Ethel Lim1, Michelle Miller1, Billingsley Kaambwa2, Bogda Koczwara3. 1. Nutrition and Dietetics, Flinders University, Bedford Park, South Australia 5042, Australia. 2. Repatriation General Hospital, 202-16 Daws Road, Daw Park, South Australia 5041, Australia. 3. Flinders Centre for Innovation in Cancer, Flinders Drive, Bedford Park, South Australia 5042, Australia. Electronic address: bogda.koczwara@flinders.edu.au.
Abstract
OBJECTIVE: The objective of this study was to identify significant cardiometabolic predictors of mortality among older cancer survivors and develop and validate a screening instrument to assess individual risk of mortality. MATERIALS AND METHODS: Retrospective cohort study used collected data from the ALSA. Cox proportional hazards model was used to derive the risk equation for mortality that could be evaluated at 10years. Measures of discrimination and calibration were calculated in the validation cohort. RESULTS: The equation was developed using 294 cancer survivors and validated in 127 different cancer survivors. Significant cardiometabolic predictors of mortality included in the final model are age, sex, history of cerebrovascular disease, non-adherence to exercise guidelines (150min moderate activity per week), and smoking. Discrimination and calibration were acceptable with minimal differences in C statistics (0.0442, 95% CI: -0.0149 to 0.103) and adjusted R2 values (0.0407, 95% CI: -0.181 to 0.0998) between the development and validation cohorts, respectively. CONCLUSION: We have developed and validated the first screening tool to predict cardiometabolic risk of mortality in older cancer survivors and defined centile values for risk classification. Further validation and research on the usability and usefulness of the tool in clinical practice are recommended in order to target cancer survivors for interventions. Cost effectiveness of such an approach should also be examined.
OBJECTIVE: The objective of this study was to identify significant cardiometabolic predictors of mortality among older cancer survivors and develop and validate a screening instrument to assess individual risk of mortality. MATERIALS AND METHODS: Retrospective cohort study used collected data from the ALSA. Cox proportional hazards model was used to derive the risk equation for mortality that could be evaluated at 10years. Measures of discrimination and calibration were calculated in the validation cohort. RESULTS: The equation was developed using 294 cancer survivors and validated in 127 different cancer survivors. Significant cardiometabolic predictors of mortality included in the final model are age, sex, history of cerebrovascular disease, non-adherence to exercise guidelines (150min moderate activity per week), and smoking. Discrimination and calibration were acceptable with minimal differences in C statistics (0.0442, 95% CI: -0.0149 to 0.103) and adjusted R2 values (0.0407, 95% CI: -0.181 to 0.0998) between the development and validation cohorts, respectively. CONCLUSION: We have developed and validated the first screening tool to predict cardiometabolic risk of mortality in older cancer survivors and defined centile values for risk classification. Further validation and research on the usability and usefulness of the tool in clinical practice are recommended in order to target cancer survivors for interventions. Cost effectiveness of such an approach should also be examined.