TL;DR: The placebo effect is when a person feels subjectively better, such as less pain, anxiety or fatigue, after receiving a neutral intervention like a sugar pill, saline injection or a ritual with no biological action. That feeling is real to the person experiencing it, but reviews do not find that placebo has important clinical effects overall, and there is no evidence that placebo eliminates disease. Therefore all health concerns still need to be examined at a medical facility, not replaced by ritual or belief.
What is the placebo effect?
The placebo effect is the portion of change recorded in a person after they receive a neutral intervention with no pharmacological action, and that portion of change exceeds the level of change in a group that received no intervention at all. In other words, the feeling of relief can come from the person's own expectations, context and beliefs, rather than from the pill or the ritual itself. This is a phenomenon recorded fairly widely in the psychology and experimental medicine literature, but it is also one of the most misunderstood concepts. The point to grasp from the outset: a placebo can affect how a person perceives their condition, but it does not affect the disease process itself inside the body.
The word "placebo" in Latin means "I shall please", according to Benedetti, F. (2013), Placebo and the New Physiology of the Doctor-Patient Relationship, Physiological Reviews 93(3): 1207-1246, doi:10.1152/physrev.00043.2012. In clinical trials, placebos are used to form the control group: one group receives the intervention being tested, another receives something that looks identical but has no active property. Any change in the control group is called the "placebo response", while the difference between that response and the results of the group that received no intervention at all is what is called the "placebo effect". Separating these two concepts is very important, because most of the relief people attribute to placebos may in fact come from entirely different factors.
Why must the placebo response be separated from the placebo effect?
The placebo response bundles several things together: the body recovering on its own over time, the disease progressing in a different direction, errors in trial record-keeping, and the tendency to return to the mean. Regression to the mean is a statistical effect: when a person is measured at a moment when their symptoms are more severe or milder than their usual level, the next measurement will usually return closer to the average, regardless of whether they did anything. If this portion cannot be separated out, it is very easy to wrongly conclude that the ritual or the pill produced something.
What can placebos affect?
Reviews and meta-analyses show that placebos affect subjective and continuous outcomes most clearly, especially the perception of pain and some accompanying symptoms. In other words, it changes how a person reports and perceives their condition, but does not change the disease itself. A 2001 meta-analysis of 130 trials across 40 medical conditions concluded that placebos had no significant effect on objective outcomes, with only possible small benefits in subjective outcomes measured continuously and in the treatment of pain (Hróbjartsson, A. & Gøtzsche, P. C. (2001), Is the placebo powerless? An analysis of clinical trials comparing placebo with no treatment, New England Journal of Medicine 344(21): 1594-1602, doi:10.1056/NEJM200105243442106). A 2010 Cochrane review by Asbjørn Hróbjartsson and Peter C. Gøtzsche also held that the placebo effect appears clearly only on subjective, continuous scales, and in the treatment of pain and related conditions; the authors concluded they "did not find that placebo interventions have important clinical effects in general" (Hróbjartsson, A. & Gøtzsche, P. C. (2010), Placebo interventions for all clinical conditions, Cochrane Database of Systematic Reviews (1): CD003974, doi:10.1002/14651858.CD003974.pub3). This review also noted that effects may appear in patient-reported outcomes, especially pain and nausea, and are very hard to separate from reporting bias; in observer-rated outcomes, the review still recorded a small effect (SMD around -0.13), while in asthma and phobia the signal was uncertain. This interpretation has been challenged, particularly on methodological grounds, so this remains a point under debate among researchers.
On the other side, some recent studies record that placebos are linked to improved motor function in people with Parkinson's, to certain immune and endocrine parameters, to organ functions governed by the autonomic nervous system, and to sports performance. These results should be read cautiously: they show a more complex picture than the conclusion "placebos are nothing", but they do not turn placebos into a method that eliminates disease.
| Outcome group | Does the placebo have a clear effect? | Notes |
|---|---|---|
| Perception of pain, nausea | Yes, recorded in many studies | Subjective outcomes, dependent on how the patient reports |
| Anxiety, fatigue, general feeling | Often recorded | Hard to separate from expectation and context |
| The disease itself, disease progression | No | The 2010 Cochrane review found no important clinical effects in general, with only possible small benefits in subjective outcomes such as pain and nausea |
| Biological parameters such as blood pressure, immune, endocrine | Signals in some studies | Still debated, more data needed |
| Motor function in Parkinson's | Signals in recent studies | Does not mean the disease is eliminated |
Why can a ceremony make a person feel lighter?
A ceremony, a prayer ritual, or a therapy presented with solemnity all create exactly the conditions that the literature describes as strengthening the placebo effect: positive expectation, a solemn context, and the care of the practitioner. Research shows that if the person giving the placebo shows concern, friendliness, empathy, or has high expectations of the result, the placebo has a stronger effect. The manner of use also matters: taking more pills is stronger than taking fewer, capsules are stronger than tablets, and injections are stronger than capsules.
This explains why after a ceremony, a person may feel lighter at heart, less tense, less headachy, or sleep more easily. Those feelings are real to the person experiencing them, and should not be dismissed as if they imagined them. But a clear distinction is needed: the feeling of lightness does not mean that the situation inside the body has been resolved. A tumour does not shrink because of a ceremony. An infection is not eliminated because of belief. A chronic disease does not stop progressing because the patient feels more at ease for a few days.
Does every ritual produce this effect?
There is no basis to say that every ritual produces the placebo effect, and no basis to say that the effect is the same in everyone. The degree of response varies from individual to individual. Some studies suggest that people with an optimistic tendency often show a stronger placebo response, while people with high anxiety levels are more prone to the nocebo effect. Nocebo is the flip side of the story: when a person expects something harmful, they may report negative effects or worsening symptoms, even though what they received is completely inert. Placebos can also cause side effects just like real ones, and withdrawal symptoms have also been recorded after stopping a placebo, for example in the Women's Health Initiative study on menopausal hormone therapy.
What factors make the effect stronger or weaker?
Many factors are recorded as influencing the strength of the placebo effect. A review in JAMA Psychiatry found that in antipsychotic drug trials, the level of change in the placebo group increased significantly from 1960 to 2013, and the authors named several possible causes, including inflated baseline scores and the recruitment of fewer severely ill patients (Rutherford, B. R. et al. (2014), Placebo response in antipsychotic clinical trials: a meta-analysis, JAMA Psychiatry 71(12): 1409-1421, doi:10.1001/jamapsychiatry.2014.1319). Another analysis in the journal Pain in 2015 recorded that the placebo response increased significantly in neuropathic pain trials in the US from 1990 to 2013, possibly related to trials becoming larger and longer. These are observations about trends in trial data, not evidence that rituals or beliefs can replace treatment.
One notable line of research is the open-label placebo, meaning patients are told clearly that what they are receiving is inert. Some trials show that open-label placebos can have positive effects compared with no intervention at all, but a review notes that these trials were conducted with small numbers of participants, so they should be interpreted cautiously until better-controlled trials are available. An updated systematic review and meta-analysis in 2021 based on 11 studies also found an overall significant effect, though slightly smaller, and noted that "research on open-label placebos is still in its infancy" (von Wernsdorff, M. et al. (2021), Effects of open-label placebos in clinical trials: a systematic review and meta-analysis, Scientific Reports 11: 3855, doi:10.1038/s41598-021-83148-6).
Why is measuring the placebo effect difficult?
Measuring the magnitude of the placebo effect is very difficult because many confounding factors are involved. A person may feel better after receiving a placebo because the disease recovers on its own or symptoms change on their own, and this can be ruled out by comparing the placebo group with a no-intervention group. But distinguishing a true placebo effect from response bias, observer bias, and other errors in trial methodology is harder, because a trial comparing placebo with no intervention would not be a blinded trial. It was Hróbjartsson and Gøtzsche themselves who argued that even if placebos had no real effect, people would still record a difference between the placebo group and the no-intervention group due to bias tied to the lack of blinding.
Another way of measuring is the "open or hidden" study design: some patients receive a painkiller and are told they are receiving it, others receive the same drug but do not know. Studies of this kind show that painkillers have a significantly stronger effect when patients know they are receiving the drug. This is fairly clear evidence for the role of expectation, but it is still about the perception of pain, not about eliminating the disease.
The depression case: an example showing the debate continues
In 2008, a meta-analysis led by psychologist Irving Kirsch, based on data submitted to the US Food and Drug Administration, recorded that the placebo group achieved more than 80% of the improvement of the antidepressant group (Kirsch, I. et al. (2008), Initial Severity and Antidepressant Benefits: A Meta-Analysis of Data Submitted to the Food and Drug Administration, PLoS Medicine 5(2): e45, doi:10.1371/journal.pmed.0050045). Specifically, the drug group improved by 9.60 points on the HRSD scale while the placebo group improved by 7.80 points. However, other authors expressed doubt about the method and interpretation, particularly the use of 0.5 as the threshold for effect size. A re-analysis and full recalculation on the same dataset argued that Kirsch's study had "important errors in the calculations"; this group concluded that although a large proportion of the placebo response is due to expectation, that is not true for the active drug (Fountoulakis, K. N. & Möller, H.-J. (2011), Efficacy of antidepressants: a re-analysis and re-interpretation of the Kirsch data, International Journal of Neuropsychopharmacology 14(3): 405-412, doi:10.1017/S1461145710000957). Another meta-analysis recorded that 79% of depressed patients on placebo remained well for 12 weeks after 6 to 8 weeks of successful initial treatment, compared with 93% in the antidepressant group; however in the maintenance phase, the placebo group relapsed significantly more often (Khan, A., Redding, N. & Brown, W. A. (2008), The persistence of the placebo response in antidepressant clinical trials, Journal of Psychiatric Research 42(10): 791-796, doi:10.1016/j.jpsychires.2007.10.004). This is an example showing that the same dataset can lead to different conclusions depending on how the statistics are handled, and readers should be cautious about any overly definitive claim in either direction.
Ethics and limits of using placebos
Deliberately giving a person a placebo when an effective treatment exists is a complex biomedical ethics issue. Placebo-controlled trials can provide information about the efficacy of an intervention, but at the same time deny some patients what may be the best available treatment. Informed consent is usually required for a study to be considered ethical, including informing participants that some will receive a placebo. In clinical practice, using a placebo without the patient's knowledge also raises issues about honesty in the doctor-patient relationship, and about bypassing informed consent.
What to remember when evaluating an experience for yourself
The feeling of lightness after a ceremony, a prayer session, or a psychological therapy is an experience that deserves respect, and the placebo literature shows it can have a real psychological basis. But at the same time, a clear boundary must be kept: that feeling is not evidence that the ritual affected the situation unfolding inside the body. If a person feels better mentally, that can be a real benefit to quality of life, and there is no need to deny it. If a person believes they have eliminated a disease thanks to a ritual or belief, that is the moment to check again with objective medical data.
Every health problem, whether mild or severe, needs to be examined and assessed at a medical facility by a qualified person. Rituals, beliefs and spiritual practices can accompany treatment, can help a person feel calmer, but cannot replace diagnosis and treatment. One should not delay going for an examination because of believing that a ritual will handle the problem. And if a person is taking medication or following a treatment regimen, they should not stop or change it on their own because they feel lighter after a psychological experience.
