Mortality Is Elevated During Infancy and Early Childhood When Births Are Closely Spaced

Susan London

First published online:

Short birth intervals are associated with an increased risk of death both for the new infant and for the older sibling, surveillance data from two Nairobi slums confirm. 1 Nearly one-tenth of second or higher order births in the study regions occurred within 18 months of the birth of an older sibling; compared with their counterparts delivered at least 36 months after the birth of a sibling, these infants were substantially more likely to die during the first year of life (hazard ratio, 2.1). In addition, compared with their last-born peers, children who experienced the birth of a younger sibling within 20 months of their own birth had an elevated mortality rate during the second year of life (2.7).

Although short birth intervals have long been linked to infant and child mortality, most research has relied on retrospective data, which may be inaccurate if participating parents misremember the timing of events or omit mention of deceased infants. The few prospective studies that have been conducted in Sub-Saharan Africa have yielded conflicting results, and none have focused on urban areas. Moreover, most studies have not controlled for reverse causality: Some couples may seek a new pregnancy in response to a child’s death, so that the short birth interval is the result of rather than the cause of mortality.

To address these issues, Fotso and colleagues analyzed data from the Nairobi Urban Health and Demographic Surveillance System for all infants born in the Viwandani and Korogocho slums between 2003 and 2009. All households in these informal settlements—about 28,000 in all during the study period—are visited every four months to collect information on fertility, mortality, migration, health and related topics. For each birth, investigators calculated the preceding birth interval (time between the birth of the preceding child and the birth of the index child) and the succeeding birth interval (time between the birth of the index child and the birth of the succeeding child). They constructed multivariate models to identify factors associated with death during infancy (the first year of life) and early childhood (the second year).

The analytic sample consisted of 13,502 births. The majority occurred to mothers who were younger than 25 (56%), were in a union (84%) and had no more than a primary education (74%). About half (48%) of infants were first births, 5% were born within 18 months of an older sibling, 6% were born 18–23 months after a sibling and the remaining 42% were born after an interval of at least two years. Overall, nearly one in 10 second or higher level births occurred within 18 months of the previous birth. Only 4% of index births were followed by another birth within 20 months, a proportion that partly reflects the low fertility rate in these slums.

The overall infant mortality rate was 77 per 1,000 live births, and the overall early childhood mortality rate was 20 per 1,000 live births. Both rates generally declined during the study period.

In a multivariate model that included the preceding birth interval, whether the child was a singleton and whether the interval was ascertained from surveillance or from retrospective report, infants born less than 18 months after a sibling’s birth were more likely to die during infancy than were those born after an interval of at least 36 months (hazard ratio, 2.5). This association remained significant, though slightly weaker, in a second multivariate analysis that controlled for child, maternal and household characteristics (hazard ratio, 2.1). Infants also had an elevated risk of mortality if they were a twin or triplet (3.9), or if either the preceding birth interval had been ascertained from maternal report or the infant was the mother’s first (1.4); they were less likely to die if their family was in the highest rather than lowest of three household wealth categories (0.6) or they had been born in Viwandani rather than Korogocho (0.7). An alternative model that treated birth interval as a continuous variable and was restricted to second and higher order births found that an infant’s risk of death fell with each additional year of spacing from the preceding birth (0.8).

The investigators constructed similar models for early childhood mortality. In the first model, relative to peers born at least 36 months after a sibling, infants born less than 18 months afterward were more likely to die during their second year of life (hazard ratio, 2.6). Moreover, compared with last-born children, those experiencing the birth of a sibling less than 20 months after their own birth had a highly elevated mortality risk (2.8). Risk was also elevated for subsequent birth intervals of 20 or more months (2.6), though this association may in part reflect reverse causation, given that the younger child may have been conceived after the older child had died.

In the second model, the association between birth intervals of less than 18 months and early child mortality was only marginally significant (hazard ratio, 2.1); risk remained elevated, however, for children experiencing the birth of a sibling less than 20 months or at least 20 months after their own birth (2.7 and 2.5, respectively). Infants born to families in the highest wealth tertile had a reduced risk of early childhood mortality (0.6). The third model revealed a marginally significant trend whereby an infant’s risk of early childhood mortality fell by about one-tenth with each additional year of spacing from the previous birth (0.9).

Overall, according to the investigators, the findings validate those of retrospective studies that similarly linked short birth intervals to elevated mortality early in life. Birth spacing could go a long way toward improving infant survival in Sub-Saharan Africa, where many children are exposed to "the double jeopardy of short preceding and succeeding intervals," the researchers maintain; a strategy that emphasized family planning to lengthen the time between births would likely be well received in this region, as contraceptives have been used to space (rather than limit) births to a greater extent there than in Asia or Latin America. "The findings from this analysis lend support to the importance of revitalizing birth spacing as a child survival intervention—especially in sub-Saharan Africa, where levels of unmet need for birth spacing and failure to avoid mistimed pregnancies remain unacceptably high," the investigators conclude.—S. London


1. Fotso JC et al., Birth spacing and child mortality: an analysis of prospective data from the Nairobi Urban Health and Demographic Surveillance System, Journal of Biosocial Science, 2013, 45(6):779–798.