The aim of this study is to correct misclassification of incidence of pancreatic cancer extracted from Iranian cancer registry across provinces. The data is extracted from Iranian annual of national cancer registration report 2014. Patients of a province that has covered less cancer cases than its expectation are assumed to be registered at a neighboring province with more than 100% expected coverage. A Bayesian method was used to estimate the rate of misclassification in registering cancer incidence in neighboring province which used the prior knowledge with observed data with determining a prior distribution for misclassified parameter. According to the results, provinces with more medical facilities are Tehran, Isfahan, Razavi Khorasan, East Azerbaijan, Mazandaran, Fars and Khozestan that had an expected coverage more than their expectation. Those provinces had significantly higher rates of pancreatic cancer than their neighboring provinces. Provinces with low medical facilities of Iran are south and north Khorasans, Sistan and Balouchestan, West Azerbaijan, Ardebil, Kigilouye and Boyerahmad, Golestan, Markzi, Qazvin, Arak, Ilam, Bushehr, Hormozgan that have had observed cancer cases less than their expectation and the highest estimated misclassification rates belong to more deprived provinces like Qazvin, Ilam and Bushehr.

pancreatic cancer, incidence, registry, Bayesian, Iran.