Partnership status, health and mortality : protection or selection?
Franke, S. (2018) Doctoral thesis. University of Liverpool. [ONS LS]
The migrant mortality advantage posits that immigrants have low mortality relative to non-migrants living in western countries, even among immigrants with a low socioeconomic status. Over time, the advantage is said to diminish as mortality among immigrants attenuates to the higher mortality rate among the host population. Among the descendants of immigrants, this mortality advantage can persist, wear off or even reverse. While immigrant mortality has been studied before in the U.S. and Europe, the actuality of the migrant mortality advantage remains contested and its causes (selection effects, cultural factors and the immigrant health transition) poorly understood. This is because two confounding factors exist which are rarely accounted for in studies. Registration uncertainty relates to the accuracy and timeliness of move reporting, stating that uncertainty in migrants’ entry and exits dates lead to overestimation of migrants’ time-at-risk in the host country. Health-related remigrations include returns to the country of origin at older ages which reflect a desire to die in the place of birth (a salmon bias effect) and at younger ages based on poorer general health (an unhealthy remigration effect). This can lead to an undercount of deaths in the host country statistics. Both of these confounding factors can cause a downward bias in migrant mortality rates, creating artificially low mortality and a data artefact. The aims of this thesis are to examine mortality patterns, and investigate the cause of mortality differences, between immigrants, their descendants and England and Wales-born population. The thesis makes several contributions to the literature. First, by adopting a rounded approach to the study of immigrant mortality through explicitly modelling the impact of registration uncertainty and investigating if immigrants remigrate in poor health from England and Wales, the research will show whether the migrant mortality advantage is an actuality or a data artefact. If the advantage is shown to be real, and registration uncertainty and health-related remigration can be dismissed as confounding causes, this will reduce much of the uncertainty surrounding the causes of the advantage. Second, the analysis of migrant mortality by sex, cause of death, generation, over age (as a proxy for length of stay) and migrants’ socioeconomic characteristics will significantly improve our understanding of the remaining causes of the migrant mortality advantage. There are few large-scale, robust international studies of immigrant mortality. This thesis thus offers a significant contribution to the international literature in a growing field of research. The analysis shows that the migrant mortality advantage cannot be accounted for by uncertainty in the registration of events and little support was found for health-related remigration among immigrants living in England and Wales. The migrant mortality advantage is an actuality and we can be confident that some combination of the healthy migrant effect, cultural factors and health transition explain the advantage. Low immigrant mortality (in nearly all groups after adjusting for socioeconomic characteristics) is driven primarily by low mortality from cancers (and in some populations from low cardiovascular disease mortality). Among many immigrant groups, low mortality coexists with low respiratory disease mortality and, among non-western immigrants, high infectious disease mortality. Low mortality among immigrants does wear off over time, but even at older ages some groups still have low mortality relative to non-migrants living in England and Wales. The descendants of immigrants do not experience a mortality advantage. They have higher mortality both before and after adjusting for their socioeconomic characteristics relative to immigrants and only similar mortality to the England and Wales-born population.
Output from project: 0301675