Global child deaths fell by half since 2000 — and migrated into the first month of life: newborns are now the majority of under-5 deaths in 123 of 196 countries

Two child-mortality series compiled by the UN Inter-agency Group for Child Mortality Estimation (UN IGME) and redistributed through the World Bank measure deaths against the same denominator — 1,000 live births. The under-5 mortality rate (indicator SH.DYN.MORT) counts deaths before the fifth birthday; the neonatal mortality rate (SH.DYN.NMRT) counts deaths in the first 28 days of life, a subset of the first ^mmodeled. Because both are expressed per 1,000 live births, the neonatal rate divided by the under-5 rate gives the share of under-5 deaths that occur in the newborn period. For the world, under-5 mortality fell from 76.7 per 1,000 in 2000 to 37.4 in 2024, a decline of 51% ^mmodeled. Neonatal mortality fell over the same span from 30.3 to 17.2, a decline of 43% ^mmodeled. The remainder — deaths between the 29th day and the fifth birthday — fell from 46.4 to 20.2, a decline of 57% ^mmodeled. Because the newborn rate fell more slowly than everything after it, the neonatal share of under-5 deaths rose from 39.5% in 2000 to 46.0% in 2024, climbing in nearly every intervening year ^mmodeled. Both series carry 2024 estimates for 196 countries. In 123 of those 196, the neonatal period now accounts for more than half of all under-5 deaths ^mmodeled. The direction holds across the development gradient: among the 130 countries that have driven under-5 mortality to 25 per 1,000 or below, the median neonatal share is 55.9%; among the 66 still above that line, it is 45.8% ^mmodeled. The highest-mortality countries show the lowest newborn shares — Nigeria (under-5 rate 115.6, neonatal share 34%), the Democratic Republic of Congo (89.7, 26%), Niger (110.7, 30%), Chad (97.3, 31%) — while lower-mortality countries cluster far higher: the United States (under-5 rate 6.5, share 57%), India (26.6, 63%), Egypt (22.4, 62%), Brazil (14.2, 50%) ^mmodeled. Sustainable Development Goal 3.2 sets two separate 2030 thresholds: under-5 mortality "as low as 25 per 1,000 live births" and neonatal mortality "as low as 12 per 1,000 live births". On the 2024 estimates, 130 of the 196 countries meet the under-5 threshold and 121 meet both ^mmodeled. Nine meet the headline under-5 target but still miss the neonatal one — Algeria, Azerbaijan, Egypt, Grenada, Iraq, Jamaica, Moldova, Trinidad and Tobago and Venezuela — against only 2 in the reverse position ^mmodeled. Regionally, the neonatal share runs 37% in Sub-Saharan Africa, 51% in East Asia and the Pacific, 53–56% across Europe, North America and Latin America, and 62% in both South Asia and the Middle East and North Africa ^mmodeled.

The following is analysis, not fact. The rising newborn share is the arithmetic signature of a specific kind of success. The deaths that a country eliminates first are the post-neonatal ones — pneumonia, diarrhea, malaria, measles — which WHO attributes largely to conditions that are cheap and well-understood to prevent: vaccines, oral rehydration, insecticide-treated nets, antibiotics ^mmodeled. What remains after those are driven down is the neonatal core: prematurity, complications during birth, and newborn sepsis, which require skilled delivery, functioning maternity facilities and hospital-grade newborn care rather than a campaign or a commodity ^mmodeled. The first set of deaths falls to public-health logistics; the second falls only to health-system depth. So as the logistics work, the composition of child death shifts toward the harder half — not because newborns are dying more (they are dying much less), but because everything after the first month is falling faster ^mmodeled. Read across all 196 countries at once, the neonatal share behaves almost like an odometer for how far a country has traveled down the child-survival road. Where under-5 mortality is still above 100 per 1,000 — Nigeria, Niger, Chad — the newborn share sits near a third, because infectious post-neonatal death still dominates ^mmodeled. Where under-5 mortality has fallen into the single digits, the newborn share climbs past 55% and often past 65%, because almost nothing is left to eliminate except the fragile first weeks ^mmodeled. The median-share gap between the countries that have cleared the SDG under-5 line (55.9%) and those that have not (45.8%) is that same journey, frozen as a cross-section ^mmodeled. The SDG's own two-threshold structure encodes the asymmetry. That 9 countries have cleared the under-5 target while missing the neonatal one, and only 2 the reverse, means the newborn floor is the stickier of the two — the last constraint to bind, not the first ^mmodeled. A country can reach "25 per 1,000" by winning the post-neonatal battle alone; it cannot reach "12 neonatal per 1,000" without building the delivery-room and newborn-unit capacity that the post-neonatal wins never required. The headline child-survival number can therefore be met by a health system that has not yet solved the problem the second threshold measures.

The strongest counter is that a rising share is not deterioration and should not be read as one. Neonatal mortality did not rise — it fell 43% worldwide since 2000 ^mmodeled. The share climbs purely because the denominator shrinks faster than the numerator; presenting it as a "shift into the first month" risks implying newborns are worse off when every absolute newborn number has improved. That objection is correct, and this piece asserts a change in composition, not in newborn survival. The honest statement is narrow: neonatal mortality is falling more slowly than post-neonatal mortality, and that gap is what raises the share. A second caution is that these are modeled estimates, not death registrations. Most high-mortality countries lack complete vital registration, so UN IGME derives both rates from a statistical model over surveys and censuses, with uncertainty intervals that are wide for small or data-poor countries ^mmodeled. At the extremes the share ratio becomes unstable — several small states post shares above 80% or below 35% on tiny absolute counts, where the estimate carries little weight. The pattern reported here rests on the aggregate and on large, well-surveyed populations, not on any single small-country figure. And the share is a composition metric, not a burden metric: Switzerland's 72% newborn share sits atop an under-5 rate of 3.9, while Sub-Saharan Africa's 37% sits atop 71.2 — the region with the lowest share still carries by far the heaviest absolute newborn toll.

From a high-mortality country, the neonatal transition described here is still ahead, not underway. In Sub-Saharan Africa, where the under-5 rate remains 71.2 per 1,000 and the newborn share is the world's lowest at 37%, the deaths still to be prevented are overwhelmingly the post-neonatal, infectious kind — the vaccine-and-net frontier, not the delivery-room one. From a country that has already crossed the SDG under-5 line, the calculus inverts: the remaining child deaths are concentrated in the first 28 days, and further progress depends on maternity and newborn-care capacity that no immunization campaign supplies. The same 2030 target points two different ministries of health at two different clinical frontiers, depending only on where on the curve each country currently sits.

A health correspondent covers one country's child-mortality release at a time, and each release is a story of progress: the number is down again. Because it is falling, the composition underneath it goes unread. Only by dividing the newborn rate into the under-5 rate for all 196 countries at once does the pattern surface — that the world's child deaths have not just shrunk but relocated into the first 28 days of life, that this is now the majority of under-5 deaths in 123 countries, and that the shift is sharpest exactly where mortality is lowest. The Sustainable Development Goal wrote the asymmetry into its own scorecard: 9 countries have met the target the headlines cite while missing the one that measures the newborn week, and only 2 the reverse. The last mile of keeping children alive, nearly everywhere on Earth, is the first month.