Goals of work: Cancer patients often experience multiple symptoms, many of which have been reported to correlate with each other. The goals of this study were to understand which cancer-related symptoms cluster together and to test the conceptual meanings of the revealed symptom clusters.

Patients and methods: Patients with various cancer diagnoses (N=151) were recruited from a medical center in northern Taiwan. The 13-item M.D. Anderson symptom inventory was used to assess patients' symptom severity. Selected symptoms were factored using principal-axis factoring with oblique rotation. The known-group technique was used to validate the conceptual meanings of revealed factors.

Main results: Patients' symptom severity ratings fit a three-factor solution that explained 55% of the variance. These three factors (symptom clusters) were named sickness symptom cluster, gastrointestinal symptom cluster, and emotional symptom cluster. Patients with pain and with advanced diseases had significantly higher mean scores in the sickness symptom cluster than patients without pain and with earlier-stage diseases. The patients' functional status was negatively correlated with mean scores in the sickness symptom cluster. Patients under chemotherapy demonstrated significantly higher mean scores in the gastrointestinal symptom cluster than patients under other treatments. Patients with anxiety or depression also had significantly higher mean scores in the emotional symptom cluster than patients without anxiety or depression.

Conclusions: This study identified three underlying symptom clusters and verified their conceptual meaning in cancer patients. Knowing these symptom clusters may help healthcare professionals understand plausible mechanisms for the aggregation of symptoms.