Age-Period-Cohort Robust Bayesian Models for Projecting Cancer Incidence and Mortality in Puerto Rico

Abstract

Race is not available in Puerto Rican data so the Worksheet for Completeness of Case Ascertainment can not estimate completeness. Cancer data is subject to unavoidable delays. Projections of cancer incidence and mortality provide a valuable indication of the current and future burden. They better inform planning and decision making, and assist in the e cient allocation of resources to meet the future needs for the prevention, detection, and treatment of cancer.Objective: To estimate the present and predict the future (2014) of incidence and mortality for top cancer in Puerto Rico (PR), by gender, age group and primary cancer site to design public policy; and to give an indication of the degree of cancer registry completeness.Methods: Incidence and mortality data from Puerto Rico Central Cancer Registry, were obtained for the years 1985 to 2004. A robust Age-period-cohort (APC) model with autoregressive errors were tted using Bayesian methods.