Endgames Statistical Question

The nocebo effect

BMJ 2013; 347 doi: http://dx.doi.org/10.1136/bmj.f6130 (Published 11 October 2013) Cite this as: BMJ 2013;347:f6130
  1. Philip Sedgwick, reader in medical statistics and medical education
  1. 1Centre for Medical and Healthcare Education, St George’s, University of London, London, UK
  1. p.sedgwick{at}sgul.ac.uk

Researchers investigated whether a sham device (validated sham acupuncture needle) and an inert pill exerted a similar placebo effect in patients with persistent arm pain. A single blind randomised controlled trial study design was used. The study was created from the placebo run-in periods for two randomised placebo controlled trials nested within a larger study, the aim of which was to investigate the effectiveness of acupuncture and amitriptyline in relieving arm pain. One trial compared acupuncture (twice a week) with a validated acupuncture sham device for six weeks, whereas the other compared amitriptyline (25 mg once a day) with placebo for eight weeks. Both trials had a placebo run-in period of two weeks. The primary investigation in this study was the comparison of the sham device with placebo pill during the placebo run-in periods.1

Participants were 266 adults with arm pain caused by repetitive use, which had lasted at least three months despite treatment, who scored three or more on a 10 point pain scale. Trial participants were randomised to the acupuncture (n=133) or amitriptyline (n=133) trial. The primary outcome measure was arm pain measured on a 10 point pain scale. At the end of the placebo run-in period, pain had decreased from baseline in both the sham device and placebo pill groups, but the change was not significantly different between groups. Nocebo effects were reported for the sham acupuncture needle and amitriptyline placebo groups.

Which one of the following best describes the nocebo effect in relation to the administration of a placebo or sham treatment?

  • a) No report of adverse side effects

  • b) No report of adverse side effects and no change in symptoms

  • c) Report of adverse side effects or worsening of symptoms

  • d) Significant improvement in symptoms

Answers

Statement c best describes the nocebo effect in relation to the administration of a placebo or sham treatment.

The researchers investigated whether a sham device (sham acupuncture needle) and a placebo (inert pill) exerted a similar placebo effect in patients with persistent arm pain. The study was created from the placebo run-in periods for two randomised controlled trials. One trial compared acupuncture with an acupuncture sham device for six weeks and the other compared amitriptyline with placebo for eight weeks. After recruitment, participants were randomised to one of the trials. The placebo run-in period, described in a previous question,2 lasted for two weeks in both trials. It preceded the randomisation of participants to active treatment (amitriptyline or acupuncture) or placebo (inert pill or shame device) within their allocated trial.

During the placebo run-in period all participants received the inert pill or sham device, depending on which trial they had been allocated to. However, trial participants were not informed of this. During informed consent, potential participants were informed they had a probability of 0.5 of receiving inactive treatment for the entire study and a probability of 0.5 of receiving active treatment at some time during the study. The run-in period is used to screen for those participants who may not adhere to the subsequent allocated treatment regimen. It therefore avoids randomising such participants to treatment groups, minimising the possibility of confounding arising because the balance in baseline characteristics achieved after randomisation is not maintained.

The placebo effect, sometimes referred to as the placebo response, has been described in a previous question.3 It is the phenomenon by which a trial participant shows an improvement (positive response) after the administration of a placebo or sham treatment. Placebos and sham treatments, described in a previous question,4 are obviously not expected to affect a participant’s outcome. The placebo effect is a component of the patient’s response to treatment not directly ascribed to provision of an active treatment. It comprises the non-specific treatment effects resulting from the patient’s response to investigation, including the response to a therapeutic ritual, subsequent response to observation and assessment, and response to the patient-doctor interaction. A complex association exists between these components.

In the above study, participants in the sham acupuncture and amitriptyline placebo groups reported nocebo effects during the placebo run-in periods. The nocebo effect, sometimes described as the nocebo response, is the phenomenon of a worsening of symptoms or side effects after administration of a placebo or sham treatment (statement c). The most commonly reported side effects in the sham acupuncture group were pain during treatment (n=19), increased pain after “removing” the needle (12), and redness or swelling (n=4). The most commonly reported side effects in the amitriptyline placebo group were drowsiness (n=25), dry mouth (n=23), and restlessness (n=9).

A placebo or sham treatment that exerts nocebo effects may be referred to as a nocebo. Participants in both groups reported side effects in the above study. However, both groups experienced an average improvement in their primary outcome. Therefore, it seems appropriate to refer to the intervention as a placebo rather than a nocebo.

The term nocebo (Latin for “I shall harm”) was chosen to indicate the opposite of the term placebo (Latin for “I shall please”). It is acknowledged that placebo and nocebo effects are psychogenic, originating from mental or emotional processes rather than physiological ones. The placebo effect arises from a participant’s positive expectations of a treatment or procedure, although in a trial participants are aware they may receive a placebo (or sham treatment) and not an active treatment. Similarly, nocebo effects are thought to be related to the negative expectations of worsening symptoms or adverse side effects. In the above trial, the researchers suggested that the reported side effects were totally different in the two study groups and clearly mimicked the information given at informed consent about potential risks of acupuncture and amitriptyline. The participants’ expectations and beliefs were therefore thought to have led to the reporting of nocebo effects.

Notes

Cite this as: BMJ 2013;347:f6130

Footnotes

  • Competing interests: None declared.

References