712
Predictors of patterns of pain, fatigue and insomnia in the elderly during the first year following a cancer diagnosis
Sharon Kozachik, RN, PhD, Schools of Medicine and Nursing, Johns Hopkins University, 525 North Wolfe St., Baltimore, MD 21205, Karen Bandeen-Roche, PhD, Biostatistics, Johns Hopkins Bloomberg School of Public Health, 615 N. Wolfe Street, Baltimore, MD 21205, and Victoria Mock, PhD, FAAN, School of Nursing, Johns Hopkins University, 525 North Wolfe St., Baltimore, MD 21205.
Pain, fatigue and insomnia (PFI) are among the most prevalent, distressing, and under managed symptoms experienced by cancer patients. Research has demonstrated that PFI co-occur; what remains unclear are the patterning and stability of PFI patterns, and the patient, disease and treatment characteristics that predict PFI patterns over time. Specifically, do patient age, sex, comorbidity, cancer site or stage, or treatment regimen predict patterns of PFI and their changes over time? This study was a secondary analysis of a dataset comprised of 867 elders (46% female) who were newly diagnosed with breast (27%), colorectal (18%), lung (26%) or prostate (29%) cancer and followed at 6-8, 12-16, 24 and 52 weeks following diagnosis. The Johns Hopkins Medicine IRB approved this study. Measures included sociodemographics, comorbidity, and symptoms; medical record audits confirmed cancer and treatment. Descriptive statistics and multi-state transition models using multinomial logistic regression were employed. The typical participant was 72.6 years, Caucasian, married/living with spouse, and reported 7.9 symptoms and 2.7 comorbidities. Attrition numbered 255 (death, n = 88; lost to follow-up, n = 167). Prior PFI pattern was consistently associated with significantly increased risks for subsequent PFI pattern. At observations 1-3, lung cancer, treatment, higher comorbidity with breast cancer, and late stage colorectal cancer were significantly associated with increased risks for PFI patterns. Advancing age was not significantly associated with increased risks for PFI patterns at any observation. PFI co-occurrence declined over time, shifting from 18% to 6% from observations 1 to 4. PFI co-occurrence was associated with significantly increased risks for death or loss to follow-up at observations 2-4. Additionally, elders without PFI reported, on average, 3 other symptoms; elders with PFI co-occurrence reported 9 other symptoms. Among elder cancer patients, PFI co-occurrence is associated with adverse outcomes and should be proactively targeted for intervention.