Birth intervals and childhood mortality in rural Tanzania

Family planning programmes in sub-Saharan Africa (SSA) often disseminate the proposition that birth spacing improves child survival. Yet, there are few examinations of this hypothesis that benefit from longitudinal data. This paper addresses this gap using 15 years of prospective data from three rural districts of Tanzania. The effect of birth interval durations on the risk of childhood mortality was estimated by fitting Weibull parametric hazard regression models with shared frailties to a dataset that comprised records of reproductive events and their succeeding survival trajectories of 25,762 mother-child dyads that lived in the sentinel areas of the Ifakara and Rufiji Health and Demographic Surveillance Systems from 2000 to 2015. The analysis was motivated by two hypotheses: First, that relatively short subsequent and preceding birth intervals would be associated with heightened risks of child mortality; however, that the effects of short subsequent birth intervals would be most pronounced among children between 12 and 59 months of age; and second, that the effects of short preceding birth intervals would be most acute during the neonatal and post-neonatal period. Results, which were adjusted for confounder effects at the individual, household, and contextual levels, demonstrated significant associations between subsequent and preceding birth interval durations and childhood mortality risk. Regarding subsequent birth intervals, relative to birth spacing of less than 18 months, durations 24-35 and ≥36 months were associated with 1-5-year-old mortality risks that were 0.29 and 0.21 times lower. Relative to preceding birth intervals of less than 18 months, those of 24-35 months were associated with a neonatal mortality risk that was 0.48 lower. Compared to the same referent group, preceding birth intervals of 18-23, 24-35, and ≥36 months were significantly associated with 12-23-month-old mortality risks that were 0.20, 0.39, and 0.33 times lower. The findings are compared with those from similar studies held in SSA, and the potential for family planning programmes to contribute to improved child survival in settings, such as Tanzania, is discussed.