Placebo effect

What is the placebo effect?

The classic article “The Powerful Placebo” by Henry Beecher highlighted the placebo effect and emphasised a need to account for it to evaluate the efficacy of a treatment modality properly. ¹

The placebo effect is a poorly understood phenomenon but recent work suggests that placebo represents the psychosocial aspect of every treatment and "the study of placebo is essentially the study of psychosocial context that surrounds the patient".²

Research in the field of neurobiology suggests that a person's expectation of an effect generates activity in certain areas of the brain. Brain regions identified as having activity after administration of a placebo include the anterior cingulate cortex, dorsolateral prefrontal cortex and basal ganglia.³

The placebo effect needs to be taken into account in clinical trials:

• It has been reported to affect quality-of-life assessments.⁴

• In certain settings placebo interventions can influence patient-reported outcomes, especially pain and nausea, although it is difficult to distinguish patient-reported effects of placebo from biased reporting. Variations in the effect of placebo were partly explained by variations in how trials were conducted and how patients were informed.⁵

Perceived placebo effect vs true placebo effect

It is important therefore to differentiate between the placebo effect exhibited in clinical trials (which needs to be minimised) and that seen in clinical practice (which should be maximised).⁶ The difference between a perceived placebo effect and a true placebo effect should be recognised.

Perceived placebo effect

This is seen in the placebo arm of a clinical trial, eg, new antihypertensive drugs may drop the average blood pressure by 5 mm Hg, but this may not be true placebo effect as other factors may confound the situation. Such factors include:

• The natural course of the disease - blood pressure may normally reduce over time, thus exaggerating the placebo effect.

• Regression towards the mean - biological variables often fluctuate.⁷ When first measured they are probably approaching their maximum and so further measurements are likely to show a reduction.

• Increased skill of the investigator - this may alter measurements up or down as more measurements are made.

• Variable factors within the same patient - white coat hypertension may reduce as patients become accustomed to having their blood pressure measured, which would increase the apparent placebo effect.

• Non-apparent simultaneous changes - on entering the trial, the patient may change their behaviour, on purpose or otherwise, in such a way as to confound the outcome, eg, eat less salt, take more exercise.

True placebo effect

This can only be studied if an untreated group is included along with active and placebo treated groups. These are relatively uncommon but it has been discovered that:

• Placebo treatment is more effective in relieving pain compared with no treatment. To achieve this, patients need to be conscious (placebo was given to sleeping patients and no difference noted).

• Physical placebos, eg, invasive techniques such as injection, are more powerful than simple oral placebos.⁸

• Topical placebo is also more effective than oral placebo, eg, in primary varicose veins.

• There is no standard degree of placebo effect (approximately 30% is often given) and its extent depends upon numerous factors including:
  

  • The demeanour of the person offering treatment.

  • The patient's attitude to health and their feelings about the treatment and person offering it.

  • The suggestibility of the patient.

  • The form of treatment, eg, whether it has worked before, how expensive it is, its invasiveness and the reasons given as to how it works.

• Placebo effects on pain are generally greater than on other symptoms.

• Placebo effects are not always useful and may produce adverse effects (nocebo effects).

• Variations in placebo remission and response rates reported in randomised controlled trials can be significantly influenced by the country in which the trial is conducted.⁹

• A study of osteoarthritis patients elicited that the placebo effect was effective in reducing pain and stiffness and improving function. The pain-relieving effect increased when the active treatment effect, baseline pain and sample size increased and when placebo was given through injections/needles.¹⁰

• One study found no difference in placebo effect whether the patient was suffering from functional or organic gastrointestinal disease.¹¹

• A study of migraine patients found that placebos were more effective in treating acute headache than in prophylaxis.¹²