The Protective Effect of Marriage on Mortality in a Dynamic Society

Purpose: We sought to assess whether the protective effect of marriage on overall and cause-specific mortality has changed over time in a dynamic society. Methods: Data from the census-based Israel Longitudinal Studies (ILMS) I (1983-1992) and II (1995-2004) were analyzed. Cox proportional hazard modeling adjusting for sociodemographic factors was applied to 152,150 and 209,125 individuals, ages 45-89 years from the ILMS I and II, respectively. During each study period 31,749 (ILMS I) and 37,656 (ILMS II) deaths were reported. Results: Mortality inequalities by marital status remained significant and widened over time for middle-aged and elderly men and women. Changes in cause-specific mortality indicated a widening of cardiovascular disease mortality inequalities by marital status. An increasing trend was also noted for deaths from cancer (+25%) and other causes (+38%, p < 0.05) in middle-aged men, but not women (cancer = 0%; other causes = -3%). Conclusions: The stronger beneficial effect of marriage over time may reflect societal changes that have differentially affected men and women.