Symptom clusters in elderly cancer patients

Statement of the problem. Cancer and treatment-related symptoms are a significant issue for cancer patients. Pain, fatigue and insomnia are particularly problematic, yet not all cancer patients experience those symptoms. Symptoms rarely occur in isolation, yet little is known about the nature of co-occurring symptoms over time.; Methods. A secondary analysis was conducted to examine, in a sample of elders who were newly diagnosed with breast, colorectal, lung or prostate cancer, the predictors for pain, fatigue and insomnia (PFI) patterns at 6, 12, 24, and 52 weeks following a cancer diagnosis; to explore the nature of symptom clustering across time; and to compare symptom clustering at the 24-week wave of data collection by site of cancer and by stage of disease. Multi-state transition models and hierarchical agglomerated cluster analysis were conducted. This secondary analysis was funded by a National Research Service Award through the National Institutes of Nursing Research, grant # F31-NR08959, and was approved by the Johns Hopkins Medicine, Institutional Review Board-X, NA-00000830.; Results. Pain, fatigue and insomnia (PFI) were among the ten most frequently reported symptoms over time, with elders moved from reporting PFI co-occurrence to lesser PFI patterns. At 6 weeks, treatment, lung cancer and higher comorbidity among breast cancer were associated with significantly increased risks for PFI patterns, with the exception of radiation which had a reduced risk. Over time, prior PFI patterning was most consistently associated with significantly increased risks for ongoing reports of PFI patterns. Based on the PFI pattern reported, elders reported between 3 and 9 additional symptoms. At 6 weeks, four different clusters emerged and over time, the number of clusters decreased to one; 15 symptoms were common to a cluster over time. Clusters varied by site of cancer and stage of disease, with 16 and 11 common symptoms to a cluster, respectively.; Conclusions. Among elderly cancer patients, under managed symptoms are a significant issue. Health care providers are called upon to assess for a variety of symptoms, based upon the patient, disease and treatment characteristics. Targeting interventions specifically toward PFI may ameliorate other symptoms. Additional symptom clustering research is needed to elucidate a possible underlying biological link.