Uncontrolled studies have reported that fatigue is a common symptom among patients with advanced cancer. It is also a frequent complaint among the general population. Simply asking cancer patients whether or not they feel fatigued does not distinguish between the 'background' level of this symptom in the community and any 'excess' arising as a result of illness. The aim of this study was to determine the prevalence of fatigue among palliative care inpatients in comparison with a control group of age and sex-matched volunteers without cancer. In addition, the correlates of fatigue were investigated. The prevalence of 'severe subjective fatigue' (defined as fatigue greater than that experienced by 95% of the control group) was found to be 75%. Patients were malnourished, had diminished muscle function and were suffering from a number of physical and mental symptoms. The severity of fatigue was unrelated to age, sex, diagnosis, presence or site of metastases, anaemia, dose of opioid or steroid, any of the haematological or biochemical indices (except urea), nutritional status, voluntary muscle function, or mood. A multivariate analysis found that fatigue severity was significantly associated with pain and dypnoea scores in the patients, and with the symptoms of anxiety and depression in the controls. The authors conclude that subjective fatigue is both prevalent and severe among patients with advanced cancer. The causes of this symptom remain obscure. Further work is required in order to determine if the associations reported between fatigue and pain and between fatigue and dyspnoea are causal or coincidental.