Predictors of health-related quality of life in patients with colorectal cancer

Abstract

Background: Most studies that have identified variables associated with the health-related quality of life (HRQL) of patients with colorectal cancer have been cross-sectional or included patients with other diagnoses. The objectives of this study were to identify predictors of HRQL in patients with colorectal cancer and interpret the clinical importance of the results. Methods: We conducted a population-based longitudinal study of patients identified through three regions of the California Cancer Registry. Surveys were completed by 568 patients approximately 9 and 19 months post-diagnosis. Three HRQL outcomes from the Functional Assessment of Cancer Therapy - Colorectal (FACT-C) were evaluated: social/family well-being (SWB), emotional well-being (EWB) and the Trial Outcome Index (TOI), which is a colorectal cancer-specific summary measure of physical function and well-being. Sociodemographic, cancer/health, and healthcare variables were assessed in multivariable regression models. We computed the difference in predicted HRQL scores corresponding to a large difference in a predictor variable, defined as a 1 standard deviation difference for interval variables or the difference relative to the reference category for nominal variables. The effect of an explanatory variable on HRQL was considered clinically meaningful if the predicted score difference was at least as large as the minimally important difference. Results: Common predictors of better TOI, SWB and EWB were better general health and factors related to better perceived quality of cancer care. Predictor variables in addition to general health and perceived quality of care were identified only for SWB. Being married/living as married was associated with better SWB, whereas being male or of Hispanic ethnicity was associated with worse SWB. Among the sociodemographic, cancer/health, and healthcare variables evaluated, only Hispanic ethnicity had a clinically meaningful effect on an HRQL outcome. Conclusion: Our findings, particularly the information on the clinical importance of predictor variables, can help clinicians identify patients who may be at risk for poor future HRQL. Potentially modifiable factors were related to perceived quality of cancer care; thus, future research should evaluate whether improving these factors improves HRQL. © 2008 Yost et al; licensee BioMed Central Ltd.