Patients who are treated with empathy by their doctors and other clinicians have better health care outcomes, according to a new study from Wharton health care management professors Ingrid Nembhard and Guy David. The paper is titled “A Systematic Review of Empathy in Health Care,” and it appears in the journal Health Services Research. The co-authors are Wharton undergraduate student Iman Ezzeddine, and David Betts and Jennifer Radin, both of Deloitte Consulting.

Nembhard spoke to Knowledge at Wharton about the study, which concludes with a call for more organizational-level interventions to ensure empathy for all patients, systematically.


How Do We Define Empathy in Health Care?

Angie Basiouny: Why is empathy in health care important? If my doctor is board certified and graduated at the top of her class, I know I’m going to get good care. What difference does it make if she doesn’t make me feel warm and fuzzy?

Ingrid Nembhard: You may get great clinical care, but that’s different from empathy. We assume that everyone is clear on what empathy is and what its role is in health care. Broadly speaking, empathy refers to understanding another person’s feelings and their thoughts, and feeling those congruent thoughts and states. In health care, empathy is defined as “understanding and feeling a patient’s emotions and perspective.” It’s also offering a response — for example, how you communicate with that patient — that reflects understanding and that actually aims to help them.

Why does empathy matter in health care? Well, when there’s insufficient empathy, there’s diminished understanding of the patient’s perspective. On the other hand, when there is higher empathy, there is understanding. In principle, that understanding matters because it cultivates efforts to better meet patient needs through both interpersonal choices, such as speaking with care, and operational choices, such as connecting patients with resources, whether they need mental health providers or transportation — things that can facilitate their care experience and their actual care and health.

We know now from looking at the research, and even if you thought about this theoretically, that the more that understanding is present, the more patient-centered care plans are likely to be made, the better the patient care experience will be, and the greater patient adherence to their plans will be. That all means that we can expect better patient, worker, and organizational outcomes, whether you think about clinical outcomes, or you think about worker job satisfaction, because they’re getting the information they need to be able to treat patients the way they should. And they’re getting better patient experience ratings.

So, why it matters is that it’s the beginning of a positive cascade, I think, for patients and health care, and even for workers.

Taking Stock of Empathy in Health Care

Basiouny: In the paper, you mention that there’s a great deal of disjointed information about empathy in health care, but that it’s emerging as its own research field. What were you and your co-authors hoping to contribute to the literature?

Nembhard: I do a lot of research on patient care experience in addition to understanding the organizational side of care. The data at this point is pretty robust that a lot of patients are having poor patient care experiences, and there has been this growing attention now to the relational side of health care. That has really led to a lot of investigation about what facilitates and what hinders empathy. What outcomes come from empathy? How best can we measure empathy? Who is likely to give you empathy, and who is unlikely to give empathy? Who is likely to get empathy? Who is unlikely to get empathy? And then how do you go about increasing it?

These kinds of investigations have been occurring for the last 50 years. We see that largely they’ve been occurring by individual researchers in independent investigations that have been published in a wide variety of journals. That means that we can now say that there is an actual research field of empathy, because there has been a lot of attention in this field, but it’s all disparate. You haven’t had yet the research that integrates all of that knowledge.

My colleagues and I thought we’ve reached a time now where we need to pause and take stock of the field. We need to see what lessons can be extracted from the 50 years of research, and we need to see if we can create clarity about the way empathy is operating in our health care system.

We can then start to identify where we need to do more research and where practice needs to change in order to achieve those goals. We decided to do a systematic review of 50 years’ worth of empirical, quantitative research on empathy. And our research covered 450 articles that met our criteria.

What Factors Predict Empathy in Health Care?

Basiouny: Let’s talk about the takeaways. You found that more empathy ends with better health care outcomes, and that five factors predict empathy. Can you take us through those factors?

Nembhard: Sure. The first is that provider demographics seem to matter. Those are things like the number of years a professional has been in their specialty. We also see that it varies by characteristics like gender and the specialty. Perhaps not surprisingly, studies suggest that primary care physicians and those in behavioral health tend to display more empathy than colleagues who are more on the surgical side or who have acute experiences with patients.

Other characteristics of Who is providing empathy can matter too. In that bucket, we find things like personality, whether somebody is an introvert or extrovert, their knowledge, their attitudes towards different people and the like.

The third category that we identified is how providers behave during their interactions. We pay attention to the fact that people talk in different ways and speak to people in different ways, and that certainly appears in the data. Providers vary in the way that they communicate, the tone they use, the words they use. They also vary in aspects like their body movement in the interaction. Are they closer to you or farther away? Do they create distance or not have distance? And whether they give adequate consultation.

The fourth bucket of things that we find are target characteristics. We’re referring largely to patients as the target of that empathetic interaction. It varies by the type of condition the person has or the disease that they’re battling. Some of the data would suggest that certain conditions are more likely to elicit an empathetic response than other conditions. Someone’s socioeconomic status — whether they have more income or less income — tends to influence the level of empathy that’s directed towards them.

The fifth category that we found in the literature is organizational context. Things that are organizational include how long is the visit that the patient has with their provider? In shorter visits, there’s less empathy typically found. The waiting time also is tied to perception of how much empathy there is.

The five categories are really interesting. I’m simply giving you the high-level [view]. Within each one of those categories, our research showed there are multiple factors.

What Interventions Increase Empathy in Health Care?

Basiouny: You also looked at some interventions that can increase empathy among health care givers. What are those interventions?

Nembhard: Once the field appreciated that empathy might matter, it started to think about ways to increase it. Most of those have been individual-level educational interventions, so things like training participants how to do a particular skill, like how to communicate well — in an empathetic way. We see some studies that focus on having a course, so a person goes through a series of lectures about how to be empathetic or what empathy behavior entails. Sometimes there are workshops where you’ll role play and get feedback on how you behave. Simulations, visuals, videos. The category that probably most caught my attention is treatment for empathy. There are studies that trial transcranial direct current stimulation — actually stimulating that part of the brain [associated] with empathy. There are a lot of options that are on the table for improving and increasing empathy, most of them educational interventions.

For us looking at the data, it was surprising that there were no studies of organizational interventions, because one of the factors that we found that was significant was organizational context. We know that organizations can matter. In some sense, the absence of organizational interventions may reflect the fact that we think of empathy as a human trait, so why make it part of the organization? You don’t need to be trained in empathy.

But if the provision of empathy benefits from having dedicated time and people and processes and leadership, then it totally makes sense that we need to direct greater attention to organizational interventions for improving empathy. My co-authors and I are now very much of the mindset that we need to have more empathetic systems and institutions that are structured in such a way that they create conditions for anybody to receive empathy, in a non-arbitrary way throughout their whole service of care.

We’ve seen that organizational interventions can work. We see it around patient safety. It used to be that you thought safety was the type of thing that a provider delivered to a patient. Innovation was the type of thing that an entrepreneur delivered. Yet now when we look at health care, it’s not unusual to see a chief patient safety officer or a chief innovation officer, or roles that are dedicated exclusively to ensuring those goals. Organizations are taking that route, rather than just training clinicians.

There are now role-based approaches centered on non-clinicians to deliver what is needed. We might want to move in that direction [for empathy too]. I think my colleagues and I would be excited to see more interventions that say, “OK, this is something that organizations need to be attentive to.”

How Empathy in Health Care Can Reduce Disparities

Basiouny: I want to ask you about two demographics, which are Black patients and Hispanic/Latino patients. We know those two groups have worse health care outcomes across a number of measures, whether it’s COVID-19, heart disease, or maternal mortality. Would greater empathy for those patients translate into better outcomes?

Nembhard: The simple answer is that it should. If you recall my earlier response to your first question about why does it matter, it’s largely about understanding people, their emotions, their needs, and where they are in their care in their state. If we were to have greater empathy, we would expect that there would be greater understanding such that the choices that are made and the conversations that happen in the course of care would be more attentive to the needs of the person. That means that they would get the communication they need, and it would be culturally competent.

If you need transportation, we would provide transportation, because we would understand the circumstances. We would understand and therefore make choices and make care plans that would allow people to be successful in their health care. I do think empathy is part of the process. If we want to reduce some of those disparities, we need to be better about understanding where people are.

Basiouny: This study is the first of its kind. What do you want to look at next?

Nembhard: The main motivation in taking on the last 50 years of research was that we wanted to be better prepared to do work that could make a difference. We’d love to be able to collaborate with a health system interested in trying and trialing a role-based organizational intervention. We think it’s time.

We’ve been doing the training of individuals for years, and we’re still not at the level we need to be. That appears to be insufficient to allow systematic empathetic health care. So, we would love to be able to study role-based organizational intervention. We’d love to see the field take off and other people think about other organizational interventions that might be used to build empathy.