Under-five mortality fell faster than newborn mortality in East Africa
In eight East African countries, estimated under-five mortality fell proportionally more than neonatal mortality from 2000 to 2024. In Rwanda, the under-five estimate fell from 184.2 to 37.7 deaths per 1,000 live births, while the newborn estimate fell from 43.1 to 17.4.
Sources checked
Rift & Coast Editorial Team
Child survivalTwo mortality rates per 1,000 live births, 2024
Deaths before age five
Shared scale · 0 per 1,000 to 101.1 per 1,000
Somalia101.1 per 1,000
South Sudan96.7 per 1,000
Uganda48.7 per 1,000
Burundi47.1 per 1,000
Ethiopia44.5 per 1,000
Kenya38.8 per 1,000
Rwanda37.7 per 1,000
Tanzania37 per 1,000
Deaths in the first 28 days
Shared scale · 0 per 1,000 to 101.1 per 1,000
Somalia34.2 per 1,000
South Sudan39.5 per 1,000
Uganda21.3 per 1,000
Burundi19.1 per 1,000
Ethiopia25.2 per 1,000
Kenya20.7 per 1,000
Rwanda17.4 per 1,000
Tanzania19.9 per 1,000
Switch between age windows. The national rates share one scale and stay in under-five order.
The newborn period became a larger share of the risk
In Rwanda the neonatal rate was about 23% of the under-five rate in 2000 and 46% in 2024. These are ratios of two probabilities per live birth. They describe how the estimated risk is distributed across the overlapping age windows. Source 1Source 3
Kenya illustrates how the pace can differ
Kenya’s under-five rate fell from 95.2 to 38.8 per 1,000, a decline of about 59%. Its neonatal rate moved from 26.3 to 20.7, a decline of about 21%. The data show the difference in pace. They give no direct measure of care quality or causes of death. Source 1Source 3
Somalia and South Sudan had the highest 2024 rates shown
Somalia’s estimated under-five rate was 101.1 per 1,000 and South Sudan’s was 96.7. South Sudan’s neonatal rate was 39.5 and Somalia’s was 34.2. The order changes when the age window changes. Source 1Source 3
Read precise values as estimates
The UN Inter-agency Group for Child Mortality Estimation combines evidence from registration, census and household survey data where available. A point estimate represents a modelled probability. The comparison uses that estimate for each country. Source 2Source 4