PhD student: Jonathan GRASSI

Title: Survival of Children and Adolescents with Cancer in Espírito Santo – A State in Brazil: A Retrospective Cohort Study

Supervisors Dr Luis Carlos LOPES-JUNIOR and Dr Jacqueline CLAVEL

Doctoral school: Federal University of Espirito Santo. Centre of Research in Epidemiology and StatisticS, Université Paris Cité, Université Sorbonne Paris Nord, Inserm, INRAe

Promotion: 2021

Funding:  CAPES – Brazil

Thesis abstract:

Introduction: Cancer in children and adolescents corresponds to a group of several diseases that have in common the uncontrolled proliferation of abnormal cells and that can occur anywhere in the body. In the last four decades, there have been advances in the treatment of childhood cancer. Following these advances, survival analysis techniques have assumed an important role in recent years, especially after the development and improvement of the technology of statistical methods, with numerous applications, mainly in Medicine as well as Collective Health. Objective: To investigate the overall survival and relative survival of children and adolescents with cancer, as well as to identify the competing risks and factors associated with the risk of death from pediatric neoplasms. Methods: Retrospective analytical observational cohort study. The cohort will consist of new cases diagnosed with primary malignant neoplasm of any morphology in children and adolescents between 0 and 19 years of age in a 10-year period (between 2007 and 2017), in the capital of Espírito Santo, Brazil. Secondary data will be collected from the Hospital-based Cancer Registry (RHC) of the Hospital Estadual Infantil Nossa Senhora da Glória (HEINSG) as well as the Mortality Information System – SIM/SESA. In the data analysis, the Kaplan- Meier method will be used, and in the multivariate analysis, the Cox proportional rich model and the Fine & Gray competitive risks model will be analyzed to investigate the probability of specific death from cancer or from other non-related causes. cancer. In the final model, the variables that present significance levels below 5% will remain, and the statistical procedures will be conducted in the IBM→ Statistical Package for the Social Sciences (SPSS) software, version 28.0.


Back to top