The effect of gender and other factors on early child mortality in China.

Proportional hazards models for first order and higher order births were used to examine the effects of sex and other maternal demographic covariates on infant and child mortality in China. Retrospective data were obtained for a 10% sample from the 1988 2% sample of fertility and contraception conducted by the State Family Planning Commission. The sample included 89,086 children born during 1965-79 and 1980-88 to women aged 15-57 years in 1988. The data set excluded mortality during the famine in the later 1950s and early 1960s. The results confirmed a significant effect of gender on early child mortality. Female survival is the greatest if the female is an only child. The mortality hazard is higher for a female child with an older female sibling. Sex discrimination was most obvious among second and higher order births. The probability of a second birth and a short birth interval was higher for mothers with a female first child. Socioeconomic factors had strong effects on first order births. Mortality was lower among children born in urban areas to mothers with formal education and employment in nonfarming sectors. Child mortality was lower among the majority Han nationality. Survival of second and higher order births was affected by prior infant or child mortality and preceding birth interval. A comparison of the two time periods of births showed that child mortality among first order births was lower during the 1980s. There was higher mortality among second and higher order births. Female child survival was similar or above male child survival in 1965-79, but female child survival was lower than male survival in 1980-88. The mortality hazard by sex for second and higher order births with controls was significantly different at the 5% level, but not apparent for first order births.