The aim of this study on cancer mortality among Turkish immigrants, for the first time, traditional comparisons in migrant health research have been extended simultaneously in two ways. First, comparisons were made to cancer mortality from the immigrants' country of origin and second, cancer mortality among Turkish immigrants across four host countries (Belgium, Denmark, France and the Netherlands) was compared. Population-based cancer mortality data from these countries were included. Age-standardized mortality rates were computed for the local-born and Turkish population of each country. Relative differences in cancer mortality were examined by fitting country-specific Poisson regression models. Globocan data on cancer mortality in Turkey from 2008 were used in order to compare mortality rates of Turkish immigrants with those from their country of origin. Turkish immigrants had lower all-cancer mortality than the local-born populations of their host countries, and mortality levels comparable to all-cancer mortality rates in Turkey. In the Netherlands and France breast cancer mortality was consistently lower in Turkish immigrants women than among local-born women. Lung cancer mortality was slightly lower in Turkish immigrants in the Netherlands and France but varied considerably between migrants in these two host countries. Stomach cancer mortality was significantly higher in Turkish immigrants when compared to local-born French and Dutch. Our findings indicate that exposures both in the country of origin and in the host country can have an effect on the cancer mortality of immigrants. Despite limitations affecting any cross-country comparison of mortality, the innovative multi-comparison approach is a promising way to gain further insights into determinants of trends in cancer mortality of immigrants.
European Journal of Epidemiology, 2012, Vol 27, Issue 12, p. 915-921