Irene Quarshie, M.D, MPH, MBAIrene Quarshie, M.D, MPH, MBA
Mar 8, 2026
Maternal mortality dropped 77% in 13 years.
No new hospitals. No specialists imported.
Rwanda held an election instead.
-----
In 1994, Rwanda emerged from genocide
with almost no institutions standing.
⭕ No hospitals in most villages.
⭕ No specialists.
⭕ No infrastructure.
⭕ Almost no money.
And a maternal mortality crisis
that demanded an answer.
🔵 So they asked a question:
Who does this mother already trust?
And can we build around that?
The result: community health workers.
Launched 1995.
Not assigned by a ministry.
Not credentialed by a licensing board.
🟢 Elected by neighbors.
🟢 Community members literally lining up
in front of the person they trusted most.
🟢 One maternal health worker per village.
Every pregnant woman, identified.
Every new mother, followed.
By someone who already knew her name.
The intervention wasn't clinical:
It was relational.
✅ Continuity of relationship is an intervention.
✅ Geographic accountability is an intervention.
✅ Removing the mother's burden
of navigating the system is an intervention.
All of them required a design decision.
✅ Rwanda's trajectory:
2000: 1,020 per 100,000 live births
2013: 320 per 100,000 live births
2020: 259 per 100,000 live births
Now run that math everywhere else.
🍃 A 77% reduction from 2023 figures:
US: 18.6 → 4.3
Ghana: 234 → 53.8
Nigeria: 1,047 → 241
The US rate for Black women: 50.3
A 77% reduction: 11.6
Lower than the overall US rate today.
When Rwanda's numbers began to plateau,
they didn't abandon the model.
They added to it.
Phase 1: design around who is already there.
Phase 2: build specialist capacity on top of that foundation.
Most systems try to start with Phase 2.
Rwanda started with Phase 1.
The foundation held.
The specialists had something to build on.
♻️ What phase is your system in?
Share with someone in maternal care.
If you're working in maternal care,
or have a maternal care story,
this newsletter is for you- real stories from real women: