What Happens When a Hospital Closes?

What happens when a hospital closes? A hospital in a rural area, where there aren’t a lot of other hospitals nearby? This is a situation that tens of millions of Americans have faced over the last two decades. Since 2005, more than 180 rural hospitals in the United States have closed, and more than 800 rural hospitals are at risk of closing in the near future. You’d expect that these closures would lead to more people going without hospital care, and maybe also worse outcomes for patients in that region.

And that’s exactly what a 2018 study showed. It was published in the Journal of the American Medical Association, and it looked at what happens when an obstetric unit closes in a rural hospital. The researchers found that those areas saw more babies born outside of the hospital or in hospitals without obstetric units, and more babies born early, before 37 weeks.

WHITE: That paper was actually kind of one of the inspirations for our work.

Corey White is an economist at Monash University in Australia. He studies access to health care, and he’s been keeping an eye on rural hospital closures in the United States.

WHITE: We thought, “There hasn’t been a study that’s looked at a really long time period and a wide range of different outcomes. But there’s this potentially really interesting trade off involved in these rural obstetric unit closures.

This interesting trade off that Corey’s mentions is that not all rural hospital closures are the same. Some closures could be bad, but others could have unexpected benefits for patients, and might even improve the quality of their care.

But does that silver lining happen outside of obstetrics?

CARROLL: There are significant costs when a rural hospital closes and patients have to travel farther for care. But, of course, that’s not the whole story.

Radio Network, this is Freakonomics, M.D. I’m Bapu Jena. I’m an economist and I’m also a medical doctor. Each episode, I dissect an interesting question at the sweet spot between health and economics. Today: What do rural hospital closures mean for the 46 million Americans who rely on these hospitals for care?

WHITE: We haven’t as a community, had a great understanding of what the effects of these closures are.

CARROLL: I’m from Lincolnville, Maine. It’s a town of about 2,000 people, about two hours up the coast from Portland.

Caitlin Carroll grew up in a rural area, which is how she got interested in researching rural health care and hospital quality.

CARROLL: I actually had two hospitals relatively close by when I was growing up, in sort of vaguely different directions — both about 15 to 20 minutes away, driving.

Caitlin’s experience is on par with other rural Americans. Where I used to live in Boston, there were four hospitals within a half-mile of my house. The situation is very different in rural communities, where the average travel time to the nearest hospital is about 17 minutes. That’s seven minutes longer than it is in cities, where the hospitals might also be larger and better equipped.

According to the Centers for Disease Control and Prevention, people who live in rural areas are more likely to die from heart disease, cancer, unintentional injury, chronic respiratory disease, and stroke than their urban counterparts. These disparities are at least partially due to the long distances rural residents need to travel for specialty or emergency care.

A lot of dynamics are at play when a rural hospital, or part of a rural hospital, closes. It’s a big deal in a local community, but not all patients are affected in the same way. So, why do these hospitals typically close? And then, what happens to patients?

CARROLL: A key challenge for rural hospitals is low patient volumes. When hospitals have low patient volumes, it limits revenues, and makes it really hard to cover the high cost of care delivery in the hospital setting. So, these low patient volumes can lead hospitals to become unprofitable and in extreme cases, can lead to closure.

JENA: Tell me about your research on what happens when rural hospitals close.

CARROLL: My work looks at the effects of rural hospital closure on Medicare beneficiaries. So, we’re thinking about an older population. And I find that two things happen when a rural hospital closes. The first thing that happens is that Medicare spending goes down. And this is entirely driven by a decrease in hospital admissions. When people lose their closest hospital, they get less care. The second thing that happens when a hospital closes is that mortality rates go up for patients with time-sensitive health conditions, like a heart attack or a stroke. So, these patients have to travel farther for care and their health outcomes get worse.

JENA: How do they get worse? What are the sorts of outcomes that you’re talking about?

CARROLL: I think about outcomes in terms of mortality. So, what share of people die within one year, specifically among patients that go to a hospital for a time-sensitive condition? For example, if you show up at a rural hospital for heart attack treatment, how likely are you to survive for the whole year following that?

JENA: So, you find that heart attack mortality goes up when someone who lives near a rural hospital, when the hospital closes.

CARROLL: That’s right. I look not just at heart attack mortality, but I look at a group of time-sensitive health conditions. So, heart attack, stroke, heart failure, in some cases, asthma attacks — things like that. Conditions where you really need care as soon as possible.

JENA: Do you have an example of the kinds of patients who don’t suffer? What types of patients are we talking about?

CARROLL: Well, for example, a patient with an urgent condition that’s not time-sensitive might be somebody who fell and broke their hip or fractured their hip. Certainly, it’s important for them to get care quickly, but it’s not a minute-by-minute issue in terms of “What is their mortality going to be?” We don’t think that when the hospital closes, those people are more likely to die if they have to travel another, say 15 minutes for care.

JENA: Are there any outcomes that don’t get any worse or that actually get better?

CARROLL: I look at mortality rates among patients with these time-sensitive health conditions. Those outcomes get worse. But then I also look at outcomes among a broader set of patients that have any urgent condition. And when I look at mortality rates in that group, I find no change. So, it doesn’t actually look like they’re getting worse before and after the hospital closure. The mortality effect is really concentrated among people where getting to the hospital as soon as possible is really important.

JENA: Is it the case that there might be a tradeoff? So, if you close a hospital, you reduce Medicare spending — that’s because people don’t go to the hospital — that harms people who have time-sensitive conditions, but people who don’t have time-sensitive conditions — they aren’t any worse off?

CARROLL: That could be true. That’s not something I look at specifically in my work, but I think it’s a great question. One interesting question for future work is how hospital closure impacts patients with, say, a more complex or chronic condition. Cancer jumps out to me as a really interesting area for future research. Is it better to have local access to a close-by facility, or is it better to travel farther to a potentially higher quality facility?

JENA: Caitlin, is it always bad to close a rural hospital?

CARROLL: It’s difficult to say. I would suspect that the answer is no — that sometimes when you close a rural hospital, it can be quite harmful, and creates real declines in access to care. In other situations, you might think that, if the rural hospital is nearby to, say, a higher quality facility, then there could be a real tradeoff between the decreased access to care and higher quality treatment at a nearby facility.

CARROLL: What I find is that when rural hospitals close — specifically these small rural hospitals that tend to be fairly isolated — patients do in fact have to travel farther for care, but they don’t actually receive treatment at a higher quality facility. And when I dig into this what I find is patients don’t receive care at a higher quality hospital, largely because there isn’t one in their local area. So, it’s not as if the first-closest hospital is a small rural facility and the second-closest hospital is M.G.H.

JENA: Thanks for giving a shoutout to M.G.H.

CARROLL: The second closest hospital, you know, it tends to be another small rural facility that looks pretty similar in terms of quality compared to the closed hospital.

JENA: That’s super interesting, because that means that there’s a heterogeneity in this effect, right? And maybe there could be this benefit when the closest hospital that someone would go to A) is not so far away, but B) is higher quality, but at least from the hospitals that you looked at, that wasn’t really on the table.

CARROLL: My work shows that there are these significant costs when a rural hospital closes and patients with time-sensitive health conditions have to travel farther for care and their mortality rates increase. But, of course, that’s not the whole story. It’s not as if we can flip a switch and decide, okay, these hospitals stay open and these hospitals close. We need to think about, how much does it cost to keep a rural hospital open? And could we better use those resources in some other way to promote rural health?

JENA: Do you have a view on whether or not standalone emergency departments may then make sense in areas where rural hospitals close? Like, something that’s not quite the hospital level, but an emergency department that is sufficiently well-resourced, has the infrastructure to stabilize patients and get them to where they need to be.

CARROLL: A key issue in rural areas is what services should be provided locally and what services should be provided regionally. For some services like emergency care, there’s good evidence that local access is important.

What about other services that aren’t emergencies? Coming up, find out what happens when an obstetric unit closes in a rural hospital, and pregnant women need to travel farther for care.

WHITE: It might just be that, accidentally, providers were using the best practices, but we didn’t know that those were the best practices at the time.

I’m Bapu Jena, and this is Freakonomics, MD.

*          *          *

When a rural hospital closes, there are downstream effects, and as Caitlin Carroll told us, not all patients may be affected in the same way. My next guest’s work has shown that in the last 30 years, over 400 counties in the U.S. have lost their sole hospital-based obstetric unit. When an O.B. unit closes, pregnant patients are forced to deliver their babies in other hospitals, farther away from where they live, or outside of a hospital setting altogether. They might get a different kind of care before they deliver, too.

Recently, Monash University economist Corey White and his colleagues wondered how these events leading up to a birth could impact outcomes after. That includes for the pregnant patient and the baby being born. Does closing an OB unit in a rural hospital put them at risk? And what can we learn when closures force changes in how care is delivered?

WHITE: We know that rural hospitals tend to be smaller, tend to have fewer resources, and they tend to do a little worse on a range of quality metrics relative to their urban counterparts. So it could be that when these rural obstetric units close, mothers are redirected to hospitals that have more resources and offer higher quality care. That’s fundamentally what we were interested in — this tradeoff between the fact that yes, women might have to travel farther to receive care after the closure occurs, but the quality of the care they might receive could be higher

JENA: Do you have a sense of why a rural OB unit would close?

WHITE: There’s not one answer to that question. If I was going to give a very general answer, it would be declining profitability of these rural obstetric units or the hospitals in general. Larger hospital networks are buying up smaller hospitals, merging some of the services provided by those hospitals or soaking up the entire hospital itself and closing the entire hospital, moving it to a more urban center, and providing those resources at the more concentrated urban hospital.

 JENA: So, it’s not because the quality of care is declining in those obstetric units over time. And so, they say, “Okay, look, the quality of care is diminished to such a point that we need to close this.”

WHITE: It’s hard to directly measure changes in quality at a hospital level over time. In terms of outcomes, we don’t see these changing trends in the things we really care about — things like infant mortality — that occur right before a closure occurs. We do see a declining number of patients going to these hospitals. So, in the several years before this obstetric unit closure, you do see that women are already starting to move to the nearby hospital. And at some point, there’s a straw that breaks the camel’s back. There’s no profitability in these hospitals anymore, for these specific services, at least, .and the obstetric unit closes.

JENA: What did prior research in this area show before you did your study?

WHITE: If you look at the beginning of our study period, infant mortality rates in 1989 were about equal between rural and urban counties. And that starts to change immediately. And what we see is, there’s this gap in infant mortality rates that emerges, and by the end of the sample period — right in 2019 — infant mortality rates in rural counties are about 15 percent higher than they are in urban counties. And you see these steady trends coinciding with each other. And I think a lot of people have asked in prior research: are these trends related? We thought, “There hasn’t been a study that’s looked at a really long time period and a wide range of different outcomes. So, maybe we can do that.” What we did in the study is fairly straightforward. We took 30 years of data, starting in 1989, going to 2019, and we first identified all the obstetric unit closures we could. We’re looking at counties that go from one or more hospital based obstetric unit to zero. There’s about 3,000 counties in the United States. So, 480 counties lose all obstetric services. That’s a pretty large percent of geographic units in the United States that experienced this shock. Once we identified all those closures, We’re comparing counties that lost all their obstetric services to similar looking counties that didn’t. We look at how far are women traveling after a closure occurs? Where are they going? Are they giving birth in a hospital at all? And then looking at procedures — things like, rates of inductions and C-sections, birth outcomes that might be determined by things like prenatal care — like birth weight and gestational length — and then, things like maternal morbidity, infant morbidity, and then the most extreme outcomes, things like infant mortality. When we look at the effect of these rural obstetric unit closures on a broad range of maternal and infant health outcomes, what we find is there’s not really any effect. And if anything, there’s actually a slightly beneficial effect for mothers,

 JENA: So, your paper obviously speaks to a very relevant policy issue, which is: closure of obstetric units doesn’t seem to be associated with any harm to moms and their infants, and may actually be associated with improvements.

WHITE: Rural obstetric units can provide two types of services: prenatal care services and labor and delivery services. So, when a rural obstetric unit closes, women are losing nearby access to both types of these services. So, when we actually look at gestational length, we do see a slight decline in, how many weeks gestation these mothers had. But what’s interesting is that it turns out that this is really just kind of a shift from gestation at 40, 41, 42 weeks to gestation right around 39 weeks. And what we think is happening, at least, is that providers know that these rural hospitals are closing. They know that these women are having to travel much larger distances in order to give birth in a hospital. And so, they’re scheduling inductions right around 39 weeks.

An interesting addition to that is that there’s this recent study in the New England Journal of Medicine that randomized, induction at 39 weeks, rather than expectant management. Just kind of waiting. And what they found is this induction at 39 weeks looks like it actually might be beneficial rather than waiting to 40, 41, 42 weeks. And the reason is, one, there’s about a 16 percent decline in C-sections, if you, have a scheduled induction at 39 weeks. And that study also finds small improvements in infant health. That study came out in 2018, essentially after the sample period of our study. So, this new evidence on the benefits of induction at 39 weeks wasn’t being factored in by the providers in our study at all. It might just be that accidentally providers were essentially using the best practices, but we didn’t know that those were the best practices at the time.

 JENA: So, you find that there’s fewer prenatal visits. There’s a shortening in gestational age. And that seems like it’s because women are more likely to have inductions performed on them, probably because if a woman’s water breaks, it’s going to be hard to drive two hours to get to the hospital. So, the doctors may want to do the delivery in a more scheduled way. And you also then find that C-section rates fall by a pretty meaningful amount.

WHITE: We actually see about a 1-percentage-point reduction in the probability of C-section. Part of that decline in C-sections is likely coming through an increase in inductions. but much more than that, what we find is that women are traveling to hospitals that have lower risk adjusted C-section rates. So, you’re giving birth in a hospital where the providers in that hospital just tend to perform C-sections less often And this contributes to this ever expanding literature on geographic variation in healthcare. And what we’re showing in this paper is that, well, when women are randomly reallocated to a hospital with a lower C-section rate, yeah, they’re actually much less likely to have a C-section.

JENA: So, there’s like a policy story here, which is quite interesting, but there’s also, something that speaks to our understanding about safety in obstetric care.

WHITE: We don’t really see any changes in infant mortality, infant morbidity, maternal mortality, and most of our measures of maternal morbidity. Although we do find pretty robust evidence of an improvement in one of our measures of maternal morbidity and, the reason we think we see that is of course this phenomenon that women are traveling to higher quality hospitals.

JENA: One thing that your study reminds me of is this study — which we’ve actually talked about on the show — when Hurricane Katrina,, hit,, mortality of people who lived New Orleans actually,, improved in the years following, which is sort of surprising. And it’s because they moved to areas that were different, in which healthcare or healthcare outcomes tended to be better.

WHITE: Yeah. So, this is very closely related to the Hurricane Katrina paper. Both papers are in the spirit of this emerging literature on the causal effects of place in healthcare, right? Where you live matters. Where you get services matters for your outcomes. I think we find something that’s maybe a little less dramatic than that, but people that live in counties that experience an obstetric unit closure, yeah, they are traveling to hospitals that have higher quality services. And their birth outcomes might be slightly better as a result of that.

JENA: It’s almost like the closure that occurred in your area forced your hand. And then, in doing so, allowed you to choose among places that may be better than what you had direct access to, because it was the closest to you. Were you surprised by the findings?

WHITE: What I would’ve expected to find is some negative impacts, at least for some groups, But I wouldn’t say that I expected to see on average slight improvements in health and that we really can’t identify any negative effects in this study. But in terms of thinking about the tradeoff between, driving a long distance and higher quality of care, I think the effects that we find are much more in favor of the higher quality story than I would’ve anticipated.

JENA: Your findings suggest overall either a benefit or no harm. I’m curious if you have a sense of why do we then hear so much uproar about O.B. unit closures in rural areas?

WHITE: So, you see these very steady trends over time in the number of rural obstetric units that close. You see rising gaps between rural and urban areas in, say, infant mortality. And it’s very natural to put those two pictures together and think that those phenomena must be connected. I think we haven’t as a community, had a great understanding of what the effects of these closures are. And I think it’s been really a fair thing to be worried about given the trends that have been occurring over the last 30 years.

JENA: You focused on obstetric care and there is some discretion. There is some time that mom can take to travel to a hospital that’s farther away. What happens when we start talking about things like stroke and heart attacks?

WHITE: I do want to emphasize that, what we’ve done in this paper — this is not generalizable to all health conditions. And there’s certainly going to be health conditions where time really matters, and immediate care is really going to outweigh the effects of kind of getting treated at a higher quality hospital.

JENA: Makes total sense to me. And I say that as a physician.

Rural hospitals play an incredibly important role in the communities where they operate. They’ve done so for decades and they do today. But medicine isn’t the same as it was 30 years ago, and the way we care for patients in rural settings maybe needs to evolve. What Corey and Caitlin’s research shows, though, is that there might be a tension between providing rapid care and providing specialized care.

One way to reconcile these tradeoffs would be to implement a rural health system that’s strong in primary care, and also strong in its ability to treat and triage the most time-sensitive medical conditions. When rural hospitals close, the solution can’t be to leave these regions abandoned, but to think about alternatives.

CARROLL: The Rural Emergency Hospital Program is a new policy from C.M.S. It’s coming online in 2023. So, it’s right around the corner. And the idea is that small rural hospitals can eliminate their inpatient service and deliver emergency care and other outpatient services.

It’ll be interesting to see what happens in rural areas when this program begins early next year, so let’s all stay tuned. In the meantime, I’d like to thank my guests today, Caitlin Carroll and Corey White, and thanks to all of you for listening! Don’t forget to leave us a review wherever you get your podcasts, and tell your friends and family to listen! It really helps us out.

Coming up next week: In medicine, when you analyze data, it can take you to some unexpected places.

BRENNER: It started to tell a story that was horrible, which is that a small sliver of patients were going back over and over for all kinds of reasons.

Some of these patients were showing up in Dr. Jeffrey Brenner’s office in Camden, NJ, and he noticed something alarming about them.

BRENNER: They were going back and forth all the time to the hospital. But they weren’t getting care. They were getting treatment, but they weren’t getting care.

These complex patients, also called “super-utilizers,” have lots of needs, and it costs a lot of money to take care of them. They were also falling through the cracks. Jeff and his team developed an innovative approach to help them. But did improving their care also decrease costs?

DOYLE: They were getting a lot of praise for their program, and they honestly wanted to rigorously determine whether it was saving money on its own.

That’s all coming up next week on Freakonomics, M.D. Thanks again for listening.

*          *          *

Freakonomics, M.D. is part of the Freakonomics Radio Network, which also includes Freakonomics Radio, No Stupid Questions, and People I (Mostly) Admire. All our shows are produced by Stitcher and Renbud Radio. You can find us on Twitter and Instagram at @drbapupod. This episode was produced by Julie Kanfer and mixed by Eleanor Osborne, with help from Jasmin Klinger. We also had help this week from Jacob Clemente. Our staff also includes Neal Carruth, Gabriel Roth, Greg Rippin, Rebecca Lee Douglas, Morgan Levey, Zack Lapinski, Ryan Kelley, Jeremy Johnston, Emma Tyrrell, Lyric Bowditch, Alina Kulman, and Stephen Dubner. Original music composed by Luis Guerra. If you like this show, or any other show in the Freakonomics Radio Network, please recommend it to your family and friends. That’s the best way to support the podcasts you love. As always, thanks for listening.

*          *          *

CARROLL: So, then we can think about what was the value of those saved lives — the saved life years?

JENA: Caitlin. You’re — you’re — you’re so cruel. How could — how could you — how could you put a value on a life?

CARROLL: it is very problematic to think about the value of a life year. So, I want that to be clear.

JENA: This is not anything that’s personal — just blame it on the field of economics.