Abstract:
Introduction: Infant mortality is the death of an infant before his or her first birthday. In 2021,
around 3.8 million infants lost their lives, and half of them, approximately 1.9 million were Sub Saharan region. Although the intervention is successful in lowering infant mortality rates
worldwide Sub-Saharan Africa, particularly East Africa, continues to struggle with high rates of
infant and under-five mortality. Though many studies have been conducted limited literature
accounts for the impact of spatial effect. Therefore this study aims to add the spatial random effect
to identify factor associated with infant mortality.
Objective: This study aims to determine the IMR, spatial variation, and its time to death
determinants in East Africa from 2015-2022.
Method: This study included a total of 101,532 infants. The IMR was estimated and the spatial
pattern was checked by Moran’s I statistics. The Bayesian frailty spatial frailty analysis model
with the AFT approach was used to identify the predictor variable. The model was diagnosed
through a Cox-Snell plot, and convergence was checked through a trace plot and density plot.
Results: The IMR was 39.83 per 1000 live births (95% CI: 35.81, 44.27) in nine East African
countries. The spatial distribution of IMR shows a nonrandom pattern and the identified hotspot
area was northwestern, northeastern, and eastern Ethiopia, Burundi, western, northern, and
southern Malawi, the northeast of Mozambique, southern Tanzania, southwestern Zambia, and
northwestern and eastern Zimbabwe. Factor significantly associated with infant survival time from
the lognormal AFT model was started breastfeeding after 24 hours (AF = 0.0905, 95%CI: 0.084,
0.106), didn’t have ANC follow-up during her pregnancy (AF = 0.652, 95%CI: 0.535, 0.794) a
mother age between 15-24 years (AF = 0.798, 95%CI: 0.709, 0.9), small-sized birth weight (AF
= 0.634, 95%CI: 0.565, 0.728), multiple births (AF = 0.25, 95%CI: 0.212, 0.3), Moreover, the
Male child's (AF = 0.784 95% CI 0.708, 0.8), parity more than five (AF = 0.79 95%CI: 0.678,0.93),
and being unemployed mother (AF = 1.35 95%CI: 1.11, 1.62).
Conclusion and Recommendation: The pooled IMR was higher than the global estimate. There
was a spatial clustering pattern in infant mortality. Breastfeeding initiation time, number of ANC
visits, maternal age, and birth weight, number of pregnancies, infant sex, parity, and
unemployment of mother were significantly associated with infant survival time. It is
recommended that enhancing ANC service close monitoring of multiple pregnancies, and enhance
immediate breastfeeding
