Do More People Die When a Rural Texas Hospital Closes? What a New JAMA Study Found

Key Takeaways

• A study in JAMA Network Open, published September 28, 2026, looked at 14 rural Texas hospitals that closed between 2007 and 2018. It counted deaths from heart attack, stroke, sepsis, and asthma or COPD (a long-term lung disease).
• After the closures, these deaths rose more, on average, in the county that lost its hospital and the counties touching it than in other Texas counties with no closure nearby. The difference was 11.6 deaths a year for every 100,000 people.
• In a county of 10,000 people, that works out to about one more death a year. That is our math, for illustration only.
• Even leaving out the county that lost its hospital, the counties touching it had about 10 more deaths a year for every 100,000 people than other Texas counties with no closure nearby. When the researchers drew the area a second way and left out the county that lost its hospital, the result for the rest of that area was smaller, and it could have been a coincidence.
• The study shows a link, not proof that the closures caused the deaths. Its records end in 2019.
• Only 3 of the 14 closed hospitals were Critical Access Hospitals, going by the UNC Sheps Center closures tracker. The smallest Texas counties were left out because they had too few deaths to count.

Table of Contents

• What Did the New JAMA Study Find About Rural Hospital Closure Mortality in Texas?
• Which Rural Texas Hospitals Closed Between 2007 and 2018?
• How Many More Deaths a Year Does the JAMA Finding Mean for a Small Texas County?
• Are Deaths Higher in Counties Next to a Closed Rural Hospital?
• Why Did the JAMA Study Leave Out the Smallest Texas Counties?
• How Much Can You Rely on the Texas Rural Hospital Closure Study?
• What to Do This Week
• Frequently Asked Questions

When a rural Texas hospital closed, deaths from four emergencies rose more in its county, and in the counties touching it, than in other Texas counties with no closure nearby. Mark Poisler and Kritee Gujral reported this in JAMA Network Open on September 28, 2026. Their study leaves out the smallest Texas counties. It also gives no separate result for Critical Access Hospitals.

What Did the New JAMA Study Find About Rural Hospital Closure Mortality in Texas?

It found more deaths from four emergencies in places that lost a rural hospital. The four are heart attack, stroke, sepsis, and asthma or COPD. The researchers call these time-sensitive, because they “have response windows measured in hours or minutes.”

Here is who was compared with whom. The researchers took each county where a hospital closed, together with the counties touching it. This article calls the counties touching it the counties next door. They looked at how deaths in those counties changed after the closure. Then they compared that with how deaths changed over the same years in other Texas counties with no closure nearby. They used federal death records from 2006 to 2019, and left out 2020 because of COVID.

After the closures, deaths from these four emergencies rose more in the county that lost its hospital and the counties next door than in the other Texas counties. The difference was 11.6 deaths a year for every 100,000 people. Counting deaths for every 100,000 people is the usual way to compare places of different sizes. In a county of 10,000 people, it is about one more death a year.

Put another way, deaths from these four emergencies were 5.4% higher than the usual number those counties had before the closures.

To check the answer, the researchers also drew the area a second way. This time they used the area whose residents mostly used the same local hospitals as the closed one, instead of the counties touching it. That gave almost the same answer.

Which Rural Texas Hospitals Closed Between 2007 and 2018?

The paper’s online supplement names all 14 closed hospitals. The study counted only hospitals that closed completely. A hospital that stopped admitting patients but kept a clinic or an emergency room open was not counted as closed.

Only 3 of the 14 closed hospitals were Critical Access Hospitals, according to the UNC Sheps Center closures tracker.

The researchers put the 14 closed hospitals at “approximately 8% of rural hospitals in Texas.” None of them was in West Texas or the Panhandle. To see which critical access hospitals are open in Texas now, our guide to critical access hospitals in Texas lists them by name.

How Many More Deaths a Year Does the JAMA Finding Mean for a Small Texas County?

About one more death a year in a county of 10,000 people, and about two in a county of 20,000. This is our math, for illustration. It is not a prediction for any one hospital or county.

• A county of 10,000 people: 11.6 x 10,000 / 100,000 = about 1.2 more deaths a year.
• A county of 20,000 people: 11.6 x 20,000 / 100,000 = about 2.3 more deaths a year.

One or two deaths a year can easily get lost in a single county’s normal ups and downs. A study needs many counties to see a change that size. This one used 175 Texas counties and 14 years of death records.

Are Deaths Higher in Counties Next to a Closed Rural Hospital?

In this study, yes. The researchers ran the numbers again counting only the counties next door, leaving out the county that lost its hospital. Those counties still had about 10 more deaths a year for every 100,000 people than the other Texas counties. The researchers wrote that their findings “were not restricted to the closure county” (the county that lost its hospital).

The second way of drawing the area gave a weaker result. When the researchers left out the county that lost its hospital, the result for the rest of that area was smaller, and it could have been a coincidence.

Why Did the JAMA Study Leave Out the Smallest Texas Counties?

Because those counties had too few deaths to count. The federal death records hide any count under 10. The researchers dropped a county if, in any year, it had fewer than 10 deaths from the four emergencies. That left out 79 of Texas’s 254 counties.

Those 79 counties averaged about 4,400 people, while the 175 counties in the study averaged about 148,000 people. Most of the 79 are in West Texas and the Panhandle. None of the 14 closed hospitals was in one of them.

The study has no finding for those counties, because it could not look at them. The researchers say this “limits generalizability to the most remote communities.”

How Much Can You Rely on the Texas Rural Hospital Closure Study?

You can rely on it as evidence of a pattern, but not as proof of cause. Here is what holds it up and where it stops.

• They redid the math several ways. They tried a second statistical method, and different lists of which deaths count as emergencies. Most of these checks also found more deaths. Some found too few extra deaths to be sure they were real, and a few found none.
• It shows a link, not proof. The researchers could not account for every difference between counties. Something else that changed at the same time could be part of the answer.
• It does not say why. The researchers could not measure why deaths were higher. They suggest possible reasons, such as delays getting patients to a hospital, crowding at nearby hospitals, local health services getting worse over time, and people going without care. Earlier studies they cite found that closures led to longer ambulance trips.
• It does not measure nurse staffing. It mentions clinicians leaving after closures only in passing, citing earlier work.
• It gives no separate result for Critical Access Hospitals.
• It ends in 2019. The records stop before COVID, and before this year’s Texas Medicaid payment freeze.

The researchers also say their numbers may be too low. Some of the other Texas counties may have been affected too, if the effects reached farther than the counties next door. If so, that would make the difference they measured smaller.

In short: the pattern showed up in most of the researchers’ checks, but the study cannot say the closures caused the deaths.

What to Do This Week

1. List the hospitals you send heart attack, stroke and sepsis patients to. Next to each one, write the drive time from your door and which of the three it can treat. Bring the list to your next leadership meeting.
2. Check your emergency protocol training list against the schedule. The federal rule says “Applicable staff, as identified by the CAH, must be trained annually on the protocols” (42 CFR 485.618(e)(2)). Surveyors now check emergency protocols under CMS tag C-0896. Pull the list of staff you identified. Confirm each one is still on the schedule and has had this year’s training, and bring back any names that are missing or overdue.
3. Ask your EMS director one question. If you do not deal with the EMS director directly, ask your administrator to raise it. If the nearest hospital in each direction closed, how long would the drive be to the next hospital that treats stroke and heart attack? Bring back those drive times, route by route.
4. Look at your next four weeks of night and weekend schedules. Write down how many nurses and which provider cover the emergency room each shift. Ask your charge nurses what they would do if two extra ambulances arrived on one of those nights. Write down the gaps they name and bring them to your next staffing meeting. Our guide to night shift coverage at a small hospital covers how night coverage is set up.

Frequently Asked Questions

Q: Did the rural hospital closures in Texas cause the deaths?

The study cannot say that. It found more deaths in and around the counties that lost a hospital than in other Texas counties with no closure nearby, which is a link, not proof of cause.

Q: Does the study apply to the smallest counties in West Texas and the Panhandle?

No. It left out 79 counties that had too few deaths to count. Those counties averaged about 4,400 people.

Q: Have studies outside Texas found the same thing?

Some have. The researchers say their findings match earlier studies that linked hospital closures to more deaths in California and among Medicare patients. Other studies found no change.

Sources

1. Poisler MA, Gujral K. “Rural Hospital Closures and Mortality From Time-Sensitive Conditions in Texas.” JAMA Network Open, 2026;9(9):e2636325, published September 28, 2026. Open access under a CC-BY-NC-ND license. It is a difference-in-differences study, meaning it compares how deaths changed in one group of counties with how they changed in another over the same years. The main estimate in this article is the paper’s result for each closure county plus the counties touching it (its adjacent-county design). That result is 11.60 additional deaths per 100,000 residents a year (95% confidence interval, 4.62 to 18.58), a 5.4% increase over the pre-closure baseline. The result for the area whose residents mostly used the same local hospitals as the closed one (its hospital service area design) is 11.52 (95% confidence interval, 0.46 to 22.58). For the counties touching the closure county, with the closure county itself left out, it is 10.46 (95% confidence interval, 2.81 to 18.11). For the rest of the hospital service area, with the closure county left out, it is 9.54, not statistically significant. doi.org
2. The same paper and its Supplement 1 at PubMed Central. eTable 1 lists the 14 hospitals. eTable 3 describes the 79 left-out counties (mean population 4,439, against 148,168 for the 175 included). eTables 6, 11 and 12 hold the extra checks, and Figure 1 is the county map. pmc.ncbi.nlm.nih.gov
3. UNC Sheps Center for Health Services Research, Rural Hospital Closures tracker, updated December 4, 2025. The Medicare payment type and bed count of each closed hospital. shepscenter.unc.edu
4. Dartmouth Atlas of Health Care, Research Methods. What a hospital service area is. dartmouthatlas.org
5. 42 CFR 485.618, Condition of participation: Emergency services, paragraph (e)(2) on annual training. Cornell Law School LII

A note on who wrote this. We build nurse scheduling software for critical access hospitals in Texas. Nothing we sell keeps a neighboring hospital open or shortens an ambulance ride, and we are not going to pretend otherwise. If you want to know what we actually do, start with how it works, or read our background pages on critical access hospital scheduling and AI nurse scheduling.

Pradeep Pandey is the co-founder of SimpleScheduleAI, an AI-native nurse scheduling service built for Critical Access Hospitals in Texas. He serves as Deputy General Manager of Operations at Apollo Hospitals and holds an MBA from IIM Trichy. LinkedIn →