Evidence guide
What Is a Placebo?
A placebo helps separate the specific effect of an intervention from expectations, attention and natural change.
Not 'nothing'
Placebo groups still experience symptoms changing over time, regression to the mean, background care and the effects of participating in a study.
Blinding
A credible placebo should resemble the intervention enough that participants and staff cannot easily guess assignment.
Limits
Placebos may be unethical when effective treatment would be withheld. In those cases, active-comparator or add-on designs are appropriate.
A practical reading checklist
- Could participants guess their group?
- Did side effects reveal assignment?
- Was effective care withheld?
- Were expectations measured?