Fertility, offspring gender, and parental mortality later in life

The relationship between childbearing and postreproductive mortality has attracted attention from researchers in biology, demography, anthropology, and gerontology. Findings from this growing literature are inconsistent. In addition to the relationship between parity and mortality, some studies also explore whether the gender of children may affect parental postreproductive longevity. Similar to studies examining the association between parity and mortality, the link between offspring gender and longevity is not conclusive. Two broad types of mechanisms have been proposed to explain how reproduction may affect postreproductive maternal mortality: those that are biological and those that are socio-cultural. Most of previous studies dealt with all biological children, instead of surviving children. Moreover, studies that used contemporary data to examine these relationships are often based on developed countries. Far less is known about these relationships in developing countries. The current study examines the number of surviving children as well as the number of surviving sons and daughters in relation to parental postreproductive mortality in China, controlling for a variety of potential confounders. Data used are based on the baseline survey of the Beijing Multidimensional Longitudinal Study on Aging (BMLSA) conducted in 1992 and mortality data collected up to 2004. Cox proportional hazards regression models are used to estimate the mortality hazard ratios. Results show that females with three to six surviving children have significantly better survival than those with only one. Males with no surviving children have higher mortality risk, compared to males with one surviving child. More daughters are associated with lower maternal mortality. It is possible that larger social support from additional children reduces females' mortality. In addition, accumulated hormonal changes due to repeated pregnancies may protect females from some diseases. Lower levels of social control and social support attributable to children may increase mortality risk among men if they have no surviving children. Our results on the interaction between fertility and urban/rural areas show that the beneficial effects of children are more pronounced in urban females with low level of fertility in relation to rural females.