Straining the System: Heat and Health-Care Outcomes, with Sandra Aguilar-Gomez

Notable Quotes

• Extreme heat taxes human health: “A temperature shock today, a hot day today, makes people seek more health care. So, it’s a positive correlation with morbidity. We see that there’s an increase in emergency-department visits and a linear increase in hospital visits as temperature rises.” (14:59)
• Heat crowds hospitals and crowds out patients: “As the temperature rises and hospitals get saturated, they’re sending sicker people home. What we find is that, unfortunately, the deaths outside the hospital system … and the percentage of the people who are dying outside the hospital system are increasing on hot days.” (17:46)
• Preparation and communication can help hospitals manage hot days: “Thinking about management protocols and emergency preparedness [on hot days] … Hospitals should be able to be prepared and have some margin to operate on the amount of staffing that they have, the sizing of their hospitals, and local coordination between hospitals.” (22:21)

Top of the Stack

• Sandra Aguilar-Gomez homepage
• Babbage podcast from the Economist
• How to Save a Planet podcast by Gimlet Media

The Full Transcript

Kristin Hayes: Hello, and welcome to Resources Radio, a weekly podcast from Resources for the Future. I'm your host, Kristin Hayes. We're recording today's episode from the annual conference of the Association of Environmental and Resource Economists (AERE). AERE is the main professional membership organization for environmental and resource economists and holds its flagship conference every year around this time. The location varies, but this year it happens to be in Washington, DC, which is Resources for the Future's backyard.

There are many strong connections between AERE and RFF, and I wanted to take just a moment to acknowledge those. These connections date back to the founding of AERE in the mid 1970s when some prominent environmental economists, including some at RFF, helped to create the organization. These included what we sometimes refer to as our elder statesmen, who were really involved in some of the most important early work in the field. These are people like John Krutilla, Allen Kneese, Emery Castle, Kerry Smith, and lots of others. They were all affiliated with RFF in one way or another.

RFF also provided a grant to help AERE get off the ground and really get started, and RFF’s support staff helped with the official paperwork. All of this was getting AERE to be the 501(c)(3) nonprofit organization that it is today. Over the years, many RFF researchers—as well as our key staff—have held leadership positions on the board and helped carry out the day-to-day management of AERE. Past AERE presidents include RFF’s own, John Krutilla—I think he was actually the first one—and current RFF Senior Fellows Alan Krupnick and Karen Palmer.

This year, members of the AERE Board reached out to us here at Resources Radio and asked if we might be willing to record episodes at the conference. Because of all this history and all of these connections, we very eagerly agreed.

Today, I'm speaking live from the conference with Sandra Aguilar-Gomez, who is an assistant professor of economics at the Universidad de los Andes in Bogotá, Columbia. (This is definitely testing my Spanish pronunciation!) Sandra received her PhD in sustainable development from Columbia University in 2021, and her research focuses broadly on environmental, gender, health, and development economics. She's particularly interested in exploring the intersections between them.

Today, we're going to be discussing a study that she's presenting here at the 2024 AERE conference, called “Killer Congestion: Temperature, Healthcare Utilization, and Patient Outcomes.” This is work that's coauthored with Joshua Graff Zivin at the University of California San Diego and Matthew Neidell at Columbia University. It uses a large health-care data set from Mexico—Sandra's home country—to parse out how temperature-induced hospital crowding influences how patients fare in terms of their trajectories and outcomes. It's a really important glimpse into how a hotter future may affect human well-being. Stay with us.

Hi Sandra. Welcome to Resources Radio. It's very nice to be sitting here with you in person.

Sandra Aguilar-Gomez: That's great. I have to say, your Spanish so far is very good.

Kristin Hayes: Well, thank you very much. That's great. I want you to know, I'm actually going to Bogotá in two weeks, so it's important that my Spanish pronunciation is up to par.

I would love to ask you our wonderful opening question. We like to ask our guests about their background. I'm going to ask you a big question here. I noticed in your bio—and as I mentioned at the outset—your interest in economics crosses environmental and gender health. It's really a range of subdisciplines of economics. I'm wondering how you came to want to bring those all together.

Sandra Aguilar-Gomez: That is a complex question, but thank you for asking it. After doing my economics undergrad in Mexico City, I went on to do a sustainable development PhD.

By the way, there are a lot of fellow alumni here. This is one of our favorite reunion venues. So, hello everyone!

I have always been interested in environmental degradation in developing countries, which, of course, carries very costly health consequences. But also, institutions, in this context, are weaker, and governments have less tools to protect the population from environmental degradation. I'm currently studying, in further papers, policies that governments implement, like air-quality warnings, our local disaster-relief policy. In general, the toolbox that governments have in this context is different from the United States, right? The Latin American versions of the Toxic Release Inventory are younger, messier, and works in progress.

In general, for all of these reasons, the findings from the studies that look at the consequences of environmental degradation and the distribution of such damages might not extrapolate to developing countries. So, the tools of a developing economy are fundamental to thinking about these questions. For example, I have research showing that environmental justice findings from the United States do not extrapolate to a country like Colombia, where skin tone, not a specific racial category, is a better predictor. Similarly, for climate inequality, the gender-climate nexus is a very different thing in developing countries. Overall, when you talk about all of these topics or all of these subfields … To me, they're part of this tool kit that is necessary to tackle this broader question of consequences of environmental degradation in developing countries that I just described. I don't know if that makes sense.

Kristin Hayes: That makes a ton of sense and is really interesting. I think it plays really well into some of what we're going to be talking about today, because one of my favorite parts of the paper was how it looks at differences in studies from high-income countries versus middle-income countries. So, that was a great explanation. I really appreciate it.

Let's turn to the topic of our conversation today. As I noted, the paper has a working title. I think it's a working title.

Sandra Aguilar-Gomez: It is a working title.

Kristin Hayes: Okay.

Sandra Aguilar-Gomez: There is debate.

Kristin Hayes: The working title of the moment is, “Killer Congestion: Temperature, Healthcare Utilization, and Patient Outcomes.” Early on in the paper, you note that there is a fairly extensive literature already on health impacts of hotter temperatures, but there's clearly some new questions that you were looking to answer. Tell us what you're hoping to answer in addition to the literature that's out there.

Sandra Aguilar-Gomez: Well, like you said, we know from a lot of studies that body temperature should only vary within a narrow range. Anything beyond that range is really costly for the body in terms of the effort needed to cool off or warm up. All of these things encompass a lot of health consequences for humans. There is a massive literature that many people in this conference have contributed to documenting these health consequences of extreme events. However, our question was, How can this affect you, not directly through the consequences that your body is facing, but through the strains that it is causing to the health-care system— particularly to a health-care system like Mexico’s that is already at capacity, as in many other contexts around the world? That's what we are looking at: the spillovers of this increased demand for health care, because morbidity translates into that. How does the system deal with this, and what are the consequences of that?

Kristin Hayes: Yeah, great. Taking it from that individual outcome to a system-wide analysis of outcomes.

Let's talk a little bit more about the differentiation between this research that's based in or on high-income countries, compared with a place like Mexico—which I'm guessing we are classifying as a middle-income country for this purpose.

Sandra Aguilar-Gomez: Yes.

Kristin Hayes: Okay, great. Why does that geographic difference matter so much?

Sandra Aguilar-Gomez: It's not only to create knowledge on a specific context that has been understudied. We think that this really contributes to a broader understanding on the differential impacts of climate change across the globe. Middle-income countries have, on average, different baseline temperatures than high-income countries. For example, the average temperature in Stockholm in June is approximately 22°C. In Lagos, it's 34°C, right? So, one degree of warming from 22°C is very different from one of degree warming from 34°C, especially when we're talking about these physiological limits that I discussed earlier. The body really has to operate within a certain window. The second part that also links to this development question is that these countries also have different abilities to respond. So, we will expect different indirect impacts of climate change in this context.

Going back to Sweden and Nigeria—Sweden has 2.1 hospital beds per 100 inhabitants, and Nigeria has 0.5. You can imagine how these health-care systems are responding differently to or have different constraints when dealing with these climate impacts. Most of what we know about climate damages to health has been based on high-income countries, which house 16 percent of the world’s population. Economists are always talking about external validity. So, I always argue very strongly for the external validity of a country, like Mexico, that is a middle-income country, where 70 percent of the world lives. We could say that our results have implications for most of the world; how most of the world would actually deal with climate change and increased temperatures.

Kristin Hayes: Can you say just a little bit more about Mexico's health-care system in particular? I'm wondering if there's any particular context about that system. You mentioned some very interesting and important stage-setting statistics around numbers of hospital beds in various countries. What does Mexico's health-care system look like in terms of capacity? What is some other foundational information you might want our listeners to know before we talk about the study itself?

Sandra Aguilar-Gomez: Most of the population goes to the public health-care system, which is what we are studying. Less than 2 percent receive their health care privately. And the public health-care system is fragmented. We're looking at the largest subsystem; it's called the noncontributory system, meaning that people who can't pay, who are informal, and who don't have social security—in a country with 50 percent of its population considered informal, this is actually the type of health care that the average person gets. This is—we have in our data—more than 800 hospitals serving 70 percent of the population and distributed throughout the entire country that take basically anyone who doesn't have social security access.

Kristin Hayes: You mentioned the magic word, “data.” There's always a question where I like to ask guests about the types of data they use. Often, there are interesting combinations of data that come together. I'm wondering if you can tell us a little bit more about the hospital data sets, the levels of data, and the types of data that you brought together for this research.

Sandra Aguilar-Gomez: Sure. This is probably where the nerds start getting excited, right?

Kristin Hayes: That's right. We love the nerds.

Sandra Aguilar-Gomez: In this public health-care system, we have detailed records on when patients were admitted, every single diagnosis that they got during their hospital stay, some sociodemographic characteristics, all of the procedures that were practiced on them, and how they were discharged. This will be important later. We have these data for emergency departments, inpatient units, and outpatient units across the entire country. We also have vital statistics and the universe of death certificates in the country. We can see the date, location, and setting—inside, in a hospital, at home, etc—of all the deaths that occurred in seven, eight years in the country. We merged that data with gridded weather data to get very geographically granular measures of daily maximum temperature.

Kristin Hayes: Sounds like a lot of data.

Sandra Aguilar-Gomez: That was a lot. This is very heavy. The other users of the server don't like me!

Kristin Hayes: The first two data sets that you mentioned, the ones that were information about hospitalizations and release rates—are those housed separately from the death-certificate data? Did you have to merge those two together, as well, or does that all come from the hospital infrastructure?

Sandra Aguilar-Gomez: No, it comes from a different subsystem, and right now, we're working on everything: on one hand, at the hospital level, and on the other hand, for the death certificates at the county level. So, no merging.

Kristin Hayes: The reason I love asking the detailed data questions is because I feel like sometimes—as a person who is not as involved in the research enterprise—I'm perpetually amazed at how much of the research enterprise is about just getting the data right and figuring out how to match data sets and bring things together in creative ways. So, extra shout-out to the creative data combinations.

I've taken us long enough without letting you talk about the key findings. Let's talk about those findings. I'm going to leave this question pretty broad and invite you to talk us through what you and your colleagues found across the various dimensions that you were looking at. Let me turn it over to you to tell us what you found.

Sandra Aguilar-Gomez: Great, thank you. First of all, we confirmed what has been shown in other contexts. A temperature shock today, a hot day today, makes people seek more health care. So, it's a positive correlation with morbidity. We see that there’s an increase in emergency-department visits and a linear increase in hospital visits as the temperature rises. More people are arriving at the emergency department.

Then, there is the question of the functions of the emergency department, one of which is triage. Some of these patients are going to be admitted into the hospital for further care, and some of these patients are going to be sent home. So, what we find is that, as temperature rises, the emergency department is forced to send more people home and admit less people to the hospital. We see that the number of people who are admitted, because they are swarming the emergency department and the hospital, is increasing, but your individual probability as a patient—conditional on your characteristics and so on—of being admitted on that day decreases as the temperature increases. So, what happens to you? The hospital is full, and instead of letting you in, they send you home.

We know from this literature that heat makes people more uncomfortable. In general, it makes them sicker, but it also changes behavior. More people are likely to seek health care as temperature rises. There's a behavioral response to heat. These people that are being sent home could just be a selection story. It could be that less-sick people are showing up. They're just more uncomfortable and show up in the hospital. “I feel bad,” but nothing is really wrong with them. Then, they send them home, and this could be a sensible way to deal with increased demand.

We show that this is absolutely not the case, and that people are showing up sicker to seek health care. To do this, we construct a severity measure. Because we have the universe of hospital information for almost a decade, we see enough deaths to get good predictors of mortality. With those predictors of mortality, like your diagnosis, age, sex, etc, we can talk about how severe you are when you show up at the hospital. What we're seeing is that this severity, measured as predicted mortality of people, is increasing among people who are sent home.

So, as the temperature rises and hospitals get saturated, they're sending sicker people home. What we find is that, unfortunately, the deaths outside the hospital system—and here's where the death-certificate data enters—and the percentage of the people who are dying outside the hospital system are increasing on hot days. So, more people are being sent home, and it seems, from our data, our findings are consistent with these people dying outside of the system instead of receiving care.

Kristin Hayes: Let me make sure I've got this right. The reality is that this overcapacity is leading to a lack of ability to care, even for patients who are sicker. So—as you mentioned with the selection-story piece—this idea, that maybe it's just uncomfortable people who aren't really worse off, is not the case. In fact, hospitals are being forced to send sicker people home, and the consequences of that are a greater number of deaths outside of the home. Did I summarize that right?

Sandra Aguilar-Gomez: Outside of the hospital, and also inside the hospital.

Kristin Hayes: Yeah, these are already pretty sobering findings.

Sandra Aguilar-Gomez: Yeah, absolutely. I hope that we get some hope when we discuss the policy implications, but for now, I think the situation is pretty dark. Overall, if Mexican people listen to this, they know that the situation of the health-care system is actually pretty dire. It's already operating very close to capacity on a normal day. These shocks are really a game changer.

For inside the hospital, we use a measure of health-care quality that is very standard in this literature known as excess mortality. We have all your diagnosis information, and like I said, we can predict the probability of death of a patient. But then, as temperature increases, excess mortality increases. This is consistent with many things, one of which is that the crowding is what causes this excess mortality. The hospitals have a lot of people to deal with, so the quality of care that each patient receives decreases as heat increases. Then, we see more deaths—inside the hospital system, as well—than what you could expect.

Kristin Hayes: That's what that expected mortality baseline is, right? It's a projection of the level of mortality you would've expected at one level, and then you're able to look at the increment on top of that on a particularly hot day.

Sandra Aguilar-Gomez: Yeah. We see the difference between expected mortality—based on diagnosis—and actual mortality, and that gap that is usually a measure of quality of care. In the United Kingdom and so on, they actually monitor hospitals with this, and we see this gap opening as temperature increases.

Kristin Hayes: All right, I'm with you. I think I need something actionable—maybe not hopeful; I won't put too much pressure on you—but at least actionable. We've both acknowledged these are pretty sobering findings, so I'd like to give you a chance to summarize, for decisionmakers in the health-care industry, whether that's in Mexico or, as you pointed out, any of these other number of countries that might be in the same income range and struggling with similar issues, what you want them to take away from this work. Either as something that they can or should change, or at least, for awareness, that they need to be cognizant of the reality that in a hotter world, this might be more stress on their systems.

Sandra Aguilar-Gomez: Absolutely. Both things, I would say. First of all, the extrapolation of temperature increases due to climate change indicates that this problem is likely to be exacerbated in the future. So, we have to do some planning here. First thing is strategic health-care infrastructure planning, like thinking about rising demand in response to climate change, where demand is going to rise, and how we can prevent these spillovers. To prevent these spillovers, I think our findings really underscore the relevance of capacity expansion. The system seems to be operating on the border most of the time. There's other things that we can do on the margin.

Thinking about management protocols and emergency preparedness … Right now, for example, even in other countries, if there is a heat wave or another natural disaster like an earthquake, there are alerts inside the health-care system that say, “It's not a regular day today; it's a special day.” So, hospitals should be able to be prepared and have some margin to operate on the amount of staffing that they have, the sizing of their hospitals, and local coordination between hospitals. If this clinic is full, how does it communicate with the nearby clinics in an efficient way? Sometimes, they don't even do that, so patients have to wander around trying to seek care. Instead of that, they should be more efficiently directed to a specific place.

Our research doesn't directly speak to getting information to the population, but other works that I have done do. On hot days, Mexico and many countries should definitely have more advanced alert systems to tell the population, “It's hot out there.”

Kristin Hayes: Right, get ready.

Sandra Aguilar-Gomez: Get ready, stay inside, and hydrate. These are very simple things, but preventative measures that people can do could also ease the strain on the health-care system. A good protocol would have both things. The prevention arm—warning the population about this. And the treatment arm—preparing the health-care system for increased demand during heat waves and other environmental extremes.

Kristin Hayes: I'm realizing, right in this moment, that Mexico, at least Mexico City, might actually be in the middle of a significant heat wave right now, if I'm not mistaken.

Sandra Aguilar-Gomez: Yeah.

Kristin Hayes: Is that true?

Sandra Aguilar-Gomez: Yeah, they are.

Kristin Hayes: Oh, gosh.

Sandra Aguilar-Gomez: I hope they read the paper.

Kristin Hayes: I know. Well, let's get this podcast out there quickly then for the good recommendations.

Sandra Aguilar-Gomez: Right—let's go viral, please!

Kristin Hayes: I do think this is a particularly timely one as we head into summer. I'm sorry that it's coming right on the heels of an actual event in Mexico City that is probably testing that system, but it sounds like this is really valuable information for hospital systems across the world—particularly, in these countries that might be operating on that capacity margin right now—to understand what the demands are likely to look like moving forward and the realities that they're facing. I really appreciate your taking the time to share this with us.

Sandra Aguilar-Gomez: Thank you for inviting me.

Kristin Hayes: Especially in real time, here at the conference.

Let me close with our regular question, which we call Top of the Stack. This is your chance to recommend to our listeners any other good content. It can be on this topic, related to what we're talking about or not—anything that you'd like to recommend to our listeners. What's on the top of your stack?

Sandra Aguilar-Gomez: Well, I am an assistant professor on the tenure track. Right now, on the top of my stack are my ungraded essays from my semester.

Kristin Hayes: Fair enough.

Sandra Aguilar-Gomez: I’m reading so much, but while I cook or walk, I listen to this nerdy podcast, Babbage from the Economist. I don't know how often you get that response. I mostly do that and How to Save the Planet, but they discontinued that fantastic podcast.

Kristin Hayes: That's okay. People can replace it with this one!

Thank you again. Enjoy the rest of your conference, and I look forward to hearing more about what you're up to.

Sandra Aguilar-Gomez: Thank you. Thank you for the opportunity.

Kristin Hayes: You've been listening to Resources Radio, a podcast from Resources for the Future, or RFF. If you have a minute, we'd really appreciate you leaving us a rating or a comment on your podcast platform of choice. Also, feel free to send us your suggestions for future episodes.

This podcast is made possible with the generous financial support of our listeners. You can help us continue producing these kinds of discussions on the topics that you care about by making a donation to Resources for the Future online at rff.org/donate.

RFF is an independent, nonprofit research institution in Washington, DC. Our mission is to improve environmental, energy, and natural resource decisions through impartial economic research and policy engagement. The views expressed on this podcast are solely those of the podcast guests and may differ from those of RFF experts, its officers, or its directors. RFF does not take positions on specific legislative proposals.

Resources Radio is produced by Elizabeth Wason, with music by Daniel Raimi. Join us next week for another episode.