The United States has the highest maternal mortality rate of any wealthy nation — and unlike its peers, that rate is getting worse. This project investigates why, where intervention would matter most, and what policy exports from top-performing countries could look like in practice.
Maternal mortality is one of the only headline health indicators where the U.S. has gone backwards while every comparable country has improved. Our goal is to make the scale of that reversal unignorable — and to trace it to specific, fixable causes.
We're building a data-first investigation that (a) examines regional and racial disparities in maternal death, (b) identifies the leading causes and how they've shifted, (c) benchmarks U.S. performance against top-performing countries to surface transferable policy, and (d) investigates why U.S. rates continue to rise while peer countries' fall.
The data sources we rely on — CDC NCHS, the World Bank, OECD, and peer-reviewed work out of NEJM and NIH — each involve methodology choices. These are the priors we're operating under:
We use the WHO definition of a maternal death: a death during pregnancy or within 42 days of its end from pregnancy-related causes.
We accept the CDC's Maternal Mortality Review Committees' standard that a death is "preventable" if reasonable changes by patient, provider, facility, system, or community could have averted it.
Cross-country comparisons use OECD harmonized rates. U.S. state and county-level numbers may be noisier and we flag small-n rows.
We focus on deaths within the 42-day window. Extending to 1 year (late maternal deaths) roughly doubles the U.S. figure, but that's for a later build.
Three live visualizations drawn from our database. Each pulls fresh data on every page load; the status indicator below confirms the connection.
Each circle is one county in Arkansas, Mississippi, or Alabama. Color shows the regional maternal mortality rate (deaths per 100,000 live births); size shows the number of birth centers in that county. Hollow rings mark counties with none.
Access alone does not explain the pattern. Some of the highest-mortality regions have many facilities; some of the lowest-mortality regions have few. The crisis is not only about distance to a hospital.
These tables are intentionally empty in the HTML. JavaScript fills them from the PHP JSON feeds when the Data page loads.
| State | County | Region | MMR | Population | Birth Centers |
|---|---|---|---|---|---|
| Loading live county data… | |||||
| State | Region | Average MMR | Total Birth Centers | Counties |
|---|---|---|---|---|
| Loading live regional data… | ||||
| State | County | Birth Center Access Ratio |
|---|---|---|
| Loading live access ratio data… | ||
Placeholder for the interactive narrative experience that ships with the final build. The full storyboard is in the linked storyboard document.
The final build will open with a short cinematic intro on the home page — roughly twenty-two seconds of timed text, an animated red underline that swipes across emphasis words, and a cross-fade from black into the live regional map. The sequence frames the problem (most U.S. maternal deaths are preventable), names a region (Southeast Arkansas at 58.6 per 100,000), and asks why three different ways before handing the viewer off to the data page.
From the data page, the user can launch short guided walkthroughs — scripted cursor moves over the live charts that highlight key findings (the access-vs-MMR contradiction, the state-by-state birth-center gap), then return control to the user. Skippable, replayable, and built on the same animation grammar as the intro.
We put the M2 build in front of five testers spanning a public-health graduate, a journalist, a designer, and two general-audience readers. Sessions ran 12–18 minutes. Each tester read the home page aloud, then explored the Data page without prompts. The full notes and verbatims live in a separate document; the action items below feed directly into M4.
The foundational material that informs every chart on this site. Peer-reviewed research, official statistical agencies, and non-partisan policy institutes only.
Clinical-review analysis finding that roughly 84% of U.S. maternal deaths are classified as preventable, even as U.S. MMR rose 136% between 1990 and 2013 while other developed nations' rates fell 38%.
Tracks state-level maternal mortality in the wake of the Dobbs decision: states with abortion bans saw an average 56% increase in maternal mortality; states with protective policies saw a 21% decrease.
Official U.S. vital-statistics data showing Black women die at 4.5× the rate of Asian women — the largest racial mortality gap for this metric of any high-income country.
Cross-country harmonized data 2000–2023. Documents the U.S.-specific finding that homicide, suicide, and drug overdose have replaced hemorrhage as leading causes of U.S. maternal death (2018–2023).
Policy-oriented comparison of maternity care across 11 high-income nations, identifying universal paid leave and midwife-led care as the two strongest correlates of lower MMR.
Long-form investigative reporting pairing statistics with named patient stories. Useful for making aggregate numbers legible and for triangulating what official data often misses.