A falling rate can still mean millions of deaths
Child and maternal mortality have declined, but counts, age-specific probabilities, a live-birth ratio, shares, and lifetime risk answer different questions.
Main takeaway. Progress is real, but it must be read through measures with explicit periods, populations, denominators, and uncertainty. No combined child-plus-maternal total is created.
Six measures, three different questions
Six cards separate counts, probabilities, and the maternal live-birth ratio. The denominator changes the question; it does not merely rescale the same number.
A falling rate can still mean millions of deaths
The global under-five mortality rate fell 60% between 1990 and 2024. The estimated number of under-five deaths fell 63%, yet 4.86 million deaths remained in 2024.
The first 28 days are a larger share of under-five deaths
Neonatal deaths fell, but more slowly than deaths later in childhood. Their share of under-five deaths rose from about 40% to 47%. Neonatal deaths are inside the under-five total.
Progress in child survival slowed after 2015
The source-reported annual rate of reduction in under-five mortality fell from 3.9% in 2000-2015 to 1.5% in 2015-2024; neonatal mortality slowed from 3.0% to 1.3%.
Maternal mortality fell, but the pace required for 2030 is much faster
The global maternal mortality ratio declined from 328 in 2000 to 197 in 2023. Reaching below 70 by 2030 would require a 14.8% annual reduction during 2024-2030 - a required scenario, not a forecast.
Rates and counts can tell different regional stories
Regional rates compare risk relative to live births, while counts show scale. The same region can look different under the two measures. Child 2024 and maternal 2023 remain separate.
Which measure answers which question?
Under-five deaths
Estimated count of deaths before exact age 5 during a year.
Under-five mortality rate
Probability of dying before exact age 5, expressed per 1,000 live births.
Neonatal deaths
Estimated count during the first 28 days; a subset of under-five deaths.
Neonatal mortality rate
Probability of dying in the first 28 days, per 1,000 live births.
Maternal deaths
Estimated count of deaths from maternal causes in the reference year.
Maternal mortality ratio
Maternal deaths per 100,000 live births, not a percentage or child mortality rate.
Adult lifetime risk
Probability that a 15-year-old woman will eventually die from a maternal cause; it reflects fertility and competing mortality as well as MMR.
Official estimates, kept in their own measurement systems
Child results use the UN Inter-agency Group for Child Mortality Estimation, Levels & Trends in Child Mortality - Report 2025, with estimates through 2024 and 90% uncertainty intervals. Maternal results use the WHO-led MMEIG report Trends in maternal mortality estimates 2000 to 2023, with 80% intervals.
Regional panels use the eight principal SDG regions. Source-native global and regional aggregates are used independently; rounded regional values are not forced to reproduce world totals.
Source lock. Seven exact official machine-readable files were validated, inspected, reconciled, and hashed. Visual Brief v1.1 applies three source-native maternal corrections: 443,000 maternal deaths in 2000; a 2000 MMR 80% interval of 309–349; and a 2023 maternal-death upper bound of 308,000.