Search the Site

Episode Transcript

Hey there, it’s Stephen Dubner. Before this week’s episode, let me ask your help for a future episode we’re working on. It’s about curses and superstitions. Which ones do you believe in, and why? Give us an example! Tell us a story! We’d like you to send us a brief audio recording. Just go in a quiet room and use your phone’s voice-memo app  — and e-mail the file to Please include your name, where you’re from, and any other information that might be relevant. And if we don’t hear from you — beware the Freakonomics Radio curse. Thanks in advance. Our episode this week is one of my very favorites, from the vault, published nearly a year ago. It’s called “How Do You Cure a Compassion Crisis?” I thought it was worth hearing again as we start the holiday season. Speaking of which: if you’re looking for gift ideas, how about some Freakonomics swag? We just got in some good new stuff: just go to and click on “Shop.” And now, “How Do You Cure a Compassion Crisis?” Thanks for listening, and happy holidays.

*      *      *

Today’s episode is really interesting. There are two ways we could start it, but I can’t decide which is better. One way is nasty and vulgar, like this:

Tiffany INGHAM: Well, why are you looking then, retard?

Or we could start with the uplifting story, like this:

Stephen TRZECIAK: Scientists define compassion as an emotional response to another’s pain or suffering, involving an authentic desire to help. 

I don’t know, I can’t make up my mind. What do you think? All right, let’s just flip a coin. Heads for nasty, tails for uplifting. And it’s tails. Okay, we’ll get to the nasty stuff later. So, let’s start here:

Anthony MAZZARELLI: My name’s Anthony Mazzarelli. I am the co-president and C.E.O. of Cooper University Health Care.

The Cooper Health System takes in about $1.5 billion in annual revenues:

MAZZARELLI: We’re a level I trauma center located — our core hospital in Camden, N.J. But we have over 100 sites.

Mazzarelli doesn’t just run the hospital.

MAZZARELLI: I’m also a practicing emergency-medicine physician.

People who know Mazzarelli call him Mazz. And Mazz is sort of an overachiever. During his medical training, for instance, at the University of Pennsylvania:

MAZZARELLI: I ended up graduating with a medical degree, a law degree, and a master’s in bioethics. And then did my residency training in emergency medicine here at Cooper and have not left Cooper since. 

In 2014, Mazzarelli was promoted to chief medical officer:

MAZZARELLI: And at that time, our institution had engaged a consultant.  

The consultant did what consultants do and suggested ways for the hospital to strengthen its bottom line. One idea: the hospital should focus on improving patient experience and physician engagement.

MAZZARELLI: And there was a list of things to ask our physicians to do, which seemed, frankly, soft. Things that were kind of mushy. Things that I was concerned that I was going to have trouble getting 450, 500 faculty members — I was going to have trouble getting them to do.

Some of these mushy things had to do with the relationship between doctors and patients. Mazz realized that most of what he was being asked to do was to get doctors to show more compassion. Now, you might assume that most people who choose medicine as a profession do so in part because they are compassionate. Or at least that they’re taught compassion during medical school. If that’s the case, where does it go? Does compassion somehow evaporate over time? If so, was there a viable way to increase it? And can compassion even be measured? Before doing anything, what Mazz needed was some research.

MAZZARELLI: So, I turned to our No. 1 N.I.H.-funded researcher, the person with the most publications, the most N.I.H. dollars—. 

TRZECIAK: I’ve been at Cooper for 17 years.

MAZZARELLI: And it was Steve.

TRZECIAK: My name’s Steve Trzeciak

Trzeciak is the chairman and chief of the department of medicine at Cooper. And also:

TRZECIAK: I’m a research nerd and I’m also a practicing intensivist, a specialist in intensive-care medicine.  

Here are two doctors — one specializing in intensive care, the other in emergency medicine — who between them had treated thousands of people who were each having one of the worst days of their lives:

TRZECIAK: So, I never doubted that compassion was essential. And I don’t know anyone in healthcare that feels otherwise, or at least no one that would admit to it. It’s what we ought to do. The way that we ought to treat patients. But does it actually move the needle on outcomes in a measurable way? That’s what I was skeptical about.

But remember, the consultants the hospital brought in did want that needle moved. Mazzarelli was willing to consider that a dose of compassion might be worthwhile. So, he reached out to Trzeciak:

MAZZARELLI: And sat with him and said, “Steve, can you science this up for me? Can you look at the data around this?” And he said, “No, you’re crazy. I don’t want to be part of this. This is mushy. This is not what I do. I’m a hard-science guy.”

DUBNER: Is that indeed what you said, Steve?

TRZECIAK: It is, but that’s because I didn’t know there was such hard science available. 

DUBNER: How much time did you think you’d have to waste on Mazz’s silly idea? 

TRZECIAK: Well, it didn’t take long before I started to see the beginning of the signal in the data. And that’s when everything started to resonate.  

What Trzeciak was seeing in the data resonated with something else that was happening in his life.

TRZECIAK: Where I sort of had an existential crisis, like, “What am I going to do with my career?” So, I just want to be clear about one thing. I was not in the market for any sort of a scientific awakening. My research program was hitting every metric for success. We were publishing in some of the best journals. Everything was fine, right? But then I had this question that was posed to me.  

He doesn’t mean the question from Mazzarelli.

TRZECIAK: The question came from my son.

Trzeciak’s son was 12 years old. The question actually came from a school homework assignment.

TRZECIAK: The question was, “What is the most pressing problem of our time?” 

Trzeciak talked over the assignment with his son. But then on his own, he kept thinking about the question, and how it applied to him.

TRZECIAK: I knew that the research that I was working on was very important, but I also knew that it wasn’t the most pressing problem of our time. And I’m not old, but I’m too old to work on things that don’t really matter. It led me to search for what is the most pressing problem of our time.

And after a couple of weeks digging around in this seemingly mushy research project that Mazzarelli put him on, he knew he’d found it — what he considered the most pressing problem of our time. At least in his field.

TRZECIAK: What I found in the data and also just looking around at healthcare, what I see is that we have a compassion crisis. You can either believe it matters or it doesn’t matter. But if it matters, how does it matter? How does it affect people? How does it affect healthcare? How does it affect the economics of healthcare? How does it affect healthcare providers and burnout.

Trzeciak and Mazzarelli wound up writing a book that tries to answer these questions. It’s called Compassionomics: The Revolutionary Scientific Evidence that Caring Makes a Difference. Today on Freakonomics Radio: how strong is that evidence? How has the pandemic changed the equation? And: is it just the healthcare system where we need more compassion?

*      *      *

Okay, let’s start by revisiting that definition of “compassion.” Stephen Trzeciak again:

TRZECIAK: Scientists define compassion as an emotional response to another’s pain or suffering involving an authentic desire to help. 

That sounds like exactly what you’re setting out to do if you become a nurse or nurse practitioner; a physician assistant or a physician.

TRZECIAK: And the hypothesis is compassion matters. We don’t just mean compassion matters in a moral or ethical or sentimental sense. We wanted to test the hypothesis that compassion matters in measurable ways for patients and for those who care for patients.  

Here’s one data point that Trzeciak and Mazzarelli cite in Compassionomics: when patients are asked what they consider “extremely important” traits in a doctor, 85 percent say yes to being treated “with dignity and respect.”

MAZZARELLI: Only 27 percent say they want them trained in one of the best medical schools. Only 58 percent say, “Has a lot of experience.”  

Anthony Mazzarelli again.

MAZZARELLI: Patients want these factors that are more on the spectrum of empathy and compassion.  

TRZECIAK: So, we do want to be crystal clear about one thing. The No. 1 driver of clinical outcomes is clinical excellence. If you’re a surgeon and you botched the surgery or if you’re a physician who prescribes the wrong medication, there is no amount of compassion that’s going to undo that. It’s not an either/or, it’s an and. So, it’s compassion and clinical excellence that produce the best clinical outcomes. 

DUBNER: When you’re going to look at data about the value and effect of compassion in medical care, how do you search for that? I’m guessing “compassion” is not a keyword in all these medical studies. 

MAZZARELLI: Yeah, it’s a great question because you know, you can look up “pneumonia” in PubMed. PubMed is the equivalent of Google. And you look up “pneumonia,” you’ll get every article on pneumonia. If you were to look up “compassion” or “empathy,” you will not necessarily get every article. So, we had to do something which was essentially the equivalent of the Dewey Decimal System. We had to go back and do a systematic review of a reference-of-references approach — whole articles and then read all of those references, and then read the references of those articles, and then the reference of those articles, and keep doing that like a giant tree. That’s why it took a couple of years to do. 

So, rather than seeking out empirical evidence on compassion per se, Mazzarelli and Trzeciak organized their research around a set of characteristics that make up what is called “patient-centered care.” These include kindness, empathy, warmth — pretty much anything that shows doctors being nice to their patients. A lot of the research they looked at involved a 10-question survey called the CARE Measure. Patients are asked questions like: How well did the doctor do “at making you feel at ease?” How well did the doctor do at “fully understanding your concerns?” At “showing care and compassion?” At “making a plan of action with you?” In seeking out evidence on patient-centered care, Mazzarelli and Trzeciak wound up reviewing 281 research articles that formed what they saw as a collage of evidence about the power of compassion.

DUBNER: Before we hear your argument and your evidence for the argument, let me just ask how persuaded are you that you’re right? Because I could imagine that we could identify benefits of compassionate care, but it may be that doctors show more compassion to patients who are more compliant. So, how persuaded are you that the outcomes are not driven by something else, whether observable or unobservable? 

MAZZARELLI: Well, Steve used to have to correct me a lot on this, because I am a lawyer also. He would always say to me, “Look, we’re not making arguments. We’re testing a hypothesis.” He would say, “We need to be equally open that compassion isn’t something that is measurable and meaningful.” 

TRZECIAK: Right. So, it’s important to recognize the difference between association and causation. And causation can only be inferred from certain study designs. But what is really compelling and to specifically get to your question, when you push all the data together and you see it all curated, essentially for the first time, the signal is so consistent across the studies that it really doesn’t make a whole lot of sense to conclude anything else. 

Okay, let’s hear some of the evidence that Trzeciak and Mazzarelli compiled.

TRZECIAK: Sure. So, first, it’s important to think about mechanisms. There are many broad categories by which compassion for patients can be beneficial, and the first is physiological. Compassion for patients can actually modulate a patient’s perception of pain. It can have immune-system effects. There are also endocrine effects, which means in patients with diabetes, there’s evidence that they have better blood-glucose control and fewer complications when they’re treated with compassion on a regular basis. There is also broadly psychological effects. So, compassion for patients can reduce symptoms of depression, reduce symptoms of anxiety, reduce emotional distress associated with somatic illnesses like having cancer.

Those are some amazing (and amazingly concrete) claims for something that both Trzeciak and Mazzarelli suspected might be mushy. So, let’s interrogate this evidence. Consider Trzeciak’s first claim:

TRZECIAK: Compassion for patients can actually modulate a patient’s perception of pain.

How did they reach that conclusion? Their book cites several research papers that take a variety of approaches. One was a randomized-controlled trial done at Harvard Medical School with patients suffering from irritable bowel syndrome. It found that compassionate care — for instance, a doctor simply saying “I can understand how difficult I.B.S. is for you” — this led patients to report significantly higher rates of symptom relief. Another study, this one from Michigan State University, also used a randomized experiment. That’s the good news. The bad news is the subject pool was tiny: just nine patients.

In any case, these nine patients were recruited from the waiting room of a primary-care clinic and randomly divided into two groups. The control group got their standard visit with a doctor. The treatment group got the “compassionate” version, with the doctor engaging in warm conversation, trying to make the patient feel at ease, and encouraging follow-up questions. Afterward, the researchers put all nine patients in an fMRI machine, in order to measure their brain activity. Each patient was then given a painful stimulus while being shown an image of the doctor who’d seen them. The treatment group — that is, the patients who’d received the “compassionate” care — showed 47 percent less activation in the region of the brain known for experiencing pain. Again: it’s a small study and fMRI evidence is hardly perfect — but still: this type of study has persuaded Trzeciak that compassion can indeed modulate pain:

TRZECIAK: So, I didn’t say eliminate pain. But attenuate pain or one’s experience of pain.

And what are the mechanisms by which this happens?

TRZECIAK: One of many potential mechanisms by which compassion can modulate their pain is the release of endorphins. So, when endorphins are circulating, they are essentially natural opioids. 

Trzeciak also believes that compassion creates trust between patient and doctor.

TRZECIAK: In many ways, the touch of a trusted other can reduce one’s experience of pain. 

A study from the University of Haifa in Israel, for instance, gave people a painful stimulus while holding the hand of either a stranger or a loved one. Holding a stranger’s hand didn’t lessen the pain at all. But people reported a 50 percent reduction in pain while holding the loved one’s hand. There’s other research showing broader claims about human connection.

TRZECIAK: There’s evidence that human connection also modulates or can affect one’s autonomic nervous system. So, the autonomic nervous system is the part of the nervous system that does everything that you don’t have to think about. Like controlling your heart rate and your cardio-respiratory system.

Trzeciak points to evidence that compassion also affects what’s known as the parasympathetic nervous system. This can boost the flow of oxytocin, a molecule known as the “trust hormone.” Now, these physiological benefits of compassion are, to me at least, quite surprising. Somewhat less surprising are the reported psychological benefits.

TRZECIAK: That’s probably intuitive to some extent, that treating someone with compassion can help their mental health. But we’ve also seen this in a study that we’d recently published here at Cooper.

This study was led by Brian Roberts, an emergency-medicine doctor.

TRZECIAK: Brian did a study on the effects of compassion and the subsequent development of P.T.S.D., post-traumatic stress disorder. 

He was studying people whose P.T.S.D. didn’t come from war or some traumatic loss. It came from spending time in the hospital.

TRZECIAK: So, approximately one-third of patients that go through the experience of critical illness in an I.C.U. end up making diagnostic criteria for P.T.S.D. at 30 days. Even if you just come to the E.R. with a life-threatening medical emergency, 25 percent of those patients end up making diagnostic criteria for P.T.S.D. at 30 days. 

Here’s the hypothesis Brian Roberts wanted to explore: that treating E.R. and I.C.U. patients with more compassion might decrease the prevalence of P.T.S.D.

TRZECIAK: And what he found was that more compassion from the patient’s perspective was associated with lower development of P.T.S.D. at 30 days. So, perhaps compassion for people while they’re going through terrifying medical emergencies can actually help them with their psychological effects down the road. 

DUBNER: So, I mentioned this compassionomics idea to one doctor friend of mine. He’s a gastroenterologist whose specialty is cancer care. He’s late 50s, early 60s. And he pushed back in the following way. He said that doctors like him used to practice lots of compassion because, he said, there wasn’t much else they could do once someone was diagnosed with cancer. And now that there are so many more treatment options, that he’d rather deliver a lot of science than a bunch of compassion. So, there’s an opportunity-cost argument to this, right? If we’re going to spend a lot of time teaching and/or focusing on these kind of softer skills, does the science suffer? 

TRZECIAK: Compassion actually takes almost no time. Like, less than a minute. There was a randomized controlled trial from Johns Hopkins in a cancer population, and the primary outcome measure was anxiety. If you have cancer or somebody close to you has, you know that anxiety is pretty important. And what they found is that the compassionate care had a significantly better effect on the patient’s anxiety level. But what was most striking is that it only took 40 seconds for the intervention. And we found five other studies which show that it is less than a minute. And some people would argue there should be no time dimension at all. Because it doesn’t take any extra time to treat somebody with compassion. 

MAZZARELLI: I think a fair response back might be, “Okay, fine, you found five studies, and it’s less than a minute. But you’re opening yourself up to a ton more questions, you’re opening yourself up to a much longer visit.” But that’s also been studied, and there is no significant increase in the total length of time that people spend together. I think that the problem is that it is sometimes very hard in medicine to take on a new paradigm shift. 

To say “it’s hard for medicine to take on a paradigm shift” — that is an understatement. The history of medicine is replete with innovations that took years, sometimes decades, to work their way into the mainstream. Hand hygiene, for instance, as basic as that now seems. Medicine is a difficult enterprise, a complicated one, and in many ways a conservative one. With good reason. Remember: first, do no harm. So, imagine you are a hospital administrator. And some researcher comes into your office preaching the virtues of compassion. It has demonstrable physiological and psychological benefits, they tell you. It doesn’t take much time or effort, they tell you. Is that enough to convince you to round up all your doctors and tell them, on top of everything else they’re doing, that they also need to show more compassion to their patients? Maybe. But if not, what if that researcher also tells you that compassion will save you a lot of money? Are you paying more attention now?

MAZZARELLI: So, compassion increased revenue and decreased costs. 

How can compassion increase revenues?

MAZZARELLI: There’s patients who will pay more for that. We have data about hospitals that have higher margins that have better patient experience. 

That’s true, according to data collected by a federal survey of hospital patients. And hospitals that perform well on this survey are also reimbursed at a higher rate by the Centers for Medicare and Medicaid Services.

MAZZARELLI: But it’s the decreasing costs that I think is the most interesting.  

TRZECIAK: There is consistent evidence that when you care deeply for patients, and they know that, they’re more likely to take their medicine. And non-adherence to medical therapy in the U.S. alone accounts for somewhere between $100 and $280 billion of avoidable downstream healthcare costs. 

MAZZARELLI: And if compassion is something that can help people be more adherent, even capturing a fraction of that could decrease costs in the healthcare system, which is approaching 19 percent of the G.D.P. That’s one way it can decrease costs. Another way is in studies where there’s really patient-centered care, the proportion of patients who were referred to specialists was 59 percent lower while those who underwent diagnostic testing was 84 percent lower

DUBNER: I can see how fewer referrals to specialists and less extra testing would certainly lower costs. But how do we know that those lower costs aren’t at the expense of better outcomes? Because obviously some referrals and some tests are necessary. 

TRZECIAK: So, there’s a whole section of the book dedicated to the data on quality of care. And we’ve found associations in the data between more caring and fewer errors. And many of us in healthcare have been exposed to folks — and fortunately they’re few and far between — who maybe don’t care as much as we think they ought to.

MAZZARELLI: Or they once did. 

TRZECIAK: Or they once did, right? If they’re burned out. If you go to medical conferences, there is one theme that is drowning out just about every other topic that’s being discussed. And that’s the topic of burnout among healthcare providers. 

Indeed, the World Health Organization recently added burnout to its international Classification of Diseases — not as a medical condition, but an occupational phenomenon. Plainly, medicine isn’t the only occupation where burnout can happen. But it is surprisingly common among doctors. So, what’s this have to do with compassion?

*      *      *

Anthony Mazzarelli and Stephen Trzeciak are doctors and administrators at Cooper University Health Care in Camden, N.J. That’s just across the Delaware River from Philadelphia. They have co-authored a book called Compassionomics, which argues that when doctors treat their patients with compassion, it improves medical outcomes and reduces costs. But there is a problem:

MAZZARELLI: There is a compassion crisis in healthcare. 

How can that be? How can the most caring of caring professions be lacking in compassion? Before we get into the causes, would you like an example? Of course you would. Remember at the beginning of this episode, I mentioned a certain nasty and vulgar incident? Let me just pause here to emphasize that this example is not representative of most healthcare professionals. Most people who get into medicine do so because they want to help people. They take a vow to uphold a standard. But occasionally that standard is violated. In 2013, for instance, a man went to have a colonoscopy at a medical facility in Reston, Va. There was the gastroenterologist who performed the procedure, an anesthesiologist, and a medical assistant. The patient planned to record the doctor’s instructions on his phone once the colonoscopy was over. But he accidentally recorded the whole procedure. From his nervous questions before things got started:

PATIENT: Sorry I have so many questions, I just — it’s just my first time doing anything like this.  

To the doctors talking about him once he was anesthetized:

INGHAM: He’s crazy.

They start talking about an earlier problem the patient had — a genital rash.

Soloman SHAH: He keeps mentioning it like it’s the first time he’s ever talked to anyone about it. I’m like, “Sir, you’ve seen two urologists. What are you telling me for?” And, also, don’t — don’t mention it to me, because I’m not interested. 

INGHAM: And I don’t care, exactly. And then he went on and on about it. And I’m like—.

SHAH: One of the nice things about being a specialist is I don’t deal with that.

INGHAM: One of the nice things about being an anesthesiologist is making people shut the hell up. 

SHAH: That’s why I didn’t become a freaking urologist. 

In case you didn’t catch that, the anesthesiologist says, “One of the nice things about being an anesthesiologist is making people shut the hell up.” And then she continues, alternately talking to her colleagues and the anesthetized patient.

INGHAM: After five minutes of talking to you in pre-op, I wanted to punch you in the face and man you up a little bit. So, just make sure you’re gowned up. Don’t want you to accidentally rub up against it. Get some syphilis on your arm or something. It’s probably tuberculosis in the penis, So, you’ll be alright. Just get a P.P.D. in like a month, and then you’ll take some I.N.H. and be fine. 

SHAH: As long as it’s not Ebola, you’re okay.

MEDICAL ASSISTANT: So, if you see a rash—.

INGHAM: It’s penis Ebola. 

“It’s penis Ebola,” she says, which is not a thing. Then she says she’s going to enter “hemorrhoids” on the patient’s chart.

INGHAM: I’m going to mark hemorrhoids even though we don’t see them and probably won’t. 

“Even though we don’t see them and probably won’t.”

INGHAM: “I’m just going to take a shot in the dark. 

This patient, after waking up and hearing the phone recording, sued the doctors. The gastroenterologist was ultimately dismissed from the case, but the anesthesiologist and her practice were ordered to pay a half-million dollars in damages. Again, this is just one incident, and an egregious example, for sure. But if you want to make an argument for the lack of compassion in healthcare, it’s a good example. On the other hand, if you want to make an argument that having compassion can save money, as Mazzarelli and Trzeciak argue, you can use this lawsuit as an example where the lack of compassion can be very expensive. And yet, they say, there is still a huge deficit in compassionate care. Here’s Trzeciak again:

TRZECIAK: The data suggests that physicians, specifically, miss approximately 60 to 90 percent of opportunities to respond to patients with compassion. 

DUBNER: Give me a simple example of a physician failing to exercise compassion. 

TRZECIAK: Let’s start with the most basic. Recent data from the Mayo Clinic show that physicians will interrupt patients in their statement of their main concern at the 11-second mark. That’s the median time to first interruption. So, patients may not even get to fully explain the main concern that they have. 

This kind of problem is driven in large part by how doctors are compensated. As you likely know, our healthcare system tends to put more value on procedures and tests than on conversation or prevention. This is hard for doctors themselves; many of them are frustrated that their profession — long viewed as a calling — has become so transactional. But there also appears to be a perception gap between physicians and patients. Consider a survey done by the Schwartz Center for Compassionate Healthcare, which included 800 recently hospitalized patients and more than 500 doctors. When asked if most healthcare professionals provide compassionate care, 78 percent of the doctors said yes; for patients, that number was just 54 percent.

TRZECIAK: So, that data, and there are other data to corroborate it, show quite clearly that many of our healthcare providers can have a blind spot with respect to how well they’re connecting with their patients. We are thinking that we’re providing them with the emotional support that they need but the data showed that’s not what we’re actually delivering. 

What’s missing in this equation? It appears to be empathy.

Helen RIESS: Yes. Empathy is a human capacity that allows us to perceive, process, and respond to others’ emotional states.

That’s Helen Riess. She is a Harvard psychiatrist who also practices at Massachusetts General Hospital, where she directs a program that does empathy research and training.

RIESS: There have been many studies, both in medical students and in practicing physicians, that demonstrate that there is definitely a deficit in empathy and compassion.

DUBNER: Let’s say on a scale of one to 10, what is the median American doctor’s empathy level? 

RIESS: I’d say it’s about four-and-a-half.

DUBNER: Oh, that’s discouraging, isn’t it?

RIESS: I think so. And it’s not to blame the doctors. I just think that our system right now is working to get the outcome that we’re seeing. First of all, medicine has become a business, and whereas we used to have time to get to know patients and to really form relationships, it’s much more about throughput now, and how many people you can squeeze into an afternoon. The incentives are much higher to see somebody for 20 minutes to just prescribe their medicine than to see them as a whole person. 

Riess argues that this scenario is a big driver of physician burnout. How is that defined?

RIESS: Burnout is defined when a few things are happening, called depersonalization, where patients are seen more like as a number, or a diagnosis, one on a list instead of like real people. A sense of decreased effectiveness, just feeling like no matter how hard I work I just don’t really feel like I’m doing a good job. And emotional exhaustion.

In 2019, the National Academy of Medicine published a report putting the rate of physician burnout in the U.S. between 40 and 54 percent. That’s roughly double the burnout rate among workers in other fields, even “after controlling for hours and other factors.” It’s also estimated that the rate of physician suicide is double that of the general population — between 300 and 400 doctors each year.

RIESS: In the general population, there are many attempts. But when physicians decide they have had enough, they know how to end their lives and they have what’s called a successful outcome. Of course it couldn’t be farther from the truth.

As bad as physician burnout has been in recent years, Covid made it worse. A recent Medscape survey found that two-thirds of the doctors who responded said their burnout symptoms had intensified during the pandemic; a quarter of them said they are considering early retirement, in part because their income has fallen. Nurses are also thought to have very high rates of burnout — although, frustratingly, there’s less data on nurses. And the lack of data on nurse suicide is even worse. Among doctors, burnout is known to start early. It’s estimated that 44 percent of medical students suffer from burnout before they even make it to their residency. 

RIESS: I talk to medical students and residents all the time. And they say, “When I chose this as a profession, I thought I’d be spending most of the time with patients.” But the average resident spends about 12 minutes a day with their patient. And the rest of the time is all work done through the computer.

This is a complaint we’ve heard before on this show, from Atul Gawande.

Atul GAWANDE: At this point, I’m a glorified data-entry clerk. 

And Gawande is among the most prominent physicians in America — a surgeon, public-health researcher, and best-selling author.

GAWANDE : I spend more time doing data entry in my office than I do seeing my patients. And that’s just broken.

RIESS: If people are feeling exhausted, disconnected from the reason that brought them to the profession, and they’re not feeling very effective in their jobs, their morale is going to decay, and cynicism can start to creep in.

There’s a cult novel, published in 1978, called The House of God, which is still popular among medical students. It follows a group of first-year residents at work in the hospital — the “house of God” is their name for the hospital itself. Here’s one passage: “Before the House of God, I had loved old people. Now they were no longer old people, they were gomers.” A gomer is doctor-slang for “get out of my emergency room.” The passage continues: “I did not, could not love them anymore. I struggle to rest, and cannot, and I struggle to love, and cannot, for I’m all leached out, like a man’s shirt washed too many times.”

MAZZARELLI: When I started medical school, compassion wasn’t a part of the curriculum—.

 Anthony Mazzarelli again:

MAZZARELLI: It wasn’t a title of any lecture. It wasn’t on any test.

And Stephen Trzeciak:

TRZECIAK: Classically, the teaching in medical education — and this wasn’t taught as part of the formal curriculum, this is just what you pick from your peers, there’s this thinking that, “Don’t get too close to patients.” Because that could make you prone to getting burned out.

Trzeciak says the current medical-school curricula are more likely to focus on empathy and compassion. He says there’s no established standard for this kind of training, nor is it likely to be evidence-based. But at least the arrow is moving in the right direction. Also, there’s new technology, like virtual reality, to help medical students learn to interact with patients. Here again is the psychiatrist Helen Riess: 

RIESS: There was a company that made this wrist device that helped you experience what it was like to have Parkinsonism. And when I tried it and I couldn’t even hold a pen, I realized I had no idea how hard it would even be to write anything or zip up your jacket. And it instantly gave me more empathy for people who can’t control their movements. 

Some years back, Riess co-founded a company called Empathetics. It uses live and virtual sessions to teach anyone — but mainly healthcare workers — how to be more empathetic.

RIESS: Empathy is how we perceive the emotional states of others and that gets mapped onto our brain. So, empathy is needed in order to show compassion.

So, how does this translate into advice for doctors?

RIESS: Empathy is in part a shared experience. And so, if your patient is really worried about something, and you’re sitting there flatline, you’re not catching any of the emotion. And we’re not suggesting that you get just as upset as the patient. But there should be a change in your physiology when something very emotionally charged is happening. And that’s why if doctors are looking at computer screens and not catching a facial expression, and they’re not really hearing it in the tone of voice, they can miss something that’s extremely important to the patient.

Reiss herself designed the program that Empathetics uses to teach empathy.

RIESS: On a walk in the woods one day, it kind of came to me that the word empathy could act as an acronym for all seven ways that we connect.

Okay, let’s start with the E:

RIESS: The first way that we connect with anybody is through eye contact that says, “I see you. You exist.” And it goes back as early as a mother-infant bonding, that a child knows they exist through the gaze of the mother or whoever’s holding them, and oxytocin is released when people gaze at one another and it bonds people. And in healthcare, when people feel afraid, small, and vulnerable, that gaze actually means a lot. The next letter is M, for “muscles of facial expression.” And I had to use that because there’s no F in “empathy” for the face. 

DUBNER: It’s fair, there are muscles in the face

RIESS: Well, they are what shape our expression. So, it works.

The idea here is that our facial expressions usually mimic someone else’s concern or sorrow. The “P” in Riess’s acronym is for posture, or body language. “A” is for affect; the “T” is for tone of voice; the “H” is for hearing the patient, the entire patient.

RIESS: In medicine, it’s so easy to focus on the injured body part — the pancreas that has abnormalities or the heart that’s got a murmur. But we’ve got to back up and realize that all these body parts are attached to a person, and only caring about how your wound is healing is not going to make that patient feel very cared about. Even though you’ve done a brilliant surgery.

And that leaves us with the Y:

RIESS: The “Y” is the most interesting one of all. And that is “your response.” And it’s not what you say next. “Your response” is your feeling of being with that person, because most feelings are mutual. And if you’re feeling good after an interaction, chances are the other person is, too. But if you’re feeling a little like something tilted there — we encourage taking some moments to reflect back on what just happened. And ask yourself, “Was I abrupt? Did I seem rushed? Did I cut the person off? Did I not answer their questions?” Like when things are off, we should not just move on and say, “Oh well.” Because oftentimes it’s that gap where you kind of know something wasn’t quite right.

Helen Riess’s argument is that if you want to increase compassion among doctors and other healthcare personnel, you have to start with empathy.

RIESS: Empathy’s the prerequisite.

And if the empathy doesn’t come naturally, or if it gets leached away over time, and if people have to be taught to exhibit empathy — well, that’s what needs to happen. And there’s one more reason why it needs to happen. This is the most radical argument that Stephen Trzeciak and Anthony Mazzarelli make in their book Compassionomics. All that stuff about how compassion is good for patients, both physiologically and psychologically? That’s not so radical. All the evidence that physician burnout is a huge problem? Also not so radical. Here’s their radical proposal: compassion is not a one-way street.

Its benefits accrue not only to patients, they argue, but to doctors and nurses as well. Compassion, in other words, will heal the healers. Several studies have linked compassion or empathy to lower levels of burnout. It’s really hard for studies like that to prove causation, but researchers have documented physiological benefits of dispensing compassion. Sometimes it’s called “the helper’s high,” driven perhaps by a spike in endorphins. Dispensing compassion can also activate the parasympathetic nervous system, which produces a calming effect. Compassion — the thing that doctors need to show — is the very thing that doctors need. That, at least, is the argument put forth by Trzeciak and Mazzarelli.

TRZECIAK: The preponderance of evidence shows that there is an inverse association between compassion and burnout. So, more compassion, lower burnout; lower compassion, higher burnout. Healthcare providers who have lower compassion for patients are more predisposed to getting burned out under the same amount of stress. So, we believe that having a fulfilling doctor-patient relationship, or a nurse-patient relationship, gives you that fulfilling part of medicine, and if you don’t have that, then it’s just one stress after another.

Soon after Trzeciak and Mazzarelli began to focus on the science of compassion, they started a program at Cooper Health System to mentor physicians on how to connect and communicate with patients. Mazzarelli reports that the hospital has since made improvements every year, in patient satisfaction; physician engagement; and financial performance — although Mazzarelli, true to what he’s learned during his compassionomics journey, was careful to note that, “Of course we can only report association rather than definitive causation from these data.”

DUBNER: Let me ask you, since you were the skeptic coming in, Steve, I’m really curious to know what kind of effect this work — the research and writing the book and trying to put it into practice — what kind of effect that’s had on you personally and/or professionally?

TRZECIAK: Sure. So, after going through all of the data and specifically seeing the signal that compassion can be beneficial for the giver too, that really left an indelible mark on me. Because after 20 years of working in an I.C.U. and meeting people on the worst day of their life, I came to the realization that I had every symptom of burnout. Every single one. And I assure you, that’s not a good place to be. So, having just synthesized all the evidence that compassion can be beneficial for the giver too, I decided to do an experiment on myself. And I tried very hard and I still do to this day, working to connect with people more, not less. It’s not only the patients for me. It’s their families. Many of my patients are so sick that they can’t talk. They’re on a ventilator, for example. But connecting more, not less, leaning in rather than pulling back — and for me that was when the fog of burnout began to lift. And you also realize that you can get better at compassion — it can be taught, it can be learned — and you have to be very intentional in practicing it every day.

DUBNER: Can you give me an example or two of something that you say that you wouldn’t have said? Or maybe it’s something that you say differently? Is it the way you touch someone that you might not have touched before? Is it eye contact?

TRZECIAK: Actually, it’s not something that I say. Oftentimes it’s something that I don’t say. It’s just being present. I practice critical care and there are a lot of times when the outcome is not something that can be changed. And sometimes you just need to sit with people and their suffering. “You’re not going to go through this alone.” “I am here with you.” In fact, just in the I.C.U. recently, I had to give — essentially news to a woman whose brother was fighting for his life. We were still hopeful that he could recover, but he was so severe that it was very likely that he might not. And it was devastating to her because he had been her rock throughout her whole life.

At the end of that discussion, she said, “You don’t remember me, do you?” And I said, “I’m sorry, I don’t.” And she said, “I wouldn’t think that you would. You see So, many patients here. It’s okay. But eight years ago, my mom was in that room right across the hall there and you were her doctor. And we had to have this talk and you had to tell me that she was dying and there was nothing we could do for her.” And what she remembered was the nurses and the fact that she never felt alone through that whole experience. She said the kindness of your nurses and how they helped me through that, she said, “It keeps coming back to me. It comes back to me all the time. I think about that because it was so hard at the time. But every time I think about it, I think about the kindness of those nurses.” And so, going back to what we were talking about earlier, even though there are 281 references in this book of original science research papers that show that compassion matters, even when it can’t make a difference in the outcome, it still makes a difference.

That, again, was a replay of Episode No. 444, “How Do You Cure a Compassion Crisis? You can get the entire archive of Freakonomics Radio on any podcast app.

*      *      *

Freakonomics Radio is produced by Stitcher and Renbud Radio. This episode was produced by Morgan Levey. Our staff also includes Alison CraiglowGreg Rippin, Jasmin Klinger, Eleanor Osborne, Mary Diduch, Zack Lapinski, Ryan Kelley, Emma Tyrrell, Lyric Bowditch, and Jacob Clemente. Our theme song is “Mr. Fortune,” by the Hitchhikers; the rest of the music was composed by Luis Guerra. You can subscribe to Freakonomics Radio on Apple PodcastsStitcher, or wherever you get your podcasts.

Read full Transcript


  • Steve Trzeciak, chairman and chief of the department of medicine at Cooper University Health Care.
  • Anthony Mazzarelli, co-president and C.E.O. of Cooper University Health Care.
  • Helen Riess, director of the Empathy and Relational Science Program.



Episode Video