@article{TGC4381,
author = {Mohamad Amin Pourhoseingholi},
title = {Bayesian adjustment for misclassification in cancer registry data},
journal = {Translational Gastrointestinal Cancer},
volume = {3},
number = {4},
year = {2014},
keywords = {},
abstract = {Background: Mortality and incidence are the familiar projections in the assessment of the burden of cancers. But in developing countries, the analysis of death statistics subject to misclassification and this is a major problem in epidemiological analysis, often leading to biased estimates, and can therefore cause one to underestimate health risks. Two statistical approaches are recommended to overcome misclassification; first is using a small validation sample and the second is Bayesian analysis in which subjective prior information on at least some subset of the parameters is used to re-estimate misclassified statistic. The aim of this study was to explain this Bayesian model and its application in estimating the burden colorectal cancer (CRC) in Iran.
Methods: National death Statistic from 1995 to 2004 included in this analysis. The Bayesian approach to correct and account for misclassification effects in Poisson count regression with a beta prior employed to reestimate the mortality rate of CRC in age and sex group. Years of life lost’ (YLL), for CRC were expressed as the annual rates/100,000, general and/or per gender, and age group.
Results: According to the Bayesian re-estimate, there were between 30 to 40 percent underestimation for YLL according to reported mortality records in death due to CRC and the rate of this YLL increased through the recent years.
Conclusions: Our findings suggested a substantial underestimate of burden due to CRC in Iranian population. So healthcare policy makers who determine research and treatment priorities on death rates should notice to this underestimation.},
url = {https://tgc.amegroups.com/article/view/4381}
}