Cause-specific mortality among first- and second-generation immigrants in the 2000s in Belgium: moving beyond the healthy migrant effect?

Background: Lower mortality among first-generation (1-gen) adult immigrants in industrialized countries is well established, whereas the underlying processes have been the subject of debate. In recent years, controversy has been solved in favour of selection processes and cultural explanations. A question that remains is whether the favourable mortality profile of first-generation immigrants applies also to the second generation. As migrants and their descendants are often living in poor socioeconomic (SE) conditions, this study hypothesizes that mortality of second-generation Turkish and Moroccan immigrants may be relatively high. The aim of this study is to compare the mortality profile of the host population with that of first- and second-generation immigrants of Turkish and Moroccan origin in Belgium, taking SE characteristics into account.
Methods: Data were extracted from record linkage between the Belgian 2001 census and information on emigration and cause-specific mortality for the period 2001-2009. The research population comprised all official inhabitants of Flanders and Brussels at the time of the census, aged 25 to 54. Directly age-standardized mortality rates (ASMRs), using the Belgian 2001 population as standard, and rate ratios, using Cox and Poisson regressions, were calculated. SE position was operationalized by education, housing quality and ownership, and employment status.
Results: First-generation immigrants have a mortality advantage, while second-generation immigrants have a disadvantage relative to the host population. Particularly striking in this respect are the high mortality rates in non-western immigrants of the second generation. The ASMR in men of this background amounts to 719.1 per 100,000 (95% CI 582.0 – 856.2), compared to 300.2 (95% CI 191.4 – 413.2) in Belgian men The same pattern was found among women, and for all causes considered. Relative differences were particularly pronounced for mortality due to drugs, respiratory diseases and infections. For some causes, 2-gen excess mortality disappeared after control for SE factors, while it weakened for others.
Conclusions: First-gen Turkish and Moroccan immigrants have a more favourable mortality profile than the host population, while the reverse applies to the 2 gen. The convergence of the health and mortality profile of second-generation immigrants towards that of the host population with similar SEP for most outcomes indicates the need for policies simultaneously addressing different areas of deprivation