Cancer · 2017 · 49 citations · 36 references
MedicineGeriatric MedicineOncologyGeriatric AssessmentHealth AssessmentMental Health In Older AdultsComprehensive Geriatric AssessmentGeriatric OncologyHealth-related Quality Of LifeHealth-related Quality Of Life MeasurementQuality Of LifeSupportive Care In CancerCancer-related Quality Of LifeEmotional NeedsGeriatric Assessment ToolsLinear ModelsQuality Of Life EndpointsRandom ForestsShort Form Health SurveySymptom Assessment Tools
BACKGROUND As the population of older adults with cancer continues to grow, the most important factors contributing to their health‐related quality of life (HRQOL) remain unclear. METHODS A total of 1457 older adults (aged ≥65 years) with cancer participated in a telephone survey. Outcomes were measured using the Physical Component Summary (PCS) and Mental Component Summary (MCS) scores of the 12‐Item Short Form Survey (SF‐12) from the Medical Outcomes Study (version 2). Statistical techniques used to identify factors in 4 domains (physical, psychological, social, and spiritual) most strongly associated with HRQOL included linear models, recursive partitioning, and random forests. Models were developed in a training data set (920 respondents) and performance was assessed in a validation data set (537 respondents). RESULTS Respondents were a median of 19 months from diagnosis, and 28.1% were receiving active treatment. The most relevant factors found to be associated with PCS were symptom severity, comorbidity scores, leisure‐time physical activity, and having physical support needs. The most relevant factors for MCS were having emotional support needs, symptom severity score, and the number of financial hardship events. Results were consistent across modeling techniques. Symptoms found to be strongly associated with PCS included fatigue (adjusted proportion of summary score's variance [R 2 ] = 0.34), pain (adjusted R 2 = 0.32), disturbed sleep (adjusted R 2 = 0.16), and drowsiness (adjusted R 2 = 0.16). Symptoms found to be strongly associated with MCS included fatigue (adjusted R 2 = 0.23), problems remembering things (adjusted R 2 = 0.17), disturbed sleep (adjusted R 2 = 0.16), and lack of appetite (adjusted R 2 = 0.16). CONCLUSIONS The findings of the current study support the importance of addressing persistent symptoms, managing comorbidities, promoting leisure‐time physical activity, and addressing financial challenges. A long‐term comprehensive approach is needed to ensure the well‐being of older adults with cancer. Cancer 2018;124:596‐605 . © 2017 American Cancer Society .
36
Statistical Power Analysis for the Behavioral Sciences
Keith E. Muller, Jacob Cohen · Technometrics · 1989 · 83.9K citations
A 12-Item Short-Form Health Survey
John E. Ware, Mark Kosinski, Susan Keller · Medical Care · 1996 · 17.1K citations
Cathy D. Sherbourne, Anita L. Stewart · Social Science & Medicine · 1991 · 6.6K citations