Placebo effects are largely misunderstood, even by professionals, and this leads to a lot of sloppy thinking about potential treatments. This problem has been exacerbated by the alternative medicine phenomenon.

Several decades ago, the proponents of so-called CAM promised that if only their preferred if unconventional treatments were properly tested medical science would discover how effective they are. “Effective” (or more precisely, “efficacy”) has a specific definition in medical science – it means that a treatment has been found to perform statistically significantly better than placebo in a blinded controlled trial. Several decades and thousands of studies later, the most popular CAM modalities (homeopathy, acupuncture, reiki, manipulation for medical indications, and more) have been shown to be no more effective than placebo. This means they don’t work.

Not to be deterred by reality, CAM proponents simply shifted the goal posts. Now many of them are saying that placebo effects are real, and therefore being as effective as placebo means that their treatments “work.” As part of this strategy they have promoted and amplified common myths about placebo effects. Let’s take a closer look at these myths and show why they are wrong.

Myth #1 – “The” placebo effect

The first and overriding myth about placebos is that there is one placebo effect (singular). This confusion is understandable, because scientists often refer to “the” placebo effect. However, they are referring to what is measured in the placebo arm of a clinical trial – that net effect (the difference between baseline or no treatment at all and a placebo treatment) is the placebo effect for that study.

There are multiple placebo effects contributing to that difference, however. Anything that might give the appearance of an improvement will contribute to the measured placebo effect. These placebo effects include: Regression to the mean – when symptoms flare, they are likely to return to baseline on their own. If you take any illness that fluctuates in severity, any treatment you take when your symptoms are at their peak is likely by chance alone to be followed by a period of less intense symptoms.

Similar to this but distinct is the reality that many illnesses are self-limiting. If you have a cold, you will likely get better even if you do nothing – so anything you do will be followed by improvement. There is also bias in perceiving and reporting subjective symptoms. People want to feel better, they want to think that the treatment is working, and they may want to please the researcher or their physician. Further, researchers and doctors want their treatments to work.

There are also many possible non-specific effects just from the act of being treated. Hope can be a very positive emotion, and that alone may make people subjectively feel better. Subjects in a trial are also getting medical attention, and are likely paying more attention to their own health. They are likely to be more compliant with other treatments.

The treatment under study itself may have several components, some specific and some non-specific. Do people sometimes feel better after a session of reiki or acupuncture because they were laying down listening to music and smelling incense during the treatment? How much of a relaxation effect is at play? Does it matter if you actually stick the needles in alleged acupuncture points (the answer is no)?

Myth #2 – Placebo effects can cause healing

Because it is often believed that “the” placebo effect is one thing, that one thing is often believed to be a real mind-over-matter physical healing. There is no evidence to support this interpretation, however. In fact researchers looking for that real healing effect of placebos have only demonstrated that it doesn’t exist.

Part of the problem here is that the term “healing” is vague. It does not have a specific definition, but the implication is that biological repair is taking place. In practice researchers distinguish objective vs subjective markers of improvement. Subjective just means that the patient feels better in some way, per their own report. They rate their own pain, for example. An objective outcome is something measurable, like blood pressure, survival, or tumor burden.

A systematic review of cancer research, for example, found that placebo interventions resulted in minor improvements in subjective symptoms, but no improvement in the cancer itself.

Placebo effects break down into several categories. One category is illusory – the misperception of improvement through regression to the mean or biased reporting. The second category is non-specific effects, such as emotional comfort from a practitioner, relaxation, or improved self-care or compliance. This third category is comprised of effects which can plausibly result from psychological interventions only. These relate mainly to stress, depression, anxiety, and the perception of pain and similar subjective symptoms. There is a mind-body connection – it’s called the brain.

There is, however, no magical control of your brain over biological or physiological processes that are not networked with the brain through nerves or hormones.

Myth #3 – Animals and babies cannot have a placebo effect

This myth results from the false assumption that in order to have a placebo effect you need to believe that you are taking an active treatment. It is the belief that is causing the effect, and therefore it is a prerequisite. The logic then follows that animals and babies, who cannot know they are receiving a treatment, can therefore not have a placebo effect. Any improvement in this context, therefore, must be a physiological response to the treatment itself.

It should already be obvious, however, that these assumptions are incorrect. There are many sources of placebo effects that do not depend upon the subject knowing they are being treated, such as regression to the mean, the self-limiting nature of many ailments, and non-specific effects or benefits from simultaneous interventions.

Further, however, someone has to determine that the animal or baby has improved. That person is vulnerable to biased perception and reporting, and will also contribute to any measured effect.

This means that studies of treatments in animals or babies still need to be properly controlled, and whoever is assessing the outcome needs to be properly blinded to treatment allocation.

Myth #4 – Fanciful or alternative treatments yield better placebo effects

Desperate to salvage a role for their preferred but ineffective treatments, many alternative practitioners will argue that their real expertise is in maximizing placebo effects. OK, sure, the scientific evidence shows that my treatment is no better than placebo, but placebo effects are real, and I am very good at eliciting them. This is the “placebo medicine” gambit.

I have already debunked the first part of that claim. There is also no evidence for the second part, that alternative practitioners elicit more of a placebo effect. What the scientific evidence shows is that all interventions will produce some placebo effect, depending mainly on the outcome to be followed. The more subjective and amenable to variables such as mood, the larger the measured effect will be.

The existence of a placebo effect does not justify using inactive or pseudoscientific treatments. You can elicit the same effects from science-based interventions. Related to this is the notion of placebo effects without deception. This is certainly possible, if you include all the non-specific and statistical effects, but most patients would likely not be happy to be receiving a treatment that they were told was completely inert, just so it may bias their perception of their symptoms. All pseudoscientific treatments, even if they are justified through placebo effects, are given with a generous helping of deception, which violates patient autonomy.

The other variable that seems to be important, but requires further study, is the therapeutic relationship between practitioner and patient. Having a positive relationship may enhance the measured placebo effect, but that may be just another measure of bias.

In any case, anything useful about placebo effects can be had with a positive therapeutic relationship, using science-based interventions, and following the ethical requirements of informed consent and patient autonomy.


Posted by Steven Novella

Founder and currently Executive Editor of Science-Based Medicine Steven Novella, MD is an academic clinical neurologist at the Yale University School of Medicine. He is also the host and producer of the popular weekly science podcast, The Skeptics’ Guide to the Universe, and the author of the NeuroLogicaBlog, a daily blog that covers news and issues in neuroscience, but also general science, scientific skepticism, philosophy of science, critical thinking, and the intersection of science with the media and society. Dr. Novella also has produced two courses with The Great Courses, and published a book on critical thinking - also called The Skeptics Guide to the Universe.