Every hospital in the state that stopped delivering babies since 2005, year by year, from the Pennsylvania Department of Health's own count of births at each hospital. Where we could confirm it, each one shows who owned the hospital when the births ended.
Beaver, Butler and Westmoreland counties each have one hospital that delivers babies, and all three changed owners this year or are about to. All three are going to systems tied to a health insurer. None of the new owners has promised to keep delivering babies. See the hospitals to watch
Hospital records, PA Department of Health Severe Maternal Morbidity report (2024) and Maternal Mortality Review (2026). "1 in" figures are rounded from the state's rates.
Drag the slider or press play. Counties turn rose when their last hospital birthing unit is gone. A dark outline marks a county that lost a unit that year.
What this meansA rose county has no hospital where a baby can be born. A mother there has to drive to the next county, sometimes farther, to give birth. In 2005 there were 10 such counties. Today there are 23, and nearly all are rural.
What this meansBirths in Pennsylvania fell 13 percent since 2005. Hospitals delivering babies fell 37 percent. The same mothers now share far fewer places to give birth, and closures came in every year but one. This isn't a one-time shock. It's a steady drain.
We extended the last two decades forward. The shaded band covers four reasonable ways of drawing the trend: a steady loss in number or a steady loss in percentage, fitted to 2005 onward or to 2015 onward. This isn't a forecast. It's where the current path leads if nothing changes.
What this meansOn today's path, by 2046 more than half of Pennsylvania's 67 counties would have no hospital where a baby can be born. The trend isn't fate. Units close one decision at a time, and decisions can be changed.
Severe complications per 10,000 deliveries, by the county where the mother lives, weighted by births. "Lost their unit": Bedford, Clarion, Clinton, Fayette, Huntingdon, McKean and Somerset, which lost their last birthing hospital between 2016 and 2021. Small numbers: read as an early signal, not proof.
What this meansSevere complications per 10,000 deliveries ran 75.2 in 2016, peaked at 110.0 in 2021, then fell to 105.2 in 2022 and 99.9 in 2023. The rise was real and the peak has passed, so this page no longer projects the 2016 trend forward. In the counties that lost their last birthing hospital, complications rose about twice as fast as in counties that kept one. National research points the same way: after rural counties lost hospital obstetric care, more babies were born early and more were born outside a hospital (Kozhimannil and colleagues, JAMA, 2018).
What could change the path: a state rule requiring notice and a community plan before a unit closes, payment that covers a small unit's round-the-clock readiness, and reopenings like Uniontown's in 2025.
Closures are one side of the story. This is the other: how many Pennsylvania mothers die or nearly die around birth, and why. All from the Department of Health's own reports.
2013 to 2018: death certificates linked to birth records. 2020 to 2022: cases reviewed by the state's maternal mortality committees. The two methods differ, so read the trend, not the step. 2019 not published.
53 of the 122 deaths were pregnancy-related: they came from a condition connected to pregnancy.
What this meansMost of these mothers didn't die in the delivery room. Of the 53 pregnancy-related deaths in 2022, 26 came between six weeks and a year after birth. Mental health crises caused 19 of those 26. Cardiac conditions caused 3, so heart disease is not what is killing women in that window. The state's own committees judged nearly nine in ten of these deaths preventable. A maternity system that ends at the hospital door can't catch them.
Pregnancy-related deaths per 100,000 live births, Pennsylvania 2020 to 2022. Rural numbers are small and uncertain.
What this meansBlack mothers in Pennsylvania die of pregnancy-related causes at twice the rate of white mothers, and mothers over 35 at nearly twice the rate of younger mothers.
Severe complications per 10,000 deliveries (hemorrhage with shock or hysterectomy, kidney failure, sepsis, heart failure, eclampsia and others; transfusion alone not counted), from hospital records. Hospitals delivering babies from state birth data.
Severe complications per 10,000 deliveries. Some of the gap is because sicker mothers more often need a cesarean.
What this meansTwo lines moved in opposite directions: fewer places to give birth, and more mothers with life-threatening complications. That doesn't prove one caused the other. It does mean mothers are facing more danger with fewer places to turn. Nearly one in three Pennsylvania babies is born by cesarean, and cesarean births carry about four times the rate of severe complications.
Share of all births, from the Department of Health's births by method of delivery.
Percent of births with an abnormal newborn condition recorded, which includes NICU admission, breathing support and treatment for suspected infection.
First baby, full term, one baby, head down. Pennsylvania residents, CDC natality data (CDC WONDER), 2016 to 2024. Dashed line: the national Healthy People 2030 target of 23.6%.
Pennsylvania residents, 2016 to 2024, CDC natality data. Read with care: a planned home or birth center birth that transferred in labor is counted as a hospital birth, mothers planning out-of-hospital births are usually lower risk, and a newborn admitted after a home birth may not be recorded on the birth certificate.
What this meansMore than one in four first-time mothers with a low-risk pregnancy in Pennsylvania has a cesarean, above the national goal and rising since 2016. A first cesarean usually leads to more: four in ten cesareans in the state are now repeats. The NICU chart shows why the state needs to record where a mother planned to give birth. Without it, a planned home birth that transferred looks like a hospital birth, and an emergency birth on the way looks like a home birth, so the numbers can't answer the question families actually ask.
The birth certificate is filed for every baby born in Pennsylvania, in every setting. It's the one record that could show how mothers fare at home, at a birth center and in the hospital. Right now it can't.
Hospital records: severe complications per 10,000 deliveries (PA SMM report). Birth certificate: births with any maternal complication checked, per 1,000 (PA birth tables). The two records count different lists of complications, so this is two different questions rather than one question answered twice.
Share of complications in hospital records that the birth certificate missed. Massachusetts, 2011 to 2013 (Luke and colleagues, Epidemiology, 2018).
What this meansHospital records say severe complications are rising. The birth certificate says they're falling. When researchers checked, birth certificates missed 88% of blood transfusions and 80% of ICU admissions. The birth certificate also can't tell a certified professional midwife from a traditional one, and it doesn't record when a mother was transferred, why, or how long it took. Pennsylvania is making decisions about maternity care without seeing what happens at the handoff.
Much of what families, reporters and legislators would need to judge maternity care already exists. It's collected, filed, and kept out of public view. Here is what we looked for and couldn't find published.
| What's missing | Who already has it | What's public | What we're asking |
|---|---|---|---|
| Serious harm during childbirth, by hospital | Hospitals and birth centers must report serious events within 24 hours to the Department of Health and the Patient Safety Authority (MCARE Act, 2002). | None of it is published by facility. | Publish childbirth serious events by facility each year, with patients unidentified. |
| Severe complications, by hospital | PHC4 collects every hospital discharge. Since 2024, Medicare (CMS) also collects each hospital's cesarean rate and severe complication rate. | The state publishes statewide and county totals only. For calendar 2024, the first mandatory year, every hospital in the country shows "Not Available." Some rows carry the footnote "Data suppressed by CMS for one or more quarters"; the rest read "Results are not available for this reporting period." CMS says hospitals reported results much higher or lower than expected and that a number had mapping issues, so it is not publicly reporting 2024 results for these two measures. | Publish each hospital's severe complication rate, adjusted for how sick its patients are. |
| Maternal deaths, with where care happened | The Maternal Mortality Review Committees review every death during pregnancy or within a year. | Reports are statewide totals, years late, and no report for 2019 deaths was found. In the 2024 report, only 10% of cases had complete records to review. | Publish on time, and report whether a transfer or a closed unit was part of the story. |
| When a hospital stops delivering babies | Hospitals know months ahead. | No statewide notice or public list. Only Philadelphia requires notice. This map is the only year-by-year list we found. | 180 days' notice, a county hearing, and a public list kept by the Department of Health. |
| Transfers between home, birth center and hospital | Midwives, EMS and hospitals each keep their own records. | The birth certificate doesn't record when a mother was transferred, why, or how long it took. No one adds it up. | Add transfer, time and reason to the birth certificate. |
| Where a mother planned to give birth, and with whom | Known to her and her midwife or doctor. | The birth certificate records where the baby was born, not where the mother planned to. The box reads “Lay midwife,” which lumps certified professional midwives with traditional midwives under a term neither uses. | Add planned place of birth and midwife credential to the birth certificate. |
| Complications, counted accurately | Hospital records capture them. | The birth-certificate checkboxes miss most of them (88% of transfusions and 80% of ICU stays in one study), and home and birth center outcomes live only in voluntary midwife registries. | Link birth certificates to hospital records and publish the linked counts. |
| Whether a hospital is ready for an emergency birth | Each hospital knows its own anesthesia coverage, blood supply and operating room readiness. | Not reported publicly for any Pennsylvania hospital. | A yearly readiness report: 24-hour anesthesia, blood on site, time to the operating room. |
| What a maternity unit costs and earns | Each hospital's books. | PHC4 publishes whole-hospital finances only, so no one can check a claim that a unit was losing money. | Require unit-level finances in any request to end births. |
| How far mothers now drive | Nobody measures it. | No state report tracks drive times after a unit closes. | Require drive-time impact in every closure plan, and publish it. |
What this meansA mother choosing where to give birth can't see how often a hospital has a serious harm event, how often its patients have severe complications, or whether it can start an emergency cesarean at 2 a.m. The state already collects most of this. It just doesn't publish it.
"Not published" means we searched the Department of Health, PHC4 and Patient Safety Authority sites and couldn't find it in public form as of September 2026. If you know where it's published, tell us and we'll correct this page.
The Pennsylvania Health Care Cost Containment Council (PHC4) publishes every hospital's finances each year. We looked up each hospital in the year births ended, and the year before, going back to 2009.
What this meansAbout half of these hospitals were losing money, and that is the reason usually given. The other half were making money, some of them well above the state's typical hospital. Holy Spirit's overall margin was about 22 percent the year its births moved. For those hospitals, ending births was a choice about where to deliver babies, not a sign the hospital was failing.
Four kinds of owner, side by side. This shows who was responsible for each hospital when births ended. It doesn't say why, and every kind of owner appears here.
What this meansEvery kind of owner has closed maternity units. But at least 21 of the 31 closures since 2015 happened at hospitals owned by a larger system or company, where one board decides for many hospitals at once. That's why this needs a statewide answer.
Birthing hospitals that changed owners in 2026 or are about to. Most closures since 2015 came after a hospital joined a larger system or company, so these are the units to keep an eye on.
Allegheny Health Network, which is owned by Highmark, an insurer, took over Heritage Valley on July 1, 2026. Heritage Valley ended births at its Sewickley hospital in 2023.
What was promised for births: None reported. AHN committed to keep the Beaver hospital open "as long as there is sufficient demand for its services." Heritage Valley Beaver also told Medicare it does not take part in a maternal safety quality program.
WVU Health System, which co-owns the Peak Health insurance plan, is buying the five-hospital Independence Health System. The deal is expected to close in late September or early October 2026. Independence’s Clarion hospital already ended births in 2021.
What was promised for births: The Attorney General’s August 31 settlement keeps the hospitals open for five years and requires 120 days’ notice before obstetrics is discontinued. That is a warning period, not a promise to keep delivering babies.
Community Health Systems, a for-profit chain, left Pennsylvania and sold Moses Taylor, Regional Hospital of Scranton and Wilkes-Barre General to Tenor Health Foundation, a nonprofit. Wilkes-Barre General had already ended births in 2023. Tenor also runs Sharon Regional, where births ended in 2021.
What was promised for births: None found. Reporting at the sale said there were no plans to restore births at Wilkes-Barre General.
What this meansBeaver, Butler and Westmoreland counties each depend on a single hospital to deliver babies, and all three hospitals changed owners this year or are about to. If one of those units closes, that county joins the 23 with none. No new owner has promised to keep delivering babies.
Medicare (CMS) now asks every hospital two things about childbirth. Here's what it shows for Pennsylvania, from CMS's public hospital data (2024 reporting period, published July 2026).
What this meansHeritage Valley Beaver is the only hospital in Beaver County that delivers babies. It told Medicare it does not take part in a maternal safety program, and in July 2026 it came under new ownership, by a system owned by an insurer, with no promise about births. The public can't see its complication rate, or any other Pennsylvania hospital's, because CMS published no 2024 results for any hospital in the country.
Source: CMS Provider Data Catalog, "Maternal Health - Hospital" (measures PC-02, PC-07a, PC-07b, SM-7) and "Birthing Friendly Hospitals," accessed September 19, 2026. SM-7 is each hospital's own yes-or-no answer.
| Births ended | Hospital | County | Owner then | What happened | Money and where births went | Sources |
|---|
What this meansEvery number on this page traces to a public record: the state's own birth counts, and the owners' own words. Anyone can check any row.
Keystone Birth Collaborative is a Pennsylvania nonprofit corporation working to make birth safer across every setting in the Commonwealth. We are independent: not funded by or tied to any hospital system or health insurer. keystonebirthcollab.org
Data current as of September 19, 2026. Download the full table as a spreadsheet (CSV). Found an error? Tell us and we will fix it and note the change.