Comparing Spatial Ecological Regression Models in Breast Cancer Incidence in Iran

Abstract

Background: The present study investigates the incidence of breast cancer in Iran at the province level and also explores the impact of the covariates of some risk factors using the temporal risk of cancer, Poisson, log-normal, and Besag-York-Mollie (BYM) models. Methods: This is an ecological study and the population includes all patients with breast cancer from 2005 and 2009 in Iran. Risk factors for the disease included physical activity and being overweight, according to an annual report on the registration of cancer cases in the country. Individuals with a positive diagnostic test or a biopsy for breast cancer were identified as definitive and registered at the breast cancer registry. Poisson, log-normal, and BYM models were used to estimate the relative risk indices, spatial risk, and temporal risk of cancer. Results: The total number of registered breast cancers from 2005 to 2009 in Iran was 32 694 cases. Among the provinces, the most observed cases are related to Isfahan Province with 2862 cases, and Khorasan Razavi Province with 2646 cases. The lowest incidence is associated with Kohgiluyeh and Boyer-Ahmad Province with 104 cases. The highest relative risk of breast cancer was observed in 2009 in Tehran Province and the lowest in Kohkiluyeh Boyer-Ahmad Province, and Sistan-Baluchestan Province. Conclusion: Due to the better fit of the BYM model, it is recommended to use the BYM model in future studies, and considering the significance of the human development index variable, this variable should be considered in similar studies.

Keywords

• disease mapping
• breast cancer
• besag-york-mollie (bym) model
• poisson regression model
• normal log model