The placebo response in the treatment of chronic fatigue syndrome: a systematic review and meta-analysis.
Cho, Hyong Jin, Hotopf, Matthew, Wessely, Simon · Psychosomatic medicine · 2005 · DOI
Quick Summary
This study looked at how often ME/CFS patients improve simply from believing they will improve (the placebo effect) in clinical trials. Researchers found that the placebo response in ME/CFS is actually lower than many people assume—only about 20% of patients improved with placebo. Interestingly, patients were less likely to show placebo improvement when the treatment was psychological or psychiatric in nature, possibly because many ME/CFS patients believe their illness is physical rather than mental.
Why It Matters
This study corrects a widespread misconception that ME/CFS treatment responses are primarily due to placebo effects, which has sometimes been used to dismiss patient symptoms as psychosomatic. Understanding the actual magnitude and determinants of placebo response helps distinguish genuine therapeutic effects from expectancy-driven improvement, strengthening the scientific basis for evaluating future ME/CFS treatments. The finding that patient beliefs about intervention type influence placebo response also has implications for trial design, informed consent, and communication strategies in ME/CFS research.
Observed Findings
Pooled placebo response across ME/CFS trials was 19.6% (95% CI: 15.4-23.7%), substantially lower than conventional assumptions.
Intervention type significantly contributed to heterogeneity in placebo response (p=0.03).
Psychological-psychiatric interventions showed lower placebo responses compared to other intervention types.
The placebo response in ME/CFS was lower than in some other medical conditions.
Patient expectations about causation (physical vs. psychological attribution) appeared to influence response rates.
Inferred Conclusions
The placebo response in ME/CFS is lower than traditionally believed and lower than in certain other medical conditions, contradicting conventional wisdom.
Patient expectations regarding intervention type and perceived rationale meaningfully influence the magnitude of placebo response.
The subjective nature and unclear pathogenesis of ME/CFS do not inherently produce high placebo responses as previously assumed.
Psychological interventions may face higher skepticism from ME/CFS patients with strong physical illness attributions, reducing expectancy-driven improvement.
Remaining Questions
Does the 19.6% placebo response vary significantly based on patient demographics, disease severity, or clinical subtype?
What This Study Does Not Prove
This meta-analysis does not prove that ME/CFS is primarily physical or psychiatric in nature—it only shows that patients' preexisting beliefs about causation influence their placebo response. The study also does not establish that the 20% placebo response is uniform across all ME/CFS patients or trial designs, as significant heterogeneity existed. Finally, a low placebo response does not indicate that psychological interventions are ineffective; it only suggests that improvement in psychological trials is less driven by expectancy alone.
About the PEM badge: “PEM required” means post-exertional malaise was an explicit required diagnostic criterion for participant inclusion in this study — not that PEM was studied, observed, or discussed. Studies using criteria that do not require PEM (e.g. Fukuda, Oxford) are tagged “PEM not required”. How the atlas works →
What specific mechanisms within psychological-psychiatric interventions reduce placebo response—is it primarily due to patient skepticism, attribution beliefs, or other factors?
How do these placebo response rates compare across different types of control or sham conditions used in ME/CFS trials?
Do interventions with explicitly physical or biomedical rationales consistently show higher placebo responses than those framed psychologically?