Search the Site

Episode Transcript

THOMASSON: When I went to graduate school it was the early nineties when then-President Clinton unleashed his healthcare plan, and it was over a thousand pages. It was so huge.

That’s the economist Melissa Thomasson.

THOMASSON: I remember being struck with the thought of, well, why do we need this? And, how did we get into this mess in the first place?

Understanding how we got into this mess, or any mess, is part of Melissa’s job as a professor at Miami University, in Ohio. She’s an economic historian, which means she spends a lot of time thinking about the events that set the stage for our current economic systems — including healthcare. In 2020, healthcare expenditures represented around 20 percent of Gross Domestic Product, or G.D.P.

And yet, back in the early 1990s when Melissa Thomasson was casting about for a dissertation topic, she realized something.

THOMASSON: It turns out, no economist had really looked at quantitatively, understanding how we did get into this mess in the first place. There’s been some terrific history books written on it. Some sociologists have looked at it, but no one had really actually tried to measure the factors that led us to our current healthcare system.

So, that’s what Melissa has spent the last two decades doing. She’s written papers about school closures during the 1916 polio epidemic in the U.S.; and about the health effects of living through the Great Depression.

A lot of her work focuses on shocks like those. Big health care bangs, and their lingering impacts. Recently, she and some colleagues wrote a paper about one especially impactful bang: the 1918 flu pandemic, also known as The Great Influenza.

THOMASSON: It really shows us how a significant shock to the healthcare system can play out even decades later

How does the 1918 flu pandemic continue to play out decades, even a century later? And what can we learn from it, as we emerge from another giant shock?

From the Freakonomics Radio Network, this is Freakonomics, M.D. I’m Bapu Jena. Today on the show: How can medicine’s past help us understand its present, and maybe see its future?

THOMASSON: Looking back, it’s really natural to say, well, of course communities built their own hospitals. But at the time, it was anything but clear.

And: we all know what they say about hindsight. But is it possible we can sharpen our foresight to better predict when crises will occur?

MARTIN: Our healthcare system is much more oriented towards disease care than healthcare.

*      *      *

THOMASSON: Right around the time of the pandemic, middle- and upper-class Americans were starting to realize that medical care might be a good option and going to hospitals might be useful.

Healthcare and medicine looked a whole lot different in 1918, when the Great Influenza hit.

THOMASSON: Let’s think about 1900, as sort of a point in time where you start seeing the modern hospital begin to develop. Before 1900, there were basically two kinds of patients who would go to hospitals. One kind of patient was what some historians have called the working poor. They were single people who lived in a boarding house who became sick and who had no one to care for them. They couldn’t hire a nurse and no one at home could nurse them. And a lot of charities and religious organizations opened hospitals to provide them with warmth, nursing care and shelter. And that started to happen maybe around the 1870s or so.

The other kinds of patients were patients with more chronic illnesses So those might have been alcoholics or unwed mothers, people who were mentally ill, whose families had just sort of kicked them to the curb. And cities operated alms houses for people without homes who were sick to be sheltered in. So these were called hospitals, but they weren’t really hospitals in the way that we think about hospitals today.

At the turn of the 20th century, if someone was ill, there wasn’t much a doctor or a hospital could do for them.

THOMASSON: Let’s say you get sick in 1905. You get the flu, you get pneumonia. Chances are you’ll have a physician visit you in your home. They’ll tell you you have the flu or you have pneumonia and give you fluids, give you rest, keep you warm, open the window, get some fresh air, and that’s about the extent of what they can do.

When the flu pandemic began in 1918, World War I was still unfolding. Hospitals were understaffed because doctors and nurses were overseas helping soldiers.

THOMASSON: Whole families were sick. Nobody was around to care for anybody. So initially, right in the thick of the pandemic, everybody just went to hospitals. It was standing room only and overflow space. As the pandemic receded, a lot of people in those communities that were harder hit started to say, “hey, we have to really do something about it.”

As the Great Influenza stormed through the country and around the world, it encountered a number of medical and scientific innovations. Germ theory, which says that specific microscopic organisms cause specific diseases, was developed in the 1860s; by around 1920 it was largely accepted. In 1895, physicians got the x-ray, which allowed them to peek inside the living body for the first time. And in 1921, insulin was discovered, making it possible to treat diabetes.

All of these new medical discoveries needed a place to happen. At the same time, a lot of people who were sick with the flu needed a place to go.

THOMASSON: So we have new technologies, anesthesia for childbirth, new understandings of disease processes. And hospitals were a natural place to locate a lot of these technologies. Things like X-ray were really bulky, a better understanding of germ theory led operating rooms to develop. Before that, physicians would routinely do surgery on kitchen tables. Hospitals became places where people would go to get medical care. And as a result of the new technology, they became more expensive.

Hospitals changed, and became an industry, one that now accounts for a third of all health care spending in the U.S., or about 6 percent of G.D.P. Not long ago, Melissa and some colleagues were thinking about the COVID-19 pandemic, and how it made commonplace an element of health care most people hadn’t interacted with before: telehealth. It reminded them of how the role of hospitals changed during the 1918 flu pandemic. People didn’t use hospitals much, until they did. So, what factors contributed to where and why hospitals proliferated around the country? And what downstream effects did this have — even today — on public health?

THOMASSON: We went way back in time and gathered a list of every hospital in the U.S. back to around 1910. And it was a little tricky. We found it in these volumes called the American Medical Directory. It, was, like, an eight-inch thick volume that listed all the hospitals and every physician in the country. And you could find the hospitals, but then it had, homes for people who were described as mentally defective or crippled children. And so trying to actually think about, okay, which of these institutions are hospitals in the sense of what we think of as hospitals?

We did a lot, of statistical techniques to sort it into what we thought were hospitals. And then in the 1920s, the American Medical Association started doing an annual survey of hospitals. So after 1925, we were able to actually categorize every hospital that was registered with the A.M.A. in the country, all of their beds. And then we got a group of cities, where we actually had really good city-level data on influenza deaths, because that’s hard to come by. And then, we look at how hospital capacity in those cities changed based on how many people died, of the flu in 1918 in those cities.

JENA: And what did you find?

THOMASSON: We found that cities that had higher mortality rates from the flu expanded hospital capacity to a greater extent than cities that didn’t. And these tended to be higher income cities. We were finding big effects in midsize to smaller cities that maybe didn’t have the hospital infrastructure like a New York or like a Chicago, and that those hospitals really, after the pandemic, caught up to the Chicagos and the New Yorks of the world. But again, that effect was much stronger in cities where the population had been more negatively affected by the flu pandemic of 1918.

JENA: So one of the challenges when we’re taking a finding like this and saying, all right, did the 1918, influenza pandemic cause more hospital capacity to be developed in cities than otherwise would’ve been developed. Is this really a causal finding? What did the trends look like in hospital capacity in those areas where influenza hit hard versus did not hit hard?

THOMASSON: The idea of correlation versus causation is really crucial. But we do look at what we call pre-trends in those cities, and those were cities that looked very similar to each other prior to the pandemic hitting. The cities that got hit harder with influenza, didn’t seem to be associated with being big or small. It was kind of just randomly dispersed. And so those cities looked similar to each other prior to the pandemic. And then there is a departure with the number of hospitals they started to build.

JENA: The growth in hospital capacity that you observed, — where did it come from? Was this public government saying, look, we recognize this is a huge public health issue, we need to build hospital capacity, or were these really efforts where private hospitals said, look, we recognize there’s a market demand for better healthcare provided in the hospital. We need to meet it. Which of those two happened?

THOMASSON: Before looking at this, you might think, well, maybe cities were really worried about having another big public health shock. Maybe this was an opportunity for cities and counties to tax themselves and to build bigger, nicer city or community hospitals. But what we find is the opposite. The growth in the hospital industry occurred solely among non-profit, private hospitals, these tended to be very philanthropic, community organized, hospitals that really were not owned and operated by cities.

JENA: These are not public hospitals.

THOMASSON: Right. These are like today’s community’s hospitals, right? That a couple, wealthy donors start, they do fund drives, bake sales, lots of things like that. And these hospitals are nonprofit hospitals.

JENA: These days, you gotta do a lot of baking. I gotta tell you, Melissa —

THOMASSON: It would be a lot of baking and a lot of wealthy donors.

JENA: So you have this expansion in hospital capacity that occurs , in your view, really by chance, depending on which communities were hardest hit by the 1918 pandemic. What is the byproduct of that? Do we then observe better quality care? Do we observe just more hospital use and greater spending on hospital care?

THOMASSON: In our data we can’t actually observe quality of care or spending per se. But what we can say is that in those cities that were harder hit, they had greater levels of hospital capacity. and it persisted till 1960. So in 1960, those cities that were harder hit by the flu had more beds per person than cities that were less harder hit by the flu. If you think that more beds leads to more spending, which economists do, then you could say yes. it did lead to more spending.

JENA: The big question of course is, was that spending helpful? Like, if you have more beds per capita, and that allows you to provide more high quality care per capita, then you would expect things to get better from a health outcomes perspective, that could outweigh the spending. But you don’t have any sort of insight as to whether care itself or the outcomes improved as a result.

THOMASSON: No, unfortunately, we don’t measure that with our data.

JENA: How long of a period did you study after the initial shock, this 1918 shock?

THOMASSON: We see those effects on hospital construction until the early 1960s, and what we suspect is that Medicare, which then was in a whole other shock, really typically evened things out. And that competitive forces and more resources from Medicare led to more of an equalization across cities.

JENA: And what impact did the differential growth in hospital capacity have on disparities, and access to hospital care, like racial or socioeconomic disparities?

THOMASSON: We can’t answer that directly with our data, but, it’s quite possible that it did. There are some other papers in the literature that suggest in some communities, when these hospitals become available there are lots of whites taking advantage of them and fewer people of color able to take advantage of them. And so we could see inequities and disparities developed because of that.

JENA: So if you were to describe in a couple sentences, what is the contribution of this particular study to the economics or health economics literature or public health literature, what would it be?

THOMASSON: Oh boy. It really shows us how a significant shock to the healthcare system can play out even decades later. And I think this paper shows us, that these shocks, even when they’re over, have lasting effects.

After the break, what could the lasting effects be of a more recent shock to the healthcare system?

MARTIN: We’ve seen a lot more people applying and matriculating into healthcare programs, more people applying to medical school, nursing school.

And, if you could predict the future of your own personal health — would you? I’m Bapu Jena, and this is Freakonomics, M.D.

*      *      *

MARTIN: I’m Greg Martin. I’m a pulmonary and critical care physician at Emory University in Atlanta, where I work predominantly at Grady Memorial Hospital. Most of my time is spent in the I.C.U. where I also conduct research and help develop new ways to take care of patients.

Developing new ways to care for patients can take years, if not longer. Unless, of course, there’s a pandemic. Then things move fast.

MARTIN: COVID came to all of us very, very quickly and the I.C.U., started off with a lot of uncertainty about how to treat these patients. One of the things that took us some time was to understand that many of the things that are the foundation of intensive care — the way we use ventilators, trying to protect the lung supporting the heart, the kidneys, the other organs — often end up being much more similar in COVID-19 than they are different. COVID occurred in surges and clusters. In the I.C.U., we had to find ways to expand the resources that we had. so that all the patients are getting the care that they need.

Greg and his colleagues relied on what they already knew, but also had to innovate quickly, with the world watching. In the earliest phases, there was rampant uncertainty surrounding testing, supplies, and even space. Where would they put all the sick people? They were overloaded with patients and also, with something else.

MARTIN: The uncertainty around care was exacerbated by a tsunami of information. And some of that was fostered and facilitated by social media. But even the biomedical literature was overwhelmed with reports and information that we were trying to share and parse through — which was the highest quality? Which was gonna apply to our patients? And how do we implement that? So there’s the lack of information on one hand, there’s a tsunami of information on the other hand, and trying to put those two together became very challenging.

It’s easy to look back at the ways medicine was, and wasn’t, prepared for COVID. It’s harder to say what comes next, as a result of it. In 2021, Greg and a group of critical care physicians from around the world published a paper that tried to do just that.

MARTIN: Much of the advance, particularly in critical care medicine, was ways to ensure the resources are available for, each individual patient as they come in. I think of that as the three Ps, the people, the places, and the paraphernalia. So the people are the specialized people that are providing the care to the patient, the nurses, the physicians, the other physician-provider groups but also, you have the pharmacists, the rehab specialists, the respiratory therapists, all working together. And that became a real challenge because, the next piece that we needed to expand was the places, meaning we needed more I.C.U. rooms or more I.C.U. beds to care for these patients.

And then the last part is the paraphernalia, which you can imagine we needed ventilators and dialysis machines and within an individual hospital, you can expand many of those things. You can open new spaces for critical care patients. You can try and expand the capability of the individuals working in the I.C.U., for instance. But the paraphernalia piece became particularly challenging because most of those are very limited. There’s only so many dialysis machines. There’s only so many ventilators. We learned a lot about the medical side of patient care, but really that earliest phase was understanding the healthcare side. What do we really need to do to care for those patients?

JENA: And what were the major medical innovations that emerged during the pandemic that were heavily used in the I.C.U.?

MARTIN: Many interventions came along and one of the earliest was the understanding of corticosteroids or immunosuppressant therapy to really help blunt some of the inflammatory response that was inducing the injury and causing the illness. So using dexamethasone, one of the most common interventions that we still use today, was shown in several trials to be effective, and particularly is effective in people who are more severely ill, and that’s one of the core things that we learned very early on. We’ve seen other interventions, things that are more targeted at very specific aspects of the immune system. So if you’re trying to blunt a specific cytokine or a specific part of the immune response, we can do that effectively too. And certainly the other part that we’ve seen is the antiviral therapy, which is more broadly effective and we tend to use that really in the earliest phases of illness. Once people become critically ill, it’s less likely that antiviral therapy is gonna work.

JENA: There’s, a part of the pandemic that we haven’t talked about yet, and that was how to deal with the supply constraint. So for example, if you had to choose between which patients would receive ventilator support versus not, how much of that did you personally witness in your own I.C.U.? And, did we have a good medical-ethical framework for thinking about who to prioritize treatments in and who not to?

MARTIN: We had a conceptual framework for how to deal with these kinds of surges and how do you make triage decisions, which patients are going to get which resources, and that’s called crisis standards of care. And in some cases, that has been developed around disaster medicine. But when you get to a pandemic and it’s no longer the urgency of an immediate disaster but you’re having people come in over hours, days, and even weeks, that pace becomes more challenging it’s not just allocating resources, it’s also potentially removing resources. Do you take a ventilator away from a patient that’s currently in the hospital, in the I.C.U. because they’re less likely to survive and you now give that to the person who’s more likely to survive?

JENA: What has changed in hospital care and critical care medicine that will change the way healthcare is delivered in the future?

MARTIN: Certainly one thing that’s changed due to COVID is we realize that there’s a need to be able to expand capacity, and that we now understand better how to do that. One of the things we also learned is that you need to be able to maintain that capability over time. And then much of the communication began to occur outside of individual hospitals. So in the past, when a patient is, let’s say sick in a community hospital in a rural area, they may call a referral center and say, I need to transfer my patient. And there were referral networks that had been built up. But when an individual hospital was overwhelmed or too full, it made it much more difficult to transfer and move patients. So having a critical care coordination center, which they’ve tried to build in some places, would be extraordinarily helpful for long-term planning and being able to respond to future issues like this.

JENA: I’m particularly interested in this point because prior to the pandemic, if you had a person who came in, to their local hospital with an acute respiratory condition where the local doctors in the hospital were thinking, we need to transfer this patient to a referral center, is it possible that in the future they would be less likely to do that because they are more equipped to handle that type of severity in the hospital? And if so, do you think that that will lead to better outcomes on average or potentially worse outcomes on average? Because you could imagine , both being true. On one hand, if you feel like you’re better able to deal with the types of patients that you’re getting, you might keep them in the hospital, but that might not be a good thing. It might be better to transfer those patients to highly specialized centers. But it could also be a good thing if you get early treatments on board, then decide whether or not to keep or transfer the patient. It might be that outcomes improve.

MARTIN: Most people desire to get their care somewhere proximate to family and friends and their support environment. What we’ve seen is that there is an expanded capacity for caring for patients now in local hospitals. I worry that that expanded capacity will wane over time as education and training and resources get scaled back or they fall back to a traditional baseline. What I hope is that we’ll continue on a trajectory with technology filling part of that gap. One example would be TeleCritical Care, which allows you to visualize a patient, see the actual data that’s being collected even see the patient on a camera where you can see , their ventilator, the waveforms and the information on their bedside monitor, And what that really facilitates is a decision for whether that patient needs to transfer or can they safely be managed at their local hospital where they’re closer to friends and family. But on the other hand, I would not expect the highest levels of trauma care, stroke care, cardiac arrest care, to be available in every hospital of every size throughout the world.

JENA: We can’t really talk about the future of, how the healthcare system, evolves without thinking about, the actual workforce. The pandemic has had a lot of effects on people, burn out in particular. But there’s another effect that I think we talk a lot less about, and that’s how the experience of doctors in training, across all specialties, actually changed tremendously. I mean, you could have been a dermatology intern but have to staff a COVID I.C.U. And I’m curious what you think that shock to how doctors were trained, how that will impact the future supply of physicians who are interested in critical care medicine or who have some expanded expertise to be able to manage those issues, in whatever line of work they go into.

MARTIN: What’s interesting and maybe not surprising, is that we’ve seen a lot more people applying and matriculating into healthcare programs, more people applying to medical school, nursing school. That’s fantastic. Are they gonna choose to work in a specialty, like pulmonary and critical care or emergency medicine where you’re gonna be expected to work long hours to provide the care, and maybe be stretched beyond your normal capacity? We’re looking at the different generations of physicians and nurses and other providers and trying to understand how we can best support that workforce.

As a critical care physician, most of Dr. Greg Martin’s clinical work focuses on acute illness, caring for patients in the throes of a current and serious medical condition. He’s often making predictions about what will happen to their health on a daily, or even hourly, basis. Over the last few years, though, Greg has become more interested in long-term predictions. It’s part of his job now as director of Emory’s Predictive Health Institute.

MARTIN: At the individual level, the idea is to try and proactively prevent illness and even to predict illness. And that could be anything from diabetes, sleep apnea, high blood pressure, all the way through cancer. And all of those diseases have a relatively long prodromal period, meaning that the disease is beginning to exist before we actually are able to make the diagnosis. If we could identify that period, that’s the opportunity for making an intervention that really puts people back on the healthy trajectory and it may completely prevent the onset of that illness. Our system is really much more oriented towards disease care than healthcare.

JENA: I think there’s sort of two benefits of prediction. So one is there’s a benefit to people because they can make lifestyle changes, they can get on treatments that can change the trajectory of disease. But is there a benefit of prediction about future health, even if there are no adequate treatments?

MARTIN: It’s something we’ve also dealt with within predictive health. And it often comes down to an educational and even an ethical issue of what do you do with the information? there are some people who would want to know and other people would say, no, I don’t want to know. We’re still really, trying to understand the best way to do that because we’re not far away from being able to genotype every individual in the world. And once we get there, we really need to understand what to do with that information.

What should we do with all of this information — about our own personal health, and also the healthcare system itself? They say what’s past is prologue, so what can we expect to see from this pandemic, based on prior ones? Melissa Thomasson has some ideas.

THOMASSON: Healthcare can develop sometimes in ways we don’t think about. For us, looking back, it’s really natural to say, well, of course communities built their own hospitals. But at the time, it was anything but clear. When I look at the pandemic today, what I see is new methods of operation. I see things like psychotherapy services that are available on a nationwide basis. There are places now that will treat obesity and it’s all teletherapy. A lot of the emergencies that paramedics respond to don’t necessarily require transport to an emergency room, they can now link up on their iPads and treat those patients themselves. That saves the healthcare system a lot of money and saves the patients a lot of discomfort. Virtual surgeries? So maybe everybody can get a Cleveland Clinic experience, even if you’re at a local community hospital. And that has important ramifications for patients in developing countries, patients in rural areas, patients where there is maybe a lack of effective medical care. And again, I think that the  COVID-19 pandemic accelerated this.

I guess we’ll just have to wait and see, but hopefully not for another century. That’s it for today’s show. I’d like to thank my guests, Melissa Thomasson and Dr. Greg Martin. And thanks to you, of course, for listening to the show, and for telling your friends and family to do the same! Let me know what you thought about today’s episode! What changes have you noticed in healthcare since the pandemic started? What do you think we’ll see 50 or even 100 years from now?

Here’s an idea to leave you with based on my conversation with Melissa and Greg. During the COVID-19 pandemic, many hospitals have been forced to develop expertise in treating critically ill patients in a way that they hadn’t before. It makes me wonder whether this current pandemic could have a silver lining if the quality and outcomes of hospital care improve as a result. One way to study this would be to look at smaller hospitals that were disproportionately affected by COVID-19, to see if their overall hospital mortality rates for other conditions fall in the future, compared to what they were before the pandemic.

Think about it and in the meantime, next week on the show: More people are insured now than ever. But does health insurance make us healthier?

BAICKER: It is an obvious question to ask, but it is a much harder question to answer than you might think.

I’ll talk with my friend Kate Baicker about a one-of-a-kind experiment she led that tries to answer that question.

BAICKER: It is a much more complicated story than simply saying, “Yes, insurance worked,” or, “No, it didn’t work.”

That’s 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 at @drbapupod. This episode was produced by Julie Kanfer and mixed by Eleanor Osborne.Lyric Bowditch is our production associate. Our executive team is Neal Carruth, Gabriel Roth, 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.

THOMASSON: I don’t know that we pay enough attention to history, but we should really pay attention to history. It seems like things tend to repeat themselves.

Read full Transcript


  • Greg Martin, professor of medicine and director of the Predictive Health Institute at Emory University; chair of critical care at Grady Memorial Hospital.
  • Melissa Thomasson, professor of economics at Miami University.



Episode Video