Maternal Mortality: the denominator problem
Global maternal mortality kept falling after 2015 — the apparent stall was COVID. What is real is that progress is being partly masked by where the world's babies are born.
The story everyone carries is that the world made huge progress on women dying in childbirth, then stalled around 2015 — and that the United States is a uniquely worsening rich country. Extended to 2023, neither claim survives. The stall was the pandemic, and it resumed; the US spike reversed. What is real is stranger and less comfortable: every income group is improving, several of them fast, while the world average barely moves — because the share of the world's births happening in the highest-risk countries keeps rising. About a third of the achievable improvement is being absorbed by that shift. It is not a story about anyone failing. It is a story about a denominator.
Maternal mortality data: UN Maternal Mortality Estimation Inter-Agency Group, *Trends in Maternal Mortality 2000–2023* (2025), via World Bank Open Data (`SH.STA.MMRT`). Births: UN World Population Prospects via Our World in Data (CC BY).
Every income group improved
yearly2015 → 2023, every group moved the right way. Low income −29.5% (491→346), the fastest-improving group in the world. Lower-middle −19.2% (291→235). Upper-middle −6.6% (61→57). High income −9.1% (11→10). Hold that against the world's −13.6% over the same years — slower than most of its own components. That is arithmetically impossible unless the mix moved, which is the next panel.
@andrewdun · 2h ago
The denominator moved
yearlyBirths per year, in millions (UN World Population Prospects). Low-income births rose from 21.8M in 2015 to 25.29M in 2023, while upper-middle fell from 39.8M to 28.5M and high income from 13.2M to 11.5M. The low-income share of world births went 15.2% → 19.2%. Weight the group rates by births and the world improves −14.5%; hold the 2015 birth-mix fixed and the same group improvements give −21.6%. The gap — a composition drag of about +9.0%, roughly 16 MMR points — means about a third of the achievable improvement is offset by where babies are now born. Every country can get safer while the world barely moves.
@andrewdun · 2h ago
Two rich countries, opposite directions
yearlyThe United States reads 17 in 2015, 22 in 2020, 31 in 2021, and 17 again in 2023 — the spike reversed, so the "uniquely worsening rich country" framing does not survive current data. A caveat that matters: US figures across 2003–2017 carry a measurement-change annotation (the phased pregnancy-checkbox rollout on death certificates), and we have not verified how the modelled series handles it — so no precise magnitude claims across that window. Australia runs 6 in 2015 to 2 in 2023, but at ~300,000 annual births an MMR of 2 is order-10 deaths: the 30-year direction is real, the precise ratio is not.
@andrewdun · 2h ago
Discussion of this dashboard
Threads here are about the dashboard as a whole — its framing, what's missing, whether the constituent indicators tell a coherent story. Per-indicator discussion lives under each panel above.
Log in to start a discussion about this dashboard.
No discussions yet. Be the first.
Maternal deaths per 100,000 live births. The world fell from 460 in 1985 to 328 in 2000 to 197 in 2023 — a 40% decline since 2000. The bump at 211 in 2020 is the pandemic, not a plateau: a series that stops there reads COVID as a structural break, which is exactly the error this dashboard exists to correct. Sub-Saharan Africa sits at 448 in 2023 against a world figure of 197, and carries roughly 70% of all maternal deaths (UN MMEIG, *Trends in Maternal Mortality 2000–2023*).
@andrewdun · 2h ago