And Then You're Dead

The Autopsy / Case File · Maternal & Infant

Postpartum Hemorrhage

The United States' overall pregnancy-related death rate has been falling since its 2021 peak. One specific cause hasn't followed that curve down. Deaths from postpartum hemorrhage have climbed in a nearly straight line every year since 2017 — through a pandemic, past it, and into the years the broader numbers finally started to improve.

A study of two decades of CDC Pregnancy Mortality Surveillance System data, published in Obstetrics & Gynecology in August 2026, found hemorrhage deaths rising at an average annual rate of 6.84% between 2017 and 2024 — a statistically significant trend distinct from, and running counter to, the broader decline the same dataset shows for pregnancy-related deaths overall since 2021. The mechanism behind most of those deaths is well understood, and a hospital protocol shown to reduce hemorrhage-related harm by nearly a fifth has existed for over a decade. It simply hasn't reached every delivery room in the country at the same pace.

And Then You're Dead · September 2026 · 9 min read

Join the discussion →
A dimly lit hospital patient room with a bed, IV pole, and bedside monitoring equipment, photographed from the doorway
A hospital patient room equipped with the kind of bedside monitoring and IV equipment used to track a patient's vital signs after delivery — the window in which most postpartum hemorrhage becomes clinically apparent.

Postpartum hemorrhage is clinically defined as cumulative blood loss exceeding 1,000 milliliters after birth, or any blood loss accompanied by signs of hypovolemia — a dangerous drop in blood volume — within 24 hours of delivery, according to the American College of Obstetricians and Gynecologists. The single most common underlying cause, accounting for an estimated 70% to 80% of cases, is uterine atony: the uterus's failure to contract firmly after delivery, which normally clamps down on the blood vessels left exposed where the placenta detached. When that contraction doesn't happen strongly enough, those vessels keep bleeding, sometimes rapidly enough to become life-threatening within minutes. Less common causes include birth-canal trauma, retained placental tissue, and underlying clotting disorders — but atony's dominance of the case count is precisely what makes hemorrhage one of the more mechanically well-understood, and in principle more preventable, causes of pregnancy-related death.

A Rising Trend Inside a Falling One

The August 2026 study, led by Fanny Njie of Tanaq Support Services and published in Obstetrics & Gynecology, examined 14,310 pregnancy-related deaths identified through the CDC's Pregnancy Mortality Surveillance System across a full 20-year window, 2005 through 2024. Cardiovascular conditions, infection or sepsis, and hemorrhage emerged as the three most common underlying causes across the entire period, together with cardiomyopathy, thrombotic embolism, and hypertensive disorders forming the six leading categories responsible for the large majority of pregnancy-related deaths nationally.

What distinguishes hemorrhage from the other leading causes is its trajectory in the second half of that period. The study's Joinpoint regression analysis — a statistical method for detecting when a trend's slope changes significantly — found that overall pregnancy-related mortality rose sharply from 2018 to 2021, driven substantially by the COVID-19 pandemic's direct and indirect effects on maternal care and infection risk, before declining again from 2021 to 2024 as pandemic-era conditions eased. Hemorrhage deaths didn't follow that same arc. Instead, they rose at an annual percent change of 6.84% from 2017 straight through 2024 — a trend that predates the pandemic, continued through it, and, unlike the broader pregnancy-related mortality rate, hasn't reversed since.

0%
annual percent increase in hemorrhage-related pregnancy deaths from 2017 to 2024 — a statistically significant rise (P=.02) running counter to the overall pregnancy-related death rate's decline since 2021
0%
reduction in severe maternal morbidity from hemorrhage at California hospitals that adopted a standardized hemorrhage-response bundle, versus a 1.2% reduction at non-participating hospitals in the same state

The Bundle That Already Works

The clearest evidence that hemorrhage deaths are more preventable than their rising trend line suggests comes from California, which built one of the country's first large-scale hospital quality-improvement collaboratives specifically targeting obstetric hemorrhage. A study led by Elliott Main and colleagues, published in the American Journal of Obstetrics and Gynecology in 2017, tracked hospitals that adopted a standardized hemorrhage "bundle" — a checklist-driven protocol covering early risk assessment, a hemorrhage cart stocked and ready on every unit, rehearsed response drills, and clear escalation triggers tied to measured blood loss rather than clinical impression alone. Hospitals that implemented the bundle between 2014 and 2016 saw severe maternal morbidity from hemorrhage fall by 20.8%, compared to just a 1.2% reduction at hospitals that hadn't yet adopted it — a gap large enough to indicate the bundle itself, not some unrelated statewide trend, was doing the work.

That California collaborative became the template for the national Alliance for Innovation on Maternal Health, or AIM, a federally authorized program — codified into law as part of the Consolidated Appropriations Act of 2022 — that funds state-level teams helping hospitals implement the same category of maternal safety bundles, hemorrhage response chief among them. Adoption has spread, but unevenly and on its own state-by-state timeline: Michigan's AIM hemorrhage-bundle participation among birthing hospitals rose from just 15.9% in January 2018 to 82.7% by December 2024, a genuine transformation of that state's obstetric units over six years — but a transformation that took six years, in one state, to reach a still-incomplete 83% of facilities. Other states report participation rates near-total among delivering hospitals; others lag well behind. There is no single national figure showing what share of U.S. births now happen inside a hospital running a fully implemented hemorrhage bundle, and that absence is itself informative: the country doesn't currently track, in one place, how close it has actually gotten to making the intervention it already knows works into a universal standard of care.

The bundle isn't experimental. It's been measured, replicated, and scaled into federal policy. What it hasn't been is delivered everywhere, at once, on the same clock — and the national hemorrhage-death curve is still climbing in the gap that leaves behind.

Where the Deaths Are Concentrated

Postpartum hemorrhage deaths don't fall evenly across the population that gives birth in the U.S. every year. CDC data on the broader maternal mortality landscape has repeatedly documented that Black women face a pregnancy-related death rate roughly 3.5 times higher than white women, and that gap is not spread evenly across every cause of death — it concentrates disproportionately in specific categories, hemorrhage and cardiomyopathy among them, according to CDC's national maternal mortality reporting. That concentration matters for how hemorrhage-bundle adoption gaps translate into outcomes: a hospital system that hasn't fully implemented standardized hemorrhage response isn't a neutral risk applied evenly to every patient who delivers there. It's a risk that lands hardest on populations who are, for a range of separate and well-documented reasons — including implicit bias in symptom recognition and unequal access to the best-resourced facilities — already more likely to experience a hemorrhage event escalate before it's caught.

A Second Proven Fix, Also Time-Sensitive

Alongside the hemorrhage bundle, a second evidence-based intervention has separately been shown to reduce hemorrhage deaths, with an even more precisely measured effect: tranexamic acid, a generic, inexpensive clotting-stabilizer drug already used widely in trauma medicine. The WOMAN trial, a randomized, placebo-controlled study spanning 193 hospitals in 21 countries and published in The Lancet in 2017, found that tranexamic acid reduced deaths from bleeding among postpartum hemorrhage patients from 1.9% to 1.5% overall — and when the drug was administered within three hours of giving birth, the reduction in bleeding deaths reached roughly 30%, a substantially larger effect than the trial found for later administration. The timing dependency is the clinically important detail: tranexamic acid works by stabilizing existing blood clots rather than by stopping bleeding outright, which means its benefit erodes the longer a hemorrhage goes unrecognized or unmedicated after delivery. Like the hemorrhage bundle itself, the drug's value depends entirely on a hospital's staff recognizing a hemorrhage is happening and acting inside a narrow window — not on inventing a treatment that doesn't already exist.

The Timing Window Most People Don't Expect

Postpartum hemorrhage is often pictured as a delivery-room emergency, and the most severe atony-driven cases usually do present within the first 24 hours after birth, matching the clinical definition's own time window. But CDC's broader pregnancy-related mortality surveillance has documented that a substantial share of pregnancy-related deaths from all causes, hemorrhage included, occur later — from one day up to a full year after delivery, a period during which postpartum medical follow-up is often thin to nonexistent by the time a delayed hemorrhage-related complication, retained tissue, or a secondary bleeding event might present. A hospital's hemorrhage bundle is built almost entirely around the acute, in-hospital delivery window. The country's broader postpartum follow-up infrastructure — how many mothers are actually seen by a clinician in the weeks after discharge — has a much weaker evidence base showing it catches the later-presenting cases with anything like the same rigor.

And Then You're Dead

The uterus either contracts or it doesn't — a single physical event, hours long at most, with a checklist that's been tested and shown to change the outcome. What determines whether that checklist was actually running in the room isn't the mother, or the pregnancy, or the birth. It's which hospital she happened to be standing in when the bleeding didn't stop.

That's it. That's the whole thing.

Comments

Loading…

Comments are anonymous and moderated.

Sources

Photo: "2011-03-04 Labor - the delivery room.jpg," by George Ruiz, via Wikimedia Commons (CC BY 2.0).