Home > DARE Reviews > The placebo response in the treatment of...

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.

Database of Abstracts of Reviews of Effects (DARE): Quality-assessed Reviews [Internet]. York (UK): Centre for Reviews and Dissemination (UK); 1995-.

Database of Abstracts of Reviews of Effects (DARE): Quality-assessed Reviews [Internet].

The placebo response in the treatment of chronic fatigue syndrome: a systematic review and meta-analysis

HJ Cho, M Hotopf, and S Wessely.

Review published: 2005.

CRD summary

This review investigated the placebo response in the treatment of chronic fatigue syndrome (CFS) and the influence of intervention type. The authors found that the placebo response in the treatment of CFS is low, and is characterised by a lower response to psychological-psychiatric interventions. The authors' conclusions accurately reflect the evidence presented, but some methodological limitations potentially affect their reliability.

Authors' objectives

To investigate the placebo response in the treatment of chronic fatigue syndrome (CFS) and to determine whether this is dependent on intervention type.


A previous review (see Other Publications of Related Interest) provided the initial source of references for this review and was reported to include database searches (inception to July 2000). Following this, the authors used a similar search strategy to search MEDLINE, EMBASE, the Cochrane Library and PsycINFO from January 2000 to August 2002 to identify further studies; the search terms were reported. The search was subsequently updated to December 2002 using PubMed. Bibliographies of the included studies were screened for additional references and experts were contacted for unpublished data. Studies reported in any language were included.

Study selection

Study designs of evaluations included in the review

Placebo-controlled randomised controlled trials (RCTs) and controlled clinical trials (CCTs) were eligible. The majority of the included trials were RCTs. The duration of follow-up ranged from 3 to 61 weeks (median 13).

Specific interventions included in the review

Studies of any intervention aimed at treating CFS compared with placebo were eligible. Placebo was defined as 'any therapeutic procedure which has an effect on a patient, symptom, syndrome or disease, but which is objectively without specific activity for the condition being treated'. In the analysis, interventions were classified according to the hypothesised degree of placebo response (i.e. low, medium or high) with which they were believed to be associated. Interventions based on infectious or immunological assumptions, or alternative therapies, were deemed to have a high placebo effect; interventions based on psychological or psychiatric assumptions, a low effect; and interventions with an obscure or neutral theory base (e.g. neuroendocrinological agents), a medium effect.

Participants included in the review

Adults and children diagnosed with CFS or any similar condition (e.g. myalgic encephalomyelitis, chronic fatigue immune deficiency syndrome, or chronic mononucleosis) based on any criteria were eligible. Studies focusing only on fibromyalgia were excluded. Patients from primary, secondary and tertiary care, along with those recruited through advertisements and from patient organisations, were included. The majority of the participants were female (mean age 38 years) with poor baseline functioning due to a perceived physical cause, and a mean illness duration of 62 months.

Outcomes assessed in the review

The primary outcomes of interest were categorised as either physical (e.g. fatigue, pain, sleep and functional ability) or general (e.g. quality of life, well-being, clinical improvement, and overall symptom measure) outcomes using a binary measure (e.g. improved or not, response or non-response). The majority of the included studies reported general outcomes, while all categorised the data into response and non-response. Given the heterogeneity of the measurement systems, these criteria were classified as strict or loose for the purposes of data extraction.

How were decisions on the relevance of primary studies made?

The authors did not state how the papers were selected for the review, or how many reviewers performed the selection.

Assessment of study quality

The validity of the primary studies was assessed using a modified version of CRD Report 4 (2nd ed.). The following criteria were adopted for the assessment of RCTs: method of randomisation; allocation concealment; blinding; baseline comparability of the groups; completeness of follow-up; drop-out analysis; intention-to-treat (ITT); objectivity of outcome assessment; appropriateness of statistical analysis; sample size; equality in how the groups were treated; description of placebo type, placebo group and placebo response. The first two criteria were adapted for the assessment of CCTs. A scoring system was used, with a maximum of 22 points attainable. One reviewer assessed the studies and a second reviewer checked them.

Data extraction

One reviewer extracted the data and a second reviewer checked them. The data were analysed on an ITT basis. Clinical improvement data were extracted in order to calculate the percentage placebo response (number of placebo responders divided by number of participants assigned to placebo), along with the 95% confidence interval (CI). If only data reporting the number of study completers were reported, it was assumed that only completers had responded.

Methods of synthesis

How were the studies combined?

The studies were combined in a meta-analysis using a random-effects model. Publication bias was not assessed.

How were differences between studies investigated?

Pre-planned subgroup analyses by intervention type and meta-regression analyses dependent on rate of placebo response were used to investigate potential sources of heterogeneity.

Results of the review

Twenty-nine trials were included in the review: 28 RCTs (n=1,002), six of which were crossover trials, and one CCT (n=14). The total number of placebo arm participants was 1,016 (median 32, range: 12 to 94).

The validity of the studies varied, with scores ranging from 7 to 21 (median 17).

Eight trials were of interventions hypothesised to have a low placebo effect, five were classified as having a medium effect, and 16 a high effect. Five trials used behavioural placebos or standardised medical care; sixteen used oral and eight used injection-based placebos.

The pooled placebo response was 19.6% (95% CI: 15.4, 23.7), but there was considerable heterogeneity amongst the trials (P<0.001). From the factors investigated, only intervention type contributed significantly to the observed heterogeneity (P=0.03). The subgroup analysis revealed an upward trend of placebo response between the groups (a 5% increase according to effect level). The low and medium effect groups had placebo responses of 14.0% (95% CI: 8.0, 19.9) and 16.5% (95% CI: 5.7, 27.4), respectively. The high effect group had the highest placebo response (24.0%, 95% CI: 18.9, 29.1).

Authors' conclusions

The placebo response in CFS treatment is low. In particular, this is characterised by a lower response to psychological-psychiatric interventions. A possible link to patient expectation theory is feasible.

CRD commentary

The review question and the inclusion criteria were explicit and clear. The search strategy represented an update of a previous review, which the authors used as a major source of reference. The updated strategy was comprehensive and included both published and unpublished literature. Publication bias was not formally assessed, but language bias was addressed. It was unclear how studies were selected for inclusion in the review, and this may be a potential source of bias. However, detailed information on the included trials was supplied; appropriate methods were used to assess their validity and the studies were, on average, reported to be of reasonably good quality. The authors acknowledged the uncertainties of pooling heterogeneous data in their meta-analysis, although this was placed in the context of comparison with other similarly conducted studies and differences were extensively explored in the subgroup analysis. This was a thorough review and the authors' conclusions reflect the evidence presented. However, the reliability of these conclusions may be hindered by the methodological limitations identified.

Implications of the review for practice and research

Practice: The authors stated that more focus is required on the non-specific, contextual aspects of CFS treatment in order to increase the effect of an active treatment. The collaborative therapeutic relationship was suggested as a key factor in the management of the condition.

Research: The authors did not state any implications for further research.

Bibliographic details

Cho H J, Hotopf M, Wessely S. The placebo response in the treatment of chronic fatigue syndrome: a systematic review and meta-analysis. Psychosomatic Medicine 2005; 67(2): 301-313. [PubMed: 15784798]

Other publications of related interest

Whiting P, Bagnall AM, Sowden AJ, Cornell JE, Mulrow CD, Ramirez G. Interventions for the treatment and management of chronic fatigue syndrome: a systematic review. JAMA 2001;286:1360-8.

Indexing Status

Subject indexing assigned by NLM


Adult; Child; Cognitive Therapy; Confidence Intervals; Controlled Clinical Trials as Topic; Exercise Therapy; Fatigue Syndrome, Chronic /psychology /therapy; Humans; Placebo Effect; Regression Analysis; Treatment Outcome



Database entry date


Record Status

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.

CRD has determined that this article meets the DARE scientific quality criteria for a systematic review.

Copyright © 2014 University of York.

PMID: 15784798


PubMed Health Blog...

read all...

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...