Risk Factors for Medium-term Adverse Outcomes and Predictive Modeling for Breast Cancer Patients

This subproject is included in a project coordinated between program groups Evaluation of Health Services of Epidemiology and Public Health (Project CAMISS). The overall objective is to evaluate different aspects of health care received by patients with breast cancer from the diagnostic process, treatment, complications, survival and quality of life, to provide information to improve the effectiveness of interventions, reduce variability, have best predictive rules and increase the quality of life. The main objective of the subproject is to create and validate prospectively predictive rules of recurrence, complications, mortality, changes in quality of life in these patients at admission and one/two years of treatment and to evaluate the external validity of our rules to predict relapses, complications and mortality in the retrospective sample of patients participating in screening programs. Observational methodology with information available from a retrospective cohort of women diagnosed between 2000 and 2009 and another 2-year prospective follow-up included 2040 incident cases of breast cancer in 18 hospitals of 5 regions. Cohorts will learn clinical and health care diagnosis, tumor, treatment, hospital, follow-up (complications, relapse, and vital status) cost and quality of life. Prediction rules are created by means of regression/Cox models at the prospective sample and also investigators will assess the external validity in the retrospective cohort in the case of recurrence, complications and mortality. Expected results: There are currently no results of cohort analysis of the diagnostic process of care that integrates different aspects. The study will create tools to assist prognostic and therapeutic decision making process in these patients