles, but not in the prescribed points, while placebo acupuncture used retractable needles which, like a dagger in a theater play, do not penetrate (Colagiuri, White).  White and coworkers also tested the effect of empathy by subjecting the volunteers to empathic or non-empathic sessions before the acupuncture.  During empathic sessions the practitioners were kind, compassionate and answered all the questions and tried to please the volunteer while in non-empathic sessions the practitioners were polite but answered in a terse way or refused to answer questions.  The results in this trial show no difference between real acupuncture, placebo acupuncture and a mock electric stimulation.  Empathy sessions did not increase the effect of the treatments, but there were differences among acupuncturists!  In spite of the results many health insurance companies pay for acupuncture for pain treatment.

What about other non-western traditional medicine treatments?

    Non-western traditional faiths and medicines were popularized by globalization  mostly through the influx of immigrants.  The three Indian traditional healing practices, Unani, Ayurvedic and Siddha, followed immigrants from the subcontinent, while traditional Chinese medicine (TCM) has been in the West, particularly in the US for longer.  The Unani derived from ancient Greek medicine which was modified and transported by Arabs to India where it developed into today's practice.  The Ayurvedic and Siddha medicines probably originated in the subcontinent.  TCM derived from ancient Chinese philosophy (see Borins, Chaudhary and Hu).
    Most if not all traditional medicines are holistic.  Furthermore, they consider not just the whole patient but also his/her environment.  They believe that everyone is composed of parts and these parts in turn have characteristics corresponding to humors or cosmic elements.  Each person has a certain balance of these elements.  When this balance drifts away from the proper mixture, disease develops.  Consequently, the treatments are personal.  The healer after examination prescribes a set of medicines and diet and exercises to neutralize the effects the environment is playing in the sickness.  Since some chronic diseases common in the western world are caused or aggravated by life style, some governments in Europe and the USA are interested in the revival of traditional medicines happening now.  They have been prescribing behavioral changes for millennia which are now recognized to play an important part in many chronic conditions in our modern society.
    One problem with the revival of traditional medicines is: How can the effect of these remedies be evaluated in a scientific way?  Science is just a method to study nature; therefore, the question we pose for each of these traditional remedies is: Is it better than placebo?  There are many hurdles for the scientific study of these remedies.  One is their definitions; some of them have wide variations in composition and preparation from region to region or among practitioners.  Another is the composition of the ingredients.  Most of them are biological products which, like everything in biology, have a wide variation from place to place and from time to time.  There are many examples of drugs identified from plants that have entered the modern pharmacopeia.  The most common example cited is digoxin, a common drug for heart problems, isolated from foxglove (common foxglove is Digitalis purpurea), a garden flower.  That the content of active ingredients in plants varied widely and that the method of preparation of the herbal remedies introduces even more variation was discovered early in the history of drug development  (see Goldman for more information).  Finding the active ingredient in a complex mixture is like the proverbial needle in the haystack and requires one to develop  assays for activity that will not take you astray.   Finally, we have the placebo effect which is always present when we measure the effect of not only the drug but the effect of the healer's interaction.  I am sure that many other drugs will be developed from traditional medicines, but the process will be slow and full of false positives (see Hu for more on how to scientifically measure the effect of traditional medicine).

Placebo effect and ethics

    The large effect of placebos in certain ailments brings to mind the question of their application in medical treatment.  But prescription of placebos by medical doctors (MD) OK?  Although some people may show a sizable placebo effect when knowing that the treatment is not effective for the affliction, the most need deceiption to have a placebo effect.  Most MDs think that deception is not ethical and that it impinges on the patient's right to approve the treatment.  But as Kaptchuk says in an interview published in The New Yorker (Specter): “what is so bad about getting better from a placebo?”  This is not a simple question and if you want more details about the ethical conundrums posed by sham treatments, I recommend that you read the complete article by Specter.  Kaptchuk also said: “placebos don't shrink tumors” (Specter) or cure a severe infection.  I would add, they do not work with everybody and they are not applicable to all diseases.  Large placebo effects are often observed in treatments for pain or for intestinal ailments.  One should consider that pain is an alarm system which the human brain creates and manipulates in ways that we do not understand completely.  Now, the gut is a strong influencer of moods and thoughts (the gut feeling), and there are good reasons to think that the brain may influence the gut as much as the gut does the brain.  Perhaps this is the reason for the susceptibility of these systems to the placebo deception.  Additionally, some of the mechanisms that produce our innate sociability may play a role in the placebo effect.  For example, our peculiar use of rituals for affirmation of membership in a group; such as dances, religious ceremonies, sports, family or patriotic gatherings, Tebowing, often produce a sensation of well-being.  Could the placebo effect be a byproduct of some of these social mechanisms?  So far nobody has proposed an ethical way to apply the placebo effect to common medical practice.
    Nevertheless, it is important, when testing new medical treatments, to design experiments that will distinguish placebo from drug effects.  The fact that a treatment is old does not excuse it from scrutiny.  For example, Bilateral internal mammary artery ligation (BIMAL) was a surgical procedure often used in the 1950s and 60s.  The basic idea was that blocking the mammary artery by ligation would redirect the blood flow towards the heart and help in cases of insufficient blood irrigation of the organ.  Very good results were observed, with up to 80% reduction in pain; until research in animals indicated that there was no increase in blood irrigation after ligation of the artery.  An RCT, in which patients received BIMAL while others only had the surgery exposing the artery but without ligation, showed relief from pain by 70 to 80%!  Other examples of placebo surgery are percutaneous laser myocardial revascularization, arthroscopic surgery for arthritis and vertebroplasty for osteoporotic spinal fractures (Jonas).


The placebo effect is real but, so far, nobody has come up with a good ethical way to take advantage of it.  It is, however, a very interesting subject for research into the “unconscious” control that the human brain exerts on our perceptions.  More important perhaps is to be aware of the placebo effect that may lurk behind many new and not well-studied treatments or drugs.