More on the Placebo Effect
by Jeremy Donovan

The quotes in the following thought piece are taken from Anne Harrington's book, The Placebo Effect: An Interdisciplinary Exploration (1999), pictured above.


A person's belief that they have been given a pain reliever can actually cause the pituitary to release the endorphins which are the brain's natural painkillers.

Not proven (Jeremy's hypothesis):

A person's belief that they have been given methods for creating altered states of reality, can actually cause the brain to activate the systems and release the neuro-transmitters which are the brain's natural means of modeling reality.

In other words, for certain "lucky" individuals who get really "out there" :-) , the hypothesis is that the brain is quite capable of modeling actual perceptual experiences of whatever you strongly believe. It can do this either in dreaming or in waking, but it is probably much easier in dreaming, because then many of the systems modeling from outside perceptual data are shut-off.

I present this not only as one who is academically theorizing that this occurs, but also as one who has experienced this sort of thing.

Experimental evidence indicates that placebo effects are not limited to humans, thus they are not limited solely to what we consider the human "rational capacity."

Some of this sort of thing appears to occur on the level of "conditioning." As an example related to the nocebo effect, giving mice a sugar solution to drink causes them to die at high rates, after they have previously been given the same solution in conjunction with a nausea-inducing immunosuppressive drug, compared to similarly treated mice who were no longer given the sugar water or the drug, and compared to mice who are given only sugar water.

The mice were conditioned to the experience of having sugar water make them sick, and their bodies kept getting sick from it even when it was no longer genuinely harmful, to the point that they died sooner.

If anyone would like to object that such results may not apply to humans, I'll let you be the one to volunteer for such experiments--and I'll also leave it to the reader to create diabolical human examples utilizing this same principle wherein we do in fact unwittingly "volunteer."

However, the placebo effect in humans need not involve pure conditioning, and can definitely be related simply to belief or expectation.

Placebos. . .

."..teach us how far we still are from closure on the question of what it will mean to create a science subtle and complex enough to encompass all that is entailed in being human."

Supported by evidence:

."..the available data supports the somewhat startling hypothesis that the history of medical treatment until recently is largely the history of the placebo effect."

Jeremy's hypothesis (not yet supported):

So too, the history of religion, mysticism, and shamanism is ALSO largely the history of the placebo effect--in accord with the ability of the brain to actually model altered perceptual states based on strong belief, as hypothesized above.

Medical treatment, in evolutionary terms, has only very recently ascended from a primitive state. Prior to very recent times, the old shamans and healers of various indigenous peoples were very likely often much better physicians than those of western civilization who called themselves "doctors." This, because for hundreds of years, western medicine was based on alchemy and all sorts of practices which outside of the placebo effect can actually be harmful to human bodies (like bleeding, dehydration, and ingesting mercury and other poisonous substances), while at least tribal healers were consumate experts in the manipulation of situations to get the full power of the placebo effect, and used mainly herbal concoctions which were more likely to be, at a minimum, benign.

As late as 1950 . . . "45 percent of published studies were uncontrolled and 18 percent inadequately controlled."

It was only in the 1970's and 1980's that the double-blind method became the standard.

The point: a truly scientific approach to medical treatment is a VERY recent development. We are in the infancy of scientific medical treatment.

A definition of "the placebo effect":

"The nonspecific, psychological, or psychophysiologic therapeutic effect produced by a placebo, or the effect of spontaneous improvement attributed to the placebo."

Please notice, our bodies have a natural ability to heal themselves, and in some small percentage of illnesses, serious or minor, people DO heal spontaneously, on their own, without treatment of any kind. It is important to realize this must be accounted for when analyzing results, and otherwise, cases of spontaneous remission of serious illness are liable, through various kinds of human drama, to end up being regarded as "miracles."

Harrington gives a rundown of prescientific medicine, and describes having found accounts of 2,000 drugs and 16,000 prescriptions used in China, 265 Sumerian-Babylonian-Assyrian remedies, 600 drugs used in India, 400 drugs mentioned in the Hippocratic body of teachings. . ." etc.

"Galen's pharmacopoeia, which dominated treatment for 1,500 years, totaled 820 placebo remedies."

Harrington claims that with only a few exceptions, the great majority of the ancient remedies were placebos. She gives examples of a few treatments which turned out to have real merit (although even these were usually inconsistently or inappropriately applied), but then presents long lists of ridiculous and/or terrifying concoctions commonly given in order to heal. Harrington claims that in spite of this, physicians were still respected because . . . they were the therapeutic agent for the placebo effect.

See, the placebo effect, whatever it is, is REAL. :-) Apart from how it is defined above, I would attempt a simple definition as: the power of our own expectations to affect our bodies (and our perceptions of reality).

Before getting too excited about that, it should also be noted that while our beliefs and expectations are amazingly effective, they are often somewhat less effective than other treatments. For example:

"Placebo effectiveness ranges from 30 percent to 50 percent in depression (Brown 1992); and when compared with effective drugs, placebo effect is 59 percent as effective as tricyclic antidepressants (93 studies), 62 percent as effective as lithium (13 studies), 58 percent as effective as nonpharmacologic treatment of insomnia, and 54 percent to 56 percent as effective as injected morphine and common analgesics (22 studies) (Evans 1985)."

It also applies to surgical procedures. If people believe they have undergone an effective surgical procedure, often they will actually feel better (even when in fact, the treatment was a sham treatment).

"In a recent survey of surgery for lumbar disc disease, although no disc herniation was present in 346 patients (negative surgical exploration), complete relief of sciatica occurred in 37 percent and from back pain in 43 percent. In another study of back pain, 20 percent to 40 percent of patients had decreased pain and improved functioning when treated with sham transcutaneous electrical stimulation and hot packs (Turner 1994)."

That means 148 people experienced complete relief of their back pain simply because they believed they had been given a surgical procedure designed to relieve that pain. While that is simply incredible, it certainly lessens one's amazement at tales of "psychic healers" like Pachita, does it not? One must remain in awe, not so much at the healer, as at the amazing power of human expectation.

Question for all to consider:

In light of this data, it would be incredible indeed if there were ANY form of healing which did not have many people walking around believing that it helped them, would it not?

Jeremy's additional question:

In light of this data, and adding my extended hypothesis regarding the brain's ability to influence and/or create altered perception, it would be incredible indeed if there were ANY form of religion, mysticism, or shamanism which did not have many people walking around believing they had real and powerful experiences related to it, would it not?

And what's the problem, you might ask? Isn't it a good thing that almost any treatment can end up making some people really feel better? Isn't it a good thing that people have real and powerful experiences of altered consciousness? Well, yeah, it is a good thing. :-) In some ways. It can also be a bad thing, because it implies that anyone can come up with anything, and no matter how idiotic it is, if they are charismatic enough to get some people to believe in it, a few will get real results, and then I'll have to spend all day arguing with them about it on the net! And . . . and . . . what is a better thing is to find treatments and practices which are highly effective, over and beyond the effect to be obtained by our beliefs alone. For you see, my contention is that there IS a "reality" which is "beyond" our beliefs alone. To illustrate I will ask another "question" . . .

The philosopher's favorite:

The very existence of external treatment methods which in double-blind studies have effectiveness rates far above and beyond that which may result from our beliefs and expectations--doesn't this alone imply that there IS a "real world," one which is "out there" apart from us, which exists and has dramatic effects APART from our beliefs and expectations? Whew! :-)

Replica Watches  Replica Watches

Those who fear that "the mysterious" is soon to be swamped by "the scientific" may be happy to learn that there is no sign that alternative healing methods are becoming less common as scientific knowledge increases, and may also be happy to hear that at this point the placebo effect is itself still somewhat mysterious.

A statistic is given regarding faith or psychic healers . . . that there are "15,000 in England and 20,000 in France." In a way, this proliferation makes the art quite "safe," for, as one researcher commented: "The only man with a right to the last word on faith healing . . . would be the last man on earth."

There appears to be a lack of conclusive data on whether or not the placebo effect can actually heal a truly serious physical illness. I had presented studies on this previously, and now come to find that they are held in doubt. For some reason, to me it seems reasonable that this should be possible (the question arises: where do you draw the borderline?), so I hope they are able to establish this eventually.

There is, on the other hand, considerable evidence that the nocebo effect can certainly contribute to serious illness. The proposed mechanism (which seems likely to me) is that "psychological factors such as depression or stress can impair immune function," but again, the hard core studies on this are not conclusive at the moment. It also stands to reason to me that if this is true, then factors such as optimism and relaxed well-being may well be found to strengthen the immune system. Intuitively, this also seems likely.

Some effort has been put into trying to determine what sort of people respond to placebos and what sort do not. The picture so far is quite complex:

"Clinical placebo effects are not correlated with susceptibility to hypnosis, tests of suggestibility, or laboratory studies using volunteers. . . . Placebo stimuli must reflect a credible therapy for the patient. Placebo effects are influenced by patient-healer interpersonal relationships and are increased in pleasant, nonthreatening, efficient clinical settings with doctors who are perceived by patients as warm, likable, and interested in them. A positive placebo effect occurs more frequently in patients with manifest or free-floating anxiety and with expectation of improvement by patients, doctors, and staff. Expectations of improvement, however, may be independent or overlap with factors such as optimism, enthusiasm, hope, faith, belief, motivation, and conditioning."

(Personally, I find that last sentence confusing, and suppose I'd have to read the actual studies to find out how "optimism" or "belief" differs from "expectation of improvement." To me, "I believe I will improve" seems awfully similar to "I expect to improve.")

At first I could not understand why "positive placebo reactors had high anxiety scores"? I could not comprehend why someone who is more anxious would be more likely to experience a positive response to a neutral treatment? But as experiments show, one of the main ways placebos relieve pain is by significantly reducing anxiety, so it follows that people with a lot of anxiety have more anxiety to reduce, and so may get even more "effect" out of a placebo. This issue is paradoxical, however, for it turns out that high anxiety itself causes the body to release an opioid mediated pain-killing response of it's own (the same opioids the body releases when one ingests what one believes to be a pain killer). It has been shown (in a very complex experiment) that administering a placebo can cause release of the same opioids, while simultaneously reducing the natural release of the same opioids (by reducing anxiety). Weird.

In any case, they have already learned a great deal about how pain responses are handled in the midbrain, brainstem and spinal cord of higher mammals, and these "pain pathways" seem to be the pathways which "mediate" a lot of placebo responses. Further research into neurotransmitters and activation pathways in the brain is called for to better understand the mechanics of placebo responses. Many of the tools necessary to accomplish this already exist (I'll spare you the gory details).

Many factors which were believed at one point or another to correlate to placebo responses have been found NOT to correlate, such as: "demography, diagnosis, past illness and treatment history, previous use of drugs . . . social desirability, acquiescence, dependency, dominance . . . ego strength, IQ . . . Introversion-Extraversion, Authoritarianism . . . .

It really does seem to be largely about "expectation" and whatever goes into creating "expectation" in the immediate environment. Thus any behavior which directly and positively affects immediate expectation would be effective, and so it is logical to find that:

"Positive placebo effects increased significantly when the placebo test instructions suggested a favorable response to treatment."

One chapter concludes with a great question:

"Does the ubiquity of the placebo effect throughout history suggest the possibility, popular but hardly testable today, or perhaps ever, that positive placebo effects are an inherited adaptive characteristic, conferring evolutionary advantages by reducing despondency, depression, and hopelessness, and that allowed more people with the placebo trait to survive than those without it?

Another definition of "the placebo effect":

"a term temporarily adopted to label a group of effects whose specific mechanisms of action we do not yet understand even though placebo's efficacy has been demonstrated in thousands of studies."

Some ailments seem to respond particularly well to placebo treatment, for reasons no one is quite certain of. For example, in a study of a medication given to heal ulcers, while the placebo did not perform as well as the active drug, it nonetheless got most impressive results, actually healing the ulcer crater within two weeks in up to a third of the patients receiving the placebo. (I guess Harrington and the other researchers do not consider an ulcer a "serious physical ailment" for they keep claiming there is inconclusive data regarding the curing of disease . . .). The most remarkable fact was that, in addition, the placebo performed equally well to the active drug at relieving the pain of the ulcer, actually curing the dyspepsia of at least half of the placebo takers in some studies. (The makers of the drug chose to use the results regarding repair of the actual ulcer crater, of course--50% compared to 33%. . .)

The same researcher who gives the above info on the ulcer studies blatantly asserts: "placebos help illness, they relieve pain, but they do not cure disease. I have examined the claims for placebos helping cancer and other diseases, but I have found not one placebo that stands up to scrutiny."

This guy was full of great info on ulcers, though. He talks about how, for years, milk was the staple of an ulcer diet, and was widely believed to really help ulcers. Then in the 1970's it was shown that drinking milk actually raises gastric secretion, and overnight milk was removed from the recommended ulcer diet. More recent data complicated the matter, showing that milk stimulates endorphin production and epidermal growth factors. The researcher concludes by asking: "Was milk a placebo or an antacid? Both?"

Two more interesting and somewhat provocative quotes from the ulcer guy (Howard Spiro):

"Spontaneous remission of cancer should not be mistaken for an effect of a placebo. Nor should survival, if we look at the bell-shaped curve: some 5 percent of persons die "too early," while the 5 percent at the other end of the curve live "too long." Swallows arrive at Capistrano unfailingly on a specific day because those who gather early are called "heralds," while those arriving late are deemed "stragglers." By the nature of things some people with the same disease live longer than others. It is only when healers who claim a beneficial influence can predict who will survive that we can begin to consider that placebos--or faith healing or homeopathy--improve disease."

Keep in mind, this is the same guy here in quote number two. . .

"The mind can surely translate its feelings into physiological function: people die of fright or anger. . ."

In spite of this admission, he seems to believe that stopping serious disease is a different ballgame altogether, and claims that the research supports him.

The nocebo effect can go as far as what is called "voodoo death." The researchers tend to argue a lot about this one, and tend to claim the data for voodoo death is bogus. But . . . I say . . . if it is true that a person can die of fright, then why couldn't a person die of fright after having been "cursed." It seems like it could happen. But most of the experts seem to think that if it exists, it is more a function of induced "hopelessness" than of fear.

The nocebo effect is the causation of sickness by expectations of sickness and by associated emotional states. Here is an account of the most impressive study I found:

"Anda et al. used a sample of the U.S. population to examine the effects of depression on ischemic heart disease (IHD) incidence and mortality in the U.S. population. They excluded subjects whose initial depressed affect might have been the consequence of chronic disease; and they examined persons who were free from heart disease at the outset of the study. Depression was assessed from the General Well-Being Schedule. Anda and colleagues found that persons with depressive affect were 1.6 times more likely to have nonfatal IHD and 1.5 times more likely to have fatal IHD than persons who did not have depressive affect. Considering that an 11.1 percent prevalence of depressed affect was assessed in the study cohort--a sample of the U.S. population--it can be estimated that more than 5 percent of U.S. IHD mortality (that is, approximately 26,000 deaths a year and more than 1% of all U.S. deaths) are attributable to depression, independent of other risk factors."

While the conclusions are somewhat questionable, the implications of this study blow me away. I'm going to resist the urge to comment at length, trusting that you can all wrestle with the implications for yourselves.

Here's another study showing the power of expectation and suggestion:

"In one experiment, 47.5 percent of asthmatics who were exposed to nebulized saline solution (normally innocuous) and told that they were inhaling irritants or allergens experienced substantially increased airway resistance. . . . The twelve subjects who developed full-blown attacks were relieved by the same saline solution when it was presented therapeutically. Controls who did not have asthma were unaffected by exposure to the same stimulus. . . . In a follow-up, double-blind experiment . . . two groups were given a bronchodilator, the other two a bronchoconstrictor; half of the group given each substance was told that they were being given a bronchodilator, the other half that they were being given a bronchoconstrictor. For each substance administered, misinformation about the substance reduced its physiologic effectiveness by 43 percent (for the bronchodilator) and 49 percent (for the bronchoconstrictor)."

Promising neurobiological connections related to neuropeptides:

"It seems very likely that at least some placebo reactions are mediated by peptides; and the fact that brain cells, immune cells, and other body tissues all share receptors for these peptides hints at the outline of a psychosomatic information network which would allow us to make much more sense of placebo data. . ."

"But it also turns out so far to be the case that these peptide receptors are thickly clustered in those parts of the brain functionally linked to emotions, and are seen least in the cerebral cortex and those centers normally associated with cognition. . ."

An attempt to simplify:

"If one recognizes that brain activity both regulates bodily function and generates conscious experience, much of the confusion about the bodily effects of subjective experience is dispelled."

(of course, that is what a lot of people would want to argue about--that it is the brain which "generates conscious experience")

So the placebo effect may be tied somehow to emotion, and somehow it is *expectancy* which seems to be the key. There is evidence that even in cases where "conditioning produces effects that are contrary to people's expectancies, the effect of expectancy may be powerful enough to reverse the conditioning effect." In other words, expectancy is often even more powerful than conditioning.

The power of expectancy never ceases to amaze. For example, side effects of placebos vary depending on the subjects knowledge of the active drug. To illustrate, actual vomiting is reported as a side effect by 8 percent of subjects given placebo estrogen, but by less than .01 percent of subjects given placebos for other substances.

It should be mentioned again that apparently this power of expectation has limits. There are specific instances where placebos do not appear to work. For example, placebos were not effective in studies related to intra-arterial ambulatory blood pressure, and were not effective with obsessive-compulsive disorder.

Another simplification:

For things placebos DO work for, belief/expectation is required. If you don't believe in them, they don't work. So skeptics are usually "right," but probably more often sick. :-)

The placebo effect puts belief and meaning (which the astute will notice can be cultural) at the center of the therapeutic encounter. There is evidence that the experience of pain can be very much altered by cultural expectations. For example, a multi-nationality study of lower back pain showed that Americans find it much more disabling than the residents of other countries. Studies have shown that patients do appreciably less well than a control group if a physician tells them: "I am not sure what is the matter with you, and I am not sure the treatment will have an effect." At the other end of the spectrum, the elaborate rituals of shamans make admirable use of the placebo effect. A complete model of the placebo effect must include the ways in which the human brain and nervous system are set into motion by the impact of culture.

The Placebo Effect: Why Various Techniques 'Work'