Sex-differences in the evolution of life expectancy and health in older age

Since the middle of the 19th century, there have been considerable gains in life expectancy at all ages, especially during the first two decades of the 20th century. Throughout this period, women have had greater survival chances than males. Nevertheless, there are some notable sex differences in survival gains. Male infant survival increased faster than female infant survival, indicating narrowing sex differentials. The opposite is observed for survival to older ages. For instance, female survival to ages 50 and 65 increased faster than male survival to these ages, indicating a widening sex differential. This finding raises a question: how do these differences in cohort survival history affect health and mortality at ages above 65 years? The role of cohort survival history was examined as an explanatory factor for sex differences in three-year survival in older persons in two surveys on health and aging, the Dutch Longitudinal Study among the Elderly (DLSE, baseline 1955-1957) and the Longitudinal Aging Study Amsterdam (LASA, baseline 1992-93). Both surveys are based on representative samples across the Netherlands, stratified by age and sex, with over 2000 persons in the common age group of 65-84 years. Cohort survival history was expressed as the percentage surviving up to ages 1, 15, 40, 50, and 65 based on cohort survival tables of the birth cohorts included in the study. Both in 1955-57 and in 1992-93, when controlling for number of chronic diseases, functional limitations, and self-rated health, multiple logistic regression models showed that survival was significantly associated with sex, indicating that females had better survival than males. Inclusion of cohort survival history in the model raised these Risk Ratios to values closer to 1. Inclusion of survival to age 1 in DLSE resulted in a reversal of the sex differential. In LASA, inclusion of survival to age 1 increased the Risk Ratio significantly, but did not bring about a reversal of the risk. Inclusion of survival to age 50 in both surveys resulted in sex being no longer significantly associated with survival. The consistent findings in both surveys suggest that the more favorable cohort survival history of females explains part of the female over male advantage in survival in later life, given the same level of health. This is true even for the infant mortality experience of the cohort. The better survival of females at all ages appears to perpetuate itself in older age.