Effects of preceding birth intervals on neonatal, infant and under-five years mortality and nutritional status in developing countries: evidence from the demographic and health surveys

Int J Gynaecol Obstet. 2005 Apr:89 Suppl 1:S7-24. doi: 10.1016/j.ijgo.2004.11.012. Epub 2005 Jan 26.

Abstract

Objective: This paper examines the association between birth intervals and infant and child mortality and nutritional status.

Methods: Repeated analysis of retrospective survey data from the Demographic and Health Surveys (DHS) program from 17 developing countries collected between 1990 and 1997 were used to examine these relationships. The key independent variable is the length of the preceding birth interval measured as the number of months between the birth of the child under study (index child) and the immediately preceding birth to the mother, if any. Both bivariate and multivariate designs were employed. Several child and mother-specific variables were used in the multivariate analyses in order to control for potential bias from confounding factors. Adjusted odds ratios were calculated to estimate relative risk.

Results: For neonatal mortality and infant mortality, the risk of dying decreases with increasing birth interval lengths up to 36 months, at which point the risk plateaus. For child mortality, the analysis indicates that the longer the birth interval, the lower the risk, even for intervals of 48 months or more. The relationship between chronic malnutrition and birth spacing is statistically significant in 6 of the 14 surveys with anthropometric data and between general malnutrition and birth spacing in 5 surveys. However, there is a clear pattern of increasing chronic and general undernutrition as the birth interval is shorter, as indicated by the averages of the adjusted odds ratios for all 14 countries.

Conclusion: Considering both the increased risk of mortality and undernutrition for a birth earlier than 36 months and the great number of births that occur with such short intervals, the author recommends that mothers space births at least 36 months. However, the tendency for increased risk of neonatal mortality for births with intervals of 60 or more months leads the author to conclude that the optimal birth interval is between 36 and 59 months. This information can be used by health care providers to counsel women on the benefits of birth spacing.

MeSH terms

  • Birth Intervals*
  • Breast Feeding
  • Child Mortality*
  • Child, Preschool
  • Developing Countries*
  • Female
  • Health Surveys
  • Humans
  • Infant
  • Infant Mortality*
  • Multivariate Analysis
  • Nutritional Status*
  • Odds Ratio
  • Pregnancy
  • Retrospective Studies