Spatio-temporal pattern of two common cancers among Iranian women: An adaptive smoothing model

Raei, Mehdi; Schmid, Volker; Moayyed, Majid und Mahaki, Behzad (2019): Spatio-temporal pattern of two common cancers among Iranian women: An adaptive smoothing model. In: Journal of Buon, Bd. 24, Nr. 3: S. 1268-1275

Abstract

Purpose: Considering the increase in incidence of breast and cervix uteri cancers in Iran, this study investigates spatio-temporal patterns of the incidence of these two cancers by estimating the step changes between pairs of adjacent regions and between the Iranian women from 2004 to 2009. Methods: Using an adaptive smoothing model, spatio-temporal mapping of the breast and cervix uteri cancers and their changes were studied. Identification of step changes between the neighboring spatial units was carried out by modeling adjacency matrix elements as random variables. Results: There was a high relative risk of breast cancer around the central northern half of Iran, and a high relative risk of cervix uteri cancer was seen in the northeastern part of Iran. Northwest and southeast of Iran had a relatively low risk of breast and cervix uteri cancer. In general, step changes were largely similar between the two diseases with an agreement coefficient of 56%. This was observed in the Chaharmahal & Bakhtiari, and Kohgiluye & Boyerahmad provinces on the central band of Iran, as well as some eastern and northern regions on the map that were distinct from their adjacent provinces from the aspect of relative risk of both cancers. Conclusion: Identifying areas with high/low incidence risk can help health authorities to make better decisions to prevent and control breast and cervix uteri cancers and allocate resources more efficiently. In addition, determining and identifying the step changes in unexplained components of the disease risk can lead to a deeper understanding of the spatial structure of unmeasured confounding factors.

• Dokumententyp: Zeitschriftenartikel
• Fakultät: Mathematik, Informatik und Statistik > Statistik
• Themengebiete: 500 Naturwissenschaften und Mathematik > 510 Mathematik
• ISSN: 1107-0625
• Sprache: Englisch