Intended for healthcare professionals


What causes chronic fatigue syndrome?

BMJ 2004; 329 doi: (Published 21 October 2004) Cite this as: BMJ 2004;329:928
  1. Peter D White (p.d.white{at}, professor of psychological medicine
  1. Barts and the London, Queen Mary School of Medicine and Dentistry, St Bartholomew's Hospital, London EC1A 7BE

    Infections, physical inactivity, and enhanced interoception may all play a part

    Chronic fatigue syndrome, also known as myalgic encephalomyelitis, is an illness of unknown nature and cause, but most medical authorities now accept its existence.13 Research about its cause has been hampered by the absence of a biological marker, the heterogeneous nature of the illness, and difficulties in differentiating cause from effect.2 3 Yet, some progress has been made, particularly when causes are divided into predisposing, triggering, and maintaining factors.

    Women get chronic fatigue syndrome more commonly than men for unknown reasons, although increasing evidence suggests a genetic influence on the illness.1 3 Premorbid mood disorders are replicated risk markers for chronic fatigue syndrome;1 3 the risks may be inflated by shared symptoms or they may be markers for those patients with comorbid mood disorders.1 35 Another replicated premorbid risk marker is increased consulting of a doctor for minor illnesses up to 15 years before diagnosis,w1 w2 suggesting a general vulnerability for either ill health or seeking health care, the latter possibly being mediated by comorbid anxiety.4

    Certain infectious illnesses, such as Epstein-Barr virus, Q fever, and viral meningitis, can trigger chronic fatigue syndrome;w3-w8 common upper respiratory infections do not.w3 w9 Little evidence exists of persistent infection in patients with chronic fatigue syndrome.w10 An immune cause has not been established,6 although preliminary research suggests that immune responses to exercise in these patients may be abnormal.w11 The symptoms of chronic fatigue syndrome are similar to the symptoms and consequent behaviour of people with acute infection.2 w12 A low cortisol level has repeatedly been found to be associated with chronic fatigue syndrome, and recent research suggests this may be secondary to the physical inactivity and sleep disturbance found with long standing chronic fatigue syndrome.7 w13 Illness maintaining factors may include illness beliefs that encourage avoidant coping.w14 w15

    Viner and Hotopf publish in this issue a 30 year cohort study of 16 567 babies born in 1970 (p 941), in which they report childhood predictions of self reported chronic fatigue syndrome and myalgic encephalomyelitis developing in adulthood.8 Their most important findings were that chronic fatigue syndrome was predicted by having a disabling illness in childhood and never or hardly ever doing sport out of school at age 10. Premorbid psychological distress in either the mother or the child did not predict the illness.

    Previous healthcare attendance for ill health does seem to predict corroborated chronic fatigue syndrome.w1 w2 Physical ill health in childhood seems to predispose to medically unexplained ill health in adulthood generally only when the childhood diagnosis is uncertain rather than established.w16 Neither the mother's nor the child's or adolescent's psychological distress predicted chronic fatigue syndrome, a finding that contrasts with most studies of both children and adults developing the illness.w17 w18 Data from Viner and Hotopf's study may be more reliable, having been gathered contemporaneously. We need to learn how certain childhood illnesses predispose people to later chronic fatigue syndrome.

    The novel finding was that 10 year old children who were reported by their mothers to “never or hardly ever” play sport in their spare time had twice the risk of chronic fatigue syndrome in adulthood. In contrast, school sport had no effect; this might be explained by the compulsory nature of some school sport. Although statistically significant being more sedentary was not a strong risk marker, since 84% of those who later developed chronic fatigue syndrome were not sedentary, although relying on self reports of chronic fatigue syndrome might have weakened predictions by including misdiagnoses.

    If the finding was not a chance one, could being sedentary itself predispose people to chronic fatigue syndrome, or is being sedentary a marker for something else? We know that ambulant patients with chronic fatigue syndrome are at least as physically deconditioned as sedentary but healthy people, even though studies have excluded the most physically disabled patients because of the difficulty in measuring their fitness.9 Lack of fitness and inactivity consistently predicted the later development of chronic fatigue syndrome after infectious mononucleosis,5 w19 suggesting that either premorbid or early deconditioning is a risk factor in some patients. Getting fitter is not necessary to feel better after a graded exercise programme,9 so is there another explanation?

    Patients with chronic fatigue syndrome perceive activity as more of an effort than healthy controls and underestimate their cognitive and physical abilities, while being more aware of their internal physiological state, a phenomenon called interoception.1 3 9 10 w2 w15 w20-22 How might this be related to being sedentary or having a disabling illness in childhood? Inactivity increases perception of effort with exercise, through both physiological deconditioning and the related cognitive, emotional, and sleep disturbance from being sedentary.11 w23 w24 This may enhance or sensitise interoception, perhaps in a similar way to that hypothesised in the related disorder of fibromyalgia.w25 The corollary is that this enhanced bodily awareness or interoception may itself cause sedentary behaviour. When an appropriate trigger supervenes in later life, enhanced interoception may predispose some people to chronic fatigue syndrome.5 Trials of prevention are required to test this idea.11 w26 Treatments that “reprogramme” interoception and increased activity, such as graded exercise therapy and cognitive behaviour therapy, seem to help most patients.12


    • Papers p 941

    • Embedded ImageReferences w1 to w26 are on

    • Competing interests PDW does consultancy work for the Department for Work and Pensions and Swiss Re, a re-insurance company.


    1. 1.
    2. 2.
    3. 3.
    4. 4.
    5. 5.
    6. 6.
    7. 7.
    8. 8.
    9. 9.
    10. 10.
    11. 11.
    12. 12.
    View Abstract