First, do no harm: graded exercise therapy and myalgic encephalomyelitis/chronic fatigue syndrome


Davenport: 0000-0001-5772-7727

Document Type


Publication Title

Annals of Internal Medicine





Publication Date



Myalgic encephalomyelitis (ME)/chronic fatigue syndrome (CFS) is a debilitating multisystem condition affecting more than 1 million adults in the United States.

Purpose: To determine benefits and harms of treatments for adults with ME/CFS and identify future research needs.

Data Sources: MEDLINE, PsycINFO, and Cochrane databases (January 1988 to September 2014); clinical trial registries; reference lists; and manufacturer information.

Study Selection: English-language randomized trials of the effectiveness and adverse effects of ME/CFS treatments.

Data Extraction: Data on participants, study design, analysis, follow-up, and results were extracted and confirmed. Study quality was dual-rated by using prespecified criteria; discrepancies were resolved through consensus.

Data Synthesis: Among 35 treatment trials enrolling participants primarily meeting the 1994 Centers for Disease Control and Prevention and Oxford case definitions of CFS, the immune modulator rintatolimod improved some measures of exercise performance compared with placebo in 2 trials (low strength of evidence). Trials of galantamine, hydrocortisone, IgG, valganciclovir, isoprinosine, fluoxetine, and various complementary medicines were inconclusive (insufficient evidence). Counseling therapies and graded exercise therapy compared with no treatment, relaxation, or support improved fatigue, function, global improvement, and work impairment in some trials; counseling therapies also improved quality of life (low to moderate strength of evidence). Harms were rarely reported across studies (insufficient evidence).

Limitation:Trials were heterogeneous and were limited by size, number, duration, applicability, and methodological quality.

Conclusion: Trials of rintatolimod, counseling therapies, and graded exercise therapy suggest benefit for some patients meeting case definitions for CFS, whereas evidence for other treatments and harms is insufficient. More definitive studies comparing participants meeting different case definitions, including ME, and providing subgroup analysis are needed to fill research gaps.