The effects of fertility behavior on child survival and child nutritional status: Evidence from the Demographic and Health Surveys 2006 to 2012.

The current study is the latest in a series of three based on DHS surveys that examine the relationship between spacing between births or pregnancies and infant and child mortality and nutritional status. The first study used data from 17 DHS surveys between 1990 and 1997. The second study used data from 52 different DHS surveys from 2000 to 2005. The present study includes data from 45 DHS surveys between 2006 and 2012 and extends the analysis by reporting on mother’s age at birth and child’s birth order in addition to the preceding birth-to-conception interval reported in the previous two studies. More than 1.1 million births that occurred in the 15 years prior to the surveys are included in the analyses of mortality under age five years and at ages one to five years. More than 420000 births in the five years preceding the surveys are included for the analyses of mortality in infancy. Nutritional status is analyzed for more than 340000 living children under age five years. The analysis methodology includes bivariate and multivariate Cox hazard regression for mortality and multivariate logistic regression for nutritional status. The results for spacing are quite similar to those found in the preceding studies confirming the importance of waiting a period of time to conceive again after a birth. While birth-to-conception intervals of fewer than 18 months (i.e. a birth-to-birth interval of 27 months) incur an increased risk of neonatal mortality intervals of fewer than 36 months birth to conception incur an increased risk of mortality at ages one to four years. Risks of mortality increase during infancy for birth-to-conception intervals longer than 60 months. Population-attributable risk calculations indicate 26 percent excess under-five mortality due to birth-to-conception intervals shorter than 36 months. The analysis of nutritional status indicates that the longer the preceding birth-to-conception interval the less likely a child is to be stunted (chronically malnourished) and underweight for his or her age. However the likelihood of wasting being a measure of acute malnutrition is not affected. Mother’s young age at birth (less than 18 years) increases the risk of dying for children at all ages under five years and a mother’s being 35 years or older increases the child’s risk of dying during infancy. After infancy however the chances of dying for children of older mothers are slightly lower than the chances for children whose mothers were 18-24 years when the child was born. Children of older mothers are also less likely to be stunted and underweight. Being born to a mother under age 18 years increases a child’s likelihood of being stunted and underweight. First-born children have a substantially higher risk of mortality at all ages under age five years even when a young mother’s age at birth is taken into account. Higher birth order does not affect neonatal mortality but increases post-neonatal ages one to four and all under-five mortality and slightly increases rates of stunting but not underweight or wasting. The study indicates that the more fertility behavioral risk factors a child faces the greater are the chances of dying and being malnourished. A child with three risk factors is twice as likely to die before reaching age five as a child with no risk factors and if the child survives she or he is 30 percent more likely to be suffering from chronic malnutrition.