Trends in total and cause-specific mortality by marital status among elderly Norwegian men and women

Abstract

Background: Previous research has shown large and increasing relative differences in mortality by marital status in several countries, but few studies have considered trends in cause-specific mortality by marital status among elderly people. Methods: The author uses discrete-time hazard regression and register data covering the entire Norwegian population to analyze how associations between marital status and several causes of death have changed for men and women of age 75-89 from 1971-2007. Educational level, region of residence and centrality are included as control variables. There are 804 243 deaths during the 11 102 306 person-years of follow-up. Results: Relative to married persons, those who are never married, divorced or widowed have significantly higher mortality for most causes of death. The odds of death are highest for divorcees, followed by never married and widowed. Moreover, the excess mortality among the non-married is higher for men than for women, at least in the beginning of the time period. Relative differences in mortality by marital status have increased from 1971-2007. In particular, the excess mortality of the never married women and, to a lesser extent, men has been rising. The widening of the marital status differentials is most pronounced for mortality resulting from circulatory diseases, respiratory diseases (women), other diseases and external deaths (women). Differences in cancer mortality by marital status have been stable over time. Conclusions: Those who are married may have lower mortality because of protective effects of marriage or selection of healthy individuals into marriage, and the importance of such mechanisms may have changed over time. However, with the available data it is not possible to identify the mechanisms responsible for the increasing relative differences in mortality by marital status in Norway.