E. Duarte, C. Cadarso-Suárez, V. Rodrigues, T. Kneib, B. de Sousa
What role does statistics play in decision-making in health? A journeyman’s tale began with the simple belief that breast cancer risk is associated with several reproductive factors such as early menarche and late menopause, and how these variables vary in space and time together with other reproductive and socioeconomic factors. The highly complex structure found in such data lead us to consider the inclusion of a trivariate interaction term between attendance, detection and mortality rates. This was implemented based on a Markov random field representation which enables efficient Bayesian inference and, when modeling breast cancer incidence rates, showed a significant improvement in terms of model fit when compared to a simpler geoadditive regression model. Findings open up to new research questions on how a woman’s rights are being fully exercised and what other factors may deter women from participating in a breast cancer screening program.
Palabras clave / Keywords: structured additive regression models, high order interaction term, breast cancer risk factors, spatial correlation, BayesX
Programado
Sesión bilateral SEIO-SPE: Statistics in Environmental Sciences and Health (Organizadores: Raquel Menezes y Carmen Cadarso)
30 de mayo de 2018 15:30
Sala Cristal
