Search the Site

Episode Transcript

 LEVINE: About one in 10 people that walk into an American hospital will have something wrong happen to them because of the medical care delivered. That’s crazy! Like we would never walk onto an airplane if one in 10 times something wrong was going to happen. But we walk into hospitals all the time.

Dr. David Levine knows a lot about hospitals.

LEVINE: We have the most advanced hospital care in the world. We have an enormous number of hospital beds. We have incredible technology and incredible people in those hospitals. Unfortunately, we haven’t built the right systems and we have created a place that is often unsafe and causes harm to folks.

David is a primary care physician at Brigham and Women’s Hospital in Boston, and he’s right: hospitals aren’t always the safest place to be. For starters, there’s the risk of infection. According to a 2018 study, at any given time, one out of every 31 patients in a U.S. hospital had an infection that they picked up during their stay. And that was before the Covid-19 pandemic. Hospitals do a lot of good for people, but they’re also a place where medical errors occur; in the U.S., these errors cause around 100,000 deaths per year. A study published just last week by today’s guest David Levine, and others found that a quarter of hospitalizations in the state of Massachusetts involve an adverse event — and that a quarter of those adverse events are preventable. Hospitalization also takes a toll on mental health, making patients feel anxious and depressed.

Last week, I talked with emergency medicine physician Ziad Obermeyer about one of the toughest decisions that emergency doctors have to make — whether to hospitalize a patient or send them home. But what if … you could do both?

From the Freakonomics Radio Network, this is Freakonomics, M.D. I’m Bapu Jena. Today on the show: how are doctors like David Levine changing what it means to be hospitalized?

LEVINE: Many folks have said, if home hospital were a pill, everyone would want to take it.

We’ll talk about how home hospital programs work — and who may benefit from them the most.

LEVINE: It appears that home hospital lifts all boats, but it actually may be lifting boats of folks who are of a lower socio-economic status more.

And: why this approach could transform hospital care altogether.

LEVINE: The sky’s the limit, frankly, in what you can provide in the home.

*      *      *

LEVINE: I ended up doing my first home visits as a high school teacher.

Before going to medical school, Dr. David Levine taught chemistry at a high school on the west side of Chicago.

LEVINE: It was a tough neighborhood and my students, they’d get shot walking outta school. Many of them would be pregnant. And they’d miss a lot of school.

So he started visiting his struggling students at their homes.

LEVINE: The second you walk in, you would get it. You would see everything that they’re living through and living with, and you could actually be much more of an advocate for them as a result. And I took that lesson of home visits as a teacher, ended up going to med school, and always had this concept that it didn’t make sense that we were bringing people to us as physicians. Instead we need to be going to where people were.  

As I talked about on the show a few weeks ago, it’s only in the last century that patients have routinely gone to the doctor’s office or to the hospital to get medical care. Before then, the doctor mostly went to the patient, at their home.

LEVINE: A hundred years ago, a person of means would never be caught in a hospital. That’s where somebody without any money went if they were going to die or were sick. And we flipped that script because we created large, mega industrial commercial complexes of medical centers and hospitals, and we have really sent an amazing message to Americans over the last hundred years that you go to a hospital when you’re sick. And I think we have a huge public service announcement that we need to make, that actually getting your care at home is either just as good or better.

Years after he visited students in their homes, this experience would come to define David’s work as a doctor. In addition to being a P.C.P., he’s also medical director of the Home Hospital program at Brigham and Women’s Hospital, which is one of the country’s largest home hospital programs.

LEVINE: I’m an old-fashioned primary care doc, Bapu. I work at a community health center called Brookside in Boston in Jamaica Plain where we see mostly patients with Medicaid and who are underserved. Most of my research is in advanced home-based care, where we try to deliver the types of care that you’d normally only see in the hospital or in a facility. And we use technology and the right kind of people to deliver that care actually in your home.

But delivering primary care and delivering hospital-level care are two very different things.

LEVINE: Even doctors get wide-eyed when we talk about some of this stuff. “Are you crazy? Are you really gonna take that patient home? They’re so sick!” The fact is there is a really big proportion of patients that we don’t serve well because they’re homebound. And so what a lot of folks many years ago said was, we need to deliver these patients all their care in their home. In the United States, we started to do experiments in the late 90s in home hospital care, simple I.V. treatments, simple lab draws, things of that sort. And then, fast forward about 20 years, we started to use models that really required a lot more technology. So things like wearable patches that would serve as remote wireless telemetry and point of care systems that would get blood test results in minutes in the home and ultrasound systems that could get images and deliver them back to radiologists really fast too.

JENA: There’s a very clear rationale for why you would need to take someone who was sick and keep them in the hospital. They need to be monitored. There needs to be adequately trained staff to diagnose problems, to treat problems. What’s the rationale for hospital at home?

LEVINE: I think the rationale is threefold. One is that we actually don’t do as good a job in hospitals as we think we do. There’s a lot of harm that we cause in hospitals. When you don’t pool all these sick patients together, they don’t share infections. When you don’t have patients across a curtain from one another, you have fewer med errors. So I think the first rationale is really there’s the potential to reduce the harm that we see in traditional inpatient hospitalization. The second one is actually capacity. Particularly at a time like now when we’re seeing the triple-demic. Home hospital offers a system the ability to create capacity without having to build brick and mortar beds. And then I think the third reason is cost. We and many others around the world have shown that when you deliver the care at home, it’s actually less expensive. And we all would love to see innovative care models that are able to deliver high quality care, but at a lower cost.

JENA: And there’s probably a fourth thing too, which is nobody wants to be in the hospital. It’s incredibly disruptive to the patient, it’s disruptive to the family. And I can imagine that the experience of care, assuming all things go well medically, would be much greater if you were able to get care at home.

LEVINE: Yes! The experience of care is just leaps and bounds better. The one big risk is that you’re further away when you’re at home than in the hospital. Thankfully most patients don’t have urgent or emergent events when they’re in the hospital. Some obviously do. We are pretty good at predicting those though. And so we have a whole set of inclusion-exclusion criteria that really help us hone in on the right patients to take home, i.e. the ones that aren’t going to need that super emergent help. And if something does go wrong, we have a team of mobile integrated health paramedics who can be out in the patient’s house within 20 to 30 minutes. We have physicians on call 24/7 who are instantaneously available by video and can be in a patient’s home 24/7. And then, as the supreme backup, we have traditional 9-1-1. Which in our urban area has an average five minute response time. So patients have a lot of tools at their disposal when they’re at home. It may not be as fast as a code team. That said, code teams take time to assemble, too, in the hospital.

JENA: Can you give me a specific example of a type of person who would be a good candidate for getting hospital care at home?

LEVINE: There are usually three input pathways for home hospital. I’ll give you the most common one, which is the patient presents to the emergency department. They’re sick. Let’s say, this patient has pneumonia and needs supplemental oxygen, I.V. antibiotics, close monitoring of their vital signs, maybe some intravenous fluids. Patients in the emergency department, they receive their triage, their diagnosis, their first round of treatment, and they get approached by the home hospital team, after discussion with the emergency department team and then would be transferred from the hospital to the home hospital team, and that team would then deploy and meet the patient in their home.

JENA: And what happens next? So, who’s seeing the patient at home? How often are they there? Who’s giving medications? Who’s doing monitoring? How does all that stuff happen?

LEVINE: So, there’s definitely a variability across the country, in how this happens. And I think that’s a healthy variability. I can tell you, our team in Boston, when that patient gets home, there’ll be a nurse or a paramedic waiting there for them. There will also be our physician who will staff them in their home on admission. And at that point, they’ll get patched, meaning they have a sticker that we put on their chest, which helps us get their continuous heart rate, respiratory rate, temperature, movement detection. And then we will do a typical admission, just like you would in the hospital, with all the extra bonuses of being in their home. So we get to see are the cupboards bare? Are the windows broken? Is the carpet torn up? Is there an infestation? All sorts of things that we can actually be advocates for them for. And then at that point, the team leaves. The monitoring stays, the patient has 24/7 connectivity directly to the doctor. 24/7 connectivity to a dispatch system. And then has a team that could come out to their house at two in the morning if there’s a problem. Maybe they become short of breath, something like that. And then daily rounds start the next day just like they would in the hospital.

JENA: I mean, this is a wonderful story. What’s the actual evidence on it?

LEVINE: I’ll start with the international evidence. There are over two dozen randomized controlled trials that have been done in places like Australia, Italy, as well as Spain. Meta-analyses have shown unbelievable reductions in readmission when people get home hospital care. So much so that many folks have said, if home hospital were a pill, everyone would want to take it. You die less, you get readmitted less, and you’re much happier when you get your care at home. A lot of people always say, “Eh, that’s not the American system. Doesn’t matter.” The first piece of evidence that came out of the United States was in 2005 in the Annals of Internal Medicine, which was a non-randomized demonstration in the Medicare Advantage population at three sites across the country. And again, really corroborated the international evidence that patients had fewer adverse events when they were in the home, and they had very high experience and lower costs. Really great evidence then came out from Mount Sinai in New York City in 2018 in JAMA Internal Medicine. Again, a non-randomized study that showed the same thing: lower cost, happier patients, less 30-day readmission. We wanted to do the first randomized control trial.

So, how did it go?

LEVINE: The trial actually stopped early because our C-suite was really, really enamored with the results and wanted it for all of our patients.

That’s after the break. I’m Bapu Jena, and this is Freakonomics, M.D.

*      *      *

LEVINE: So just to paint the picture, these patients are in the E.D., they’ve been admitted, and literally we walk in and say, “Hey, are you interested in maybe going home?”

Before the break, David Levine was telling us that there had never been a randomized controlled trial of home hospital programs in the United States … until 2017, when he and his team at Brigham and Women’s Hospital decided to do one of their own.

LEVINE: Our group started to recruit a randomized cohort in the emergency department.

They ended up recruiting 91 patients who said they were interested in the possibility of going home. Around half of them were randomly selected to participate in the home hospital program — while the other half, the control group, stayed in the hospital. Then, something unusual happened.

LEVINE: The trial actually stopped early because our C-suite was really, really enamored with the results and wanted it for all of our patients. We were able to show that readmission rates plummeted when you were home hospitalized versus traditionally hospitalized. Cost was much lower, and folks had really, really nice outcomes during their home hospitalization.

JENA: Tell me about the cost differences.

LEVINE: It was an extremely manual process that we did for both the intervention group and the control group. So we totaled every Band-Aid, every gauze pad, every bag of vancomycin, looked at nursing labor, tech labor, transport time, basically everything you could think of that we were able to total up in both arms equivalently, we did. Through that direct addition of all the costs, we were able to show that cost at home was 38 percent less than cost in the hospital.

JENA: Can you tell me about how home hospital affects different groups from an equity lens?

LEVINE: The equity lens is extremely important in home hospital. Folks at Mount Sinai have actually looked at patients who were Medicaid members and they looked at patients who were non-Medicaid members and compared their outcomes. It turns out actually that folks who were on Medicaid ended up coming back to the emergency department less often than folks not on Medicaid. And for those who know the literature, that’s unheard of. Of course folks on Medicaid unfortunately, are often coming back to the emergency department more often. It appears that home hospital lifts all boats, but it actually may be lifting boats of folks who are of a lower socioeconomic status more. And I think the reason for that is because of the power we have in the home. When we are in somebody’s home, we have this amazing opportunity and frankly responsibility to act on the social determinants of health that we see in their home. If the cupboards are bare, if the window is broken, you better believe that our team is knocking on the super’s door the same day and saying, “I’m sorry, this patient’s in a wheelchair and can’t even wheel themselves on their floor. That’s not to code. This needs to be fixed immediately.” We have a lot of agency as the medical team, and so we are able to make a whole lot of impact for patients when we’re in their homes that I think then bears out on some of that data.

JENA: Are there patients for whom this does not appeal so much? I mean, I could imagine it would be very stressful for some caregivers and in fact they might even get some respite by having the patient, their loved one or family member, in the hospital.

LEVINE: There are definitely patients that do not fit the bill. There are definitely patients that should stay in the hospital and some of those things are clear-cut environmental criteria. Patients who are experiencing homelessness are not good candidates for home hospital. Patients with things like domestic violence in the home or without running water in their home — those are what I consider as non-therapeutic homes, and so we often wouldn’t want to take a patient there, while they’re really down for the count with an acute illness. That said, your comment about caregivers is really important because there’s this notion that when you go home, it ends up putting all this burden on the caregiver to help their loved one, and I think that absolutely could be the case. But many home hospitals, they’re actually giving the caregiver the tools they need in order to keep their loved one at home. We can deploy a home health aide for a 12-hour shift at night. So you sleep, as the caregiver, and there’s somebody there to take care of your loved one while they need to get to the bathroom.

I think the flip side is really important, which is the story of how burdensome it is to be a caregiver for somebody in the hospital. In Boston, you drive half an hour in traffic, you pay $50 to park, you get lost finding your loved one’s room, you sit in a cramped two-person room. Because you got lost, you missed the team rounding and the only person that ever comes back to talk to you is a medical student. And then you have to pay $30 for terrible cafeteria food. And then it’s the end of the day and you have to drive home at night, and you’re exhausted and you feel like you didn’t do anything for your loved one.

JENA: David, you make me not want to get hospitalized anywhere.

LEVINE: I don’t wish hospitalization on you ever, Bapu, but, if you did need it I hope we could do it at home.

JENA: Is there any risk that the monitoring gives us a false sense of security that we can do something remotely that we really should be doing in person?

LEVINE: I think that’s definitely a risk. Part of my job is to innovate around new pathways, and I’m always afraid that a new pathway could cause harm. That’s why we actually do it really, really carefully. I’ll give you an example. We have a randomized controlled trial recruiting right now for bariatric surgery. We’re actually taking patients right from the O.R., and we’re taking them home. We’re doing that only for half the patients, though. Half of them stay in the hospital, and that’s by design, because I wanna make sure that the folks that go home are getting, if not the same, better outcomes than the folks that stay in the hospital.

JENA: The pandemic changed a lot about how we provide medical care. How did home hospital change?

LEVINE: C.M.S. saw the massive crush on hospital capacity that was happening, and had the brilliance to open up a regulatory and payment pathway for home hospital care.

C.M.S. stands for the Centers for Medicare and Medicaid Services. That’s the federal agency that oversees American public health insurance programs. And it plays a crucial role in the story of home hospital in the U.S. Because when a new drug or, in this case, a new care model hits the scene, C.M.S. has to figure out how to pay for it. When the pandemic started and hospitals were overflowing with sick Covid patients, C.M.S. wanted to devise a quick — but temporary — way to pay for home hospital programs like the one David runs.

LEVINE: It was a fascinating story and an amazing example of C.M.S. being really, really flexible and innovative during this pandemic.

In the fall of 2020, David and his colleague Bruce Leff from Johns Hopkins helped craft an “Acute Hospital Care at Home” waiver for C.M.S. They did this in just a few weeks, right before Thanksgiving.

LEVINE: Because of how fast everything was moving, C.M.S. wanted no changes at all to the billing system because there was no way to put those changes into effect. And so patients that got cared for at home looked exactly like patients who got cared for in the hospital in claims data. This is a waiver that you have to apply for and your hospital undergoes a pretty thorough vetting actually. C.M.S. interviews that home hospital program. And then if they’re satisfied, they end up approving you and then you’re able to deliver hospital level care at home.

This past December, Congress extended that waiver program by at least two years. So, now that there’s a payment model in place, what could the future hold for home hospital care in the U.S.?

LEVINE: I think the future is a ubiquitous home hospital care system. You walk into any hospital in America and its second nature to turn right back around to the home hospital team if you’re acutely ill. That is the exception right now in the United States. We only have it in 250 some odd hospitals. There’s over 5,000 acute care hospitals. I think with the passage of this legislation you are going to see hospitals take to this care model in a skyrocketing proportion. Patients will walk into that E.D. and it’s gonna be their expectation that that kind of service line is available at the hospital.

Other recent data suggests that we might already be trending in the direction of more out-of-hospital care: home births in the U.S. have reached their highest levels since at least 1990, jumping by 12 percent from 2020 to 2021 alone. And telemedicine now accounts for about 5 percent of all medical claims. Looking ahead, could the expansion of home hospital programs bolster these changes, and perhaps inspire other home-based care options?

LEVINE: Absolutely because once you have the pieces to deliver this care, you have the nursing team in the home, you have the paramedic team in the home, you have the physicians who are facile at video-based management as well as home-based management. You have the pharmacy system set up to deliver this kind of care. As soon as you have that chassis, the sky’s the limit, frankly, in what you can provide in the home.

The concept of getting hospital-level medical care in your own home can be tricky to wrap your head around. But after talking with David, it’s clear that a growing body of evidence shows home hospital programs are safe for patients — at the least the right ones — and could even save hospitals money. If I ever need to be hospitalized, I’d be willing to try it out. Netflix and pill, I guess.

Anyway, I hope you enjoyed the episode and learned something new about what the future of hospital care could look like. And here’s an idea I had based on my conversation with David. He mentioned that errors sometimes occur in hospitals when two patients share the same room. Medications can get mixed up, for example. Does this also happen for patients with the same last name? Suppose you had electronic health record data and looked at adverse events to patients during hospitalizations where, by chance, there happened to be another patient hospitalized on the same floor with the same last name. Would you see more medication errors or other mistakes when Ms. Jones happens to be hospitalized the same week as another Ms. Jones?

Think about it. And in the meantime let me know what you thought about today’s show. I’m at That’s B-A-P-U at

That’s it for today. I’d like to thank my guest this week, Dr. David Levine. And thanks to you, of course, for listening. Coming up next week: what can referral patterns between physicians tell us about discrimination in medicine?

 SARSONS: Other men are kind of unaffected by the actions of one man, whereas other women are affected by what happens to one woman.

We’ll talk about what the findings of a labor economics paper could mean for female surgeons — and also, for their patients.

 SALLES: When we don’t support women physicians, that means we’re not supporting the patients who are being cared for by those women physicians.

That’s 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 Lyric Bowditch. It was mixed by Eleanor Osborne with help from Jasmin Klinger. 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.

JENA: Where do health economists fall in this safety net? When do they get called?

LEVINE: Bapu, I’ve tried to call you many times. You never pick up the phone at two in the morning.

Read full Transcript


  • David Levinegeneral internist, clinician-investigator, and medical director of the Home Hospital program at Brigham and Women’s Hospital.



Episode Video