Trust in Medicine: Built By Humans, Lost at Scale

Recorded March 28, 2026. This transcript has been edited for clarity.

Robert A. Harrington, MD: Hi. This is Bob Harrington, from Weill Cornell Medicine in New York City. I’m here at ACC 26 in New Orleans. Wow, what a buzz this meeting already has. I just came out of the late-breaking clinical trial session, but we’re not going to talk about late-breaking clinical trials right now.

I always love to use these meetings as an opportunity to catch up with friends and colleagues, to pick their brains on some of the topics that I know they’ve been thinking about, and in this particular guest’s case, has been writing about. The topic for today is going to be trust in medicine, trust in healthcare, and maybe broadly, trust in society. And I have a great guest to join me here today.

I’m with Dr Lisa Rosenbaum. Lisa is a cardiologist at Beth Israel Lahey in Boston. She is the writer in residence for the Smith Center, an assistant professor of medicine at Harvard Medical School, and she is a national correspondent for The New England Journal of Medicine. That’s a lot. First, welcome.

Lisa Rosenbaum, MD: Thank you.

Harrington: Thanks for doing this. As I was getting ready to tape this, I was checking out Medscape AI and I asked them about you. That bio I just gave of you, that’s where I got it.

Rosenbaum: It’s totally correct.

Harrington: We’re going to move into a serious topic. There are some fun things I suspect we’ll get to, but it’s a really serious topic that I know you’ve been thinking about. It’s Trust in Medicine. I’ve been a doctor a long time. It seems as though trust is declining in us as a profession. Many times when you ask about a profession, if you then drill down and ask about an individual, it’s different.

If you ask, do you trust medicine? Do you trust science? The answer is now frequently no. If you ask, do you trust your doctor? The answer is most frequently yes. Let’s start with the societal ask.

‘I Forgot’ and Medication Nonadherence

Rosenbaum: This is a topic obviously, you know, that I’ve thought about often. Even though we’re talking about it right now in the context of declining trust, not just in medicine, but in other institutions like media and government, I actually came to it when I was still in training because I was doing some work about why patients don’t take their medications after they have MIs (myocardial infarction). I was working with a group of behavioral scientists who were doing an RCT trying to get patients to take their medicines by nudging them, essentially.

Harrington: It’s a really popular way, particularly with devices, etc, to actually study behaviors.

Rosenbaum: In fact, my ostensible role was to make sure people could open the pill bottle that had a glow cap to tell us whether or not they took the pill. I took the opportunity to talk to people who’d had an MI, which was a luxury. You would think as a doctor that’s something that we actually get to do, but it turns out not so much. There were 30 people who had had an MI. We know that half of patients stopped taking their medicines within a year. People have really complex feelings about taking medications, and that was really eye-opening for me.

That was around 2012. One thing you would hear all the time is that people will just say, “Well, I forgot to take my meds.” I’ll never forget, there was a professor at Brown and I was talking to him about nonadherence. He said to me, “Would you ever forget to pick up your child from school?” I was like, “I don’t have a child, but no, if I did.” He was like, “Well, the way we talk about forgetting, we say it because it’s a way to frame something else that we don’t really understand.”

I think something very similar is going on with trust. It’s become this blanket term of something where people have some friction with us as an institution of medicine. Clearly, that friction exists all across society, but I think it’s like this really complex set of behaviors. It came to the fore during COVID, of course, because people didn’t want to take vaccines or they didn’t want to wear masks, or they didn’t want to isolate. We were all affected by other people’s behaviors.

When I go back into my own career and my own beginning to become interested in this, I realized this isn’t really new.

Harrington: This has been going on a while. Let’s unpack that. Let’s go back to this notion of, “I forgot to take my medicines half the time.”

As you peeled it back, I think about saying I forgot, but also as a patient, I have agency when I choose not to do something. That’s really important to patients.

Rosenbaum: It is so important. I’m so glad you said that. I think if there was one singular finding in that work, it was that desire for control. The irony is that we’ve done incredible work as a profession in cardiology to educate people about how empowered they are to prevent cardiovascular disease. The unfortunate corollary of that is that when they get it, they feel like they’re at fault. Taking a medicine feels like a cop out.

Harrington: Oh, interesting.

Rosenbaum: Yes. If they say, “I did it to myself, now I need to undo it,” I think it’s a way to take back control. That was one of the interesting findings.

Harrington: Is there any common demographic or social or cultural link about people who stop taking medicines after a heart attack and those who continue to take it?

Rosenbaum: First of all, just to be clear, I interviewed 30 people. I’m not in your realm of mega trials.

Harrington: Usually, if it’s less than 1000 participants, I don’t consider it a real trial.

Rosenbaum: It wasn’t a real trial. It was me being a journalist pretending to be a researcher.

Harrington: I think of it as qualitative research.

Rosenbaum: It was qualitative research. Not that I could find a common demographic, but I will say the backdrop to all of this was this idea that people aren’t taking them because they’re too expensive. Then there was a trial published in The New England Journal of Medicine called MI FREEE, where they essentially gave everybody their medications for free and there were no different outcomes in terms of adherence.

I think that really told us that the things that are affecting people’s choices are maybe not what we assume. Fast forward to now, however many years, close to a decade, we’re seeing in many probably well-educated, wealthier demographics, people who have more access to knowledge and more access to how science works, they often use those tools to undermine science in some way.

Information Overload and the Death of Expertise

Harrington: Let’s talk about this. The question about more access to knowledge. I would actually say they have more access to information, but it’s not necessarily knowledge.

Rosenbaum: Such a good point.

Harrington: My sense is that, as people have immediate access to things — I mean, you and I could sit here on our phones and look up many different things — there’s this notion, and maybe they’re tied together, that we don’t need expertise because everyone is an expert. My grandmother’s an expert.

Everybody is an expert, and it really has diminished the authority of the medical profession to make recommendations that people say, “Okay, I’m going to do that.”

Rosenbaum: Right. I absolutely agree with that. Although there’s this extreme range of granularity. I don’t know what it’s like for you.

Harrington: I’m a consulting cardiologist.

Rosenbaum: By definition, when a team calls you, I suspect they want your expertise, right?

Harrington: They do.

Rosenbaum: That’s different than a patient. It’s a team. As a cardiology consultant, too, I find when I walk in the room, the patient or the family says, “We’ve been waiting all day to hear what you have to say.”

The first bucket, I think, is that when people are sick, they want our expertise. I have yet to have an encounter with a patient who says, “Well, AI says this, and you’re saying something different, so you must be wrong.”

They’ll often say something like, “I’m really good at my own research and reading the literature. I saw all of these things. What do you think?” I have yet to experience that very overt, you’re saying something different and I don’t trust you.

Harrington: As you know, I practiced in Silicon Valley for a long time before moving to New York. One of the things I loved about Silicon Valley was that people showed up all the time with, at the time, printouts — now they just show me their phone — to say, hey, I read this article, or I Googled this and this came up.

Particularly, in our world, statins remain an area where people say, “Hey, I get it. My cholesterol’s high. I’m at risk for myocardial infarction or heart attacks, but I’m not taking statins. I read too many negative things about it.” I can pull out the hundreds of thousands of patients in the meta-analyses that show that these are incredibly safe drugs, almost 40 years of experience with them, and it doesn’t do the trick.

Rosenbaum: Right. I guess that’s part of what I’m trying to get at. Maybe you get called because a patient’s having a Type 2 NSTEMI, right? On the oncology service, that’s a very different question than I’m going to put this thing into my body every single day that I read has a small chance of helping me and might hurt me. I think statins sit at the convergence of everything.

Harrington: They really do. It’s amazing to me that after all these years, all the trials, and all the data, that they still sit at that nexus.

Rosenbaum:  That’s partly why they’re so interesting, right? They’re so good. I think they should be in the water, I really do. They hit the self-optimization, biohacking, life hacking, all of those trends. Also these people who pay attention, and by the way, there are many people in medicine who probably would agree with this idea that the benefit of primary prevention is small.

Harrington: For an individual, it is. For the population, it’s pretty big.

Rosenbaum: Let’s talk about that then. The question about the individual vs the population, I think is also at the heart of some of what we’re talking about. We’re saying to them, “It might not help you, but it will help 10,000 of you.”

You and I are trained to think like that, but most people aren’t.

Harrington: Most people think, “What does it do for me?”

Rosenbaum: Exactly. That’s very human.

When we say, well, is there distrust or are people more empowered to express these human things that they’ve always felt?

Harrington: Is it with more ready access to data and information from a variety of sources, is that just giving people the tools to ask more questions? Trust is a component of it, but lack of trust or losing trust is not such a new thing. It’s that people have more tools, to go back to my original phrase, to give them agency.

Rosenbaum: Exactly. It’s like they have a new language. We democratized information. Everyone’s on the internet.

Harrington: Which is great.

Rosenbaum: Mostly. It’s great, depending on where you sit in all of this. I still think it’s great because, philosophically, I see my role as a physician and as a writer, as I want to help people live their best lives.

Harrington: It’s the AHA motto, right? We want people to live longer, healthier lives.

Rosenbaum: Part of that, though, if you’re putting a statin into your body every day, and something about that makes you feel bad. I don’t mean that it makes you feel physically bad because of a muscle problem or anything, but something psychologically bad. Well, that counts for something, too. To get back to this question of trust, I do think there are some clear changes that we can’t ignore, like social media, AI, pandemic, public health, whatever happened.

Harrington: Public figures who are saying, you shouldn’t trust us. People who have vilified organizations that promote healthy living.

There’s a group out there promoting that those people aren’t right. Those people are bad. They’re giving you bad information.

Rosenbaum: That factor, I can’t quantify it, but it’s undoubtedly contributing to all of this. I think it’s also part, again, of our tribal political milieu that everything has become sport, in a sense, but not in the fun way. In the tribal way. Science has now fallen into that trap. Again, I think that much of what we’re seeing is that people, like you said, have agency. They’ve found some agency and tools to express things that they might have been feeling for some time.

Harrington: When you think about trust, and we talked about the trust of the profession and there’s trust of the individual, there does seem to be more trust in the individual, particularly if you have a relationship.

If you make rounds in the hospital, you haven’t quite established enough of a relationship to have total trust, but if somebody’s been seeing you in the office for, let’s call it 10 years, there tends to be more of a personal relationship and they trust you.

Primary Care: The Key to Rebuilding Trust

Rosenbaum: If there’s one thing we could actually do, if we agreed upon a definition of trust and said we need to do something, it would be to give everybody a primary care doctor and let them have time to have a relationship with that doctor. So I agree.

Harrington: I’ve spent time reading your primary care series in The New England Journal of Medicine. It’s a fabulous series on primary care.

Rosenbaum: Thank you.

Harrington: It’s something that, now running a medical school, I think about all the time. Not to diverge too much, but if you ask young people when they show up in medical school their career options, I almost guarantee that primary care will be on the list. When they leave medical school, it’s very infrequently on the list, particularly at a place that stresses research, etc.

Rosenbaum: Right.

Harrington: We have to really rethink this. I’ve been following your series and trying to ask my education team, how do we rethink this? It should be a much more interesting field and a better compensated field than it currently is.

Rosenbaum: I totally agree. I’d love to talk to your education team. You could start where you are because you can do whatever you want because you’re dean.

Harrington: The thousands of faculty wouldn’t say the same.

Rosenbaum: Let’s just pretend. What if you created a curriculum that others could follow, but where we thought about primary care differently and where we began at a very different place and we brought back that notion of the consummate doctor.

Harrington: Of trust.

Rosenbaum: Yes. Really think about all the forces that they’re facing in terms of AI, isolation, and a general sense that people are spending their whole lives on the internet, lost and nihilistic, really. We could then say, well, there’s this role that you could do where you don’t just get to take care of people’s diseases and their chronic diseases. You get to really do something very tangible to address this sense of pessimism that is plaguing all of your peers. That would be a very different pitch.

Harrington: It makes me think about this notion of the concierge physician, which has become wildly popular. I think it’s in part because there is the relationship trust piece that now can be built because you’re taking care of 200 people, not 2500, which is the typical panel size for a primary care doctor. I think it’s allowing both sides, patient and physician, to recapture some of the magic of the relationship.

Rosenbaum: I absolutely think that’s why people are doing it. I wrote about concierge medicine, too. When you talk to the physicians, they’re so happy.

They’re so happy. But we have to acknowledge it’s bad for inequities.

Harrington: I said the same thing at my place. I said, I don’t mind having concierge medicine as long as you can assure me that we haven’t abandoned health equity and that the poorest person, the most vulnerable person in our system is going to get great care. And if you can assure me that along that spectrum...but it’s tough.

Rosenbaum: It’s really tough. The only way I could think about approaching it in a way that acknowledged the happiness. It’s funny because you began with agency and I think physicians having agency is important. We’re all human, we’re all responding to the same need.

When I wrote the piece about concierge medicine, the question I faced was, okay, the inequities are obvious. It’s like a math problem.

Harrington: You don’t need thousands of subjects to work this one out.

Rosenbaum: No. What can we learn? What are they reacting to within traditional medicine and what are they bringing to their practices, with the broader question being: Can we bring some of that back into traditional medicine?

Again, when you think about designing a curriculum, I think that would be a really interesting piece. Let them go and shadow concierge practices. What are they doing? What are the elements of care? What’s been really heartwarming to me in putting together this podcast, which has been about so many things in primary care. When you talk to people, Federally Qualified Health Centers (FQHCs) for instance, or many docs in primary care who have really come out in spades to remind me that there are actually many happy primary care doctors.

They find agency in their own ways. It’s quiet, and there’s this interesting tension in how do we recognize that they’re happy and they’re doing this very meaningful thing quietly without destroying that quiet joy by talking about it. Do you know what I mean?

Harrington: Well, the FQHC thing is a really important one. You talk to those docs and they’re really committed. They really believe that they’re doing important work and really care deeply about the social mission.

They are happy. They’re working their butts off. They’re frustrated by inequities in the system. They’re frustrated by many of the federal funding issues, the state funding issues for Medicaid, etc, but they do find joy in practice.

Rosenbaum: It would be really cool to have students see all of it. Again, I think trust is lost at scale, but it is built one person at a time. I think it can only be built in the context of relationships. I don’t think it exists as this abstract concept. It’s the same as the forgetting thing. It’s like, okay, it’s something I can vaguely understand, but what does it really mean?

Harrington: We can’t have a conversation about trust without having mentioned the pandemic.

Boy, we lost a large amount of trust with the public. Here’s one of the things that would always surprise me, including with my very well-educated friends. We’d be on the phone regularly. I’m catching people up, saying, this is what we’re learning. They said, but last week you told me something different. For me it’s like, well, that’s just science. In the last week, these two papers came out and now we’re going to do it this way. They say, yeah, but last week you told me something else.

Was that part of it, that you and I and our listeners understand the scientific process? You do an experiment, you learn. You go forward, you adapt, you test it again, and you learn something different.

Rosenbaum: I don’t know. The reason I don’t know is because I feel like people make up their mind about something and then they find a way to reason toward it.

Harrington: Which is the opposite of the scientific process.

Rosenbaum: Right. We would hear things like, “You didn’t admit uncertainty.” Well, some people will argue that it’s not good to say you’re uncertain about something. How much of this is just that people have a gut feeling and then they find ways to poke holes in what we’re doing? I also want to be a little bit soft on how I feel about this because I don’t think the answer is to stop changing our minds when we have new data.

Harrington: We need to keep doing science.

Rosenbaum: Exactly. The one thing we can do is keep doing science. There are so many forces shaping people’s impressions about things that I just don’t know what really fueled the disbelief.

Reuniting Public Health and Medical Care

Harrington: Well, the other thing for me with the trust has been this long time, what I’ll call, “problem” of the separation of public health from medical care.

If you go back and you read the original documents around Flexner and the adoption of the scientific method, and then you read the formation of public health several decades later, it’s really in a reaction to some of the medical care issues that they’re trying to address. Boy, there’s a separation there that’s also made it difficult.

Rosenbaum: That is so true and unexplored in our discussion about this. It’s something that I’ve been thinking about often because I think when we talk about trust, so often it’s about vaccines. Vaccines are obviously important, but they’re such a small part of this. Again, to go back to this idea about cardiovascular disease and people not taking their medications and that probably affects more people. It has become this problem that’s centered in public health, and public health has a very different set of tools than we have, and they don’t get to have the relationships.

Harrington: It’s not a one-on-one thing.

Rosenbaum: I think that’s a very interesting observation and one that I want to explore more.

Harrington: if you talk to colleagues at medical schools that are also part of a university with a public health school, often the two shall never meet.

Rosenbaum: There’s a public health school at Harvard. I have some friends there, but it’s not like we’re in dialogue.

Harrington: It’s not like our students are constantly crossing the boundaries, etc. It’s always been interesting to me as to why can’t we get those closer together? After the pandemic, it would seem that we have to get these closer together because one of the real failings of the pandemic was the failing of the public health system.

Rosenbaum: I don’t think that’s going to happen imminently.

Harrington: Believe me, I have many things to tackle.

Rosenbaum: To get back to your role and where you and I both sit in the house of medicine, I think there is tremendous opportunity, especially right now, to bring back relationships.

Harrington: To foster trust with the building of relationships.

Rosenbaum: Exactly. I think people want it. The students want it, and we need to show them what that looks like. The saddest thing would be if we become so fragmented and also disillusioned ourselves that we don’t show them anymore.

Harrington: That’s a great comment because I think about the training of young doctors — and this is a key component: How do we teach them in a world where information is immediately available. We had the AI talk a few minutes ago. They’ve got all that information, so what makes what they do special?

I refer to my colleague Abraham Verghese, that we’re in that long tradition of learned intermediaries. Part of what we do is we build relationships — he likes to call it being present — and that we’re present. People know we’re there for them. We have information and we should be learned interpreters of that information, but we can only do that if people trust us.

Rosenbaum: Not only do I completely agree with that, but I think also if we are doing our jobs well, I mean, we all want to feel special. There is always going to be a need for that human-to-human connection. Algorithms are great to a point, but there’s always that soft touch of expertise, which is about judgment. It’s about knowing somebody. It’s actually an extreme form of intelligence that I think we need to really focus on as well.

Harrington: It’s funny you say that. I was meeting with one of our deans of artificial intelligence. He’s a computer scientist and informatician, and he said to me, well, I was writing some code. And I said, “Well, wait a minute. I thought we didn’t have to write code anymore, that the machine writes it for us.” He smiled and he said just what you said. He said, “That’s fine for a couple of lines of code, but as it gets more complicated, there’s still a need for expertise.”

The way that he, as a computer scientist, asks the machine to write something is different than the way I might ask. Then we extended it into medical care. Yeah, you can ask the algorithm, how do you take care of any disease, but there’s still a need for expertise.

I go back to the early writings of Dave Sackett with evidence-based medicine that gets a large amount of criticism for cookbook, etc, but that’s not what he said. He said that evidence-based medicine is using the best quantitative information available, coupled with expert wisdom and judgment, and put into the context of that patient’s values and preferences. That, to me, sums it up. That’s what we’re trying to do.

Rosenbaum: Maybe AI will force us to be more clear about what we mean by expertise. We all want to feel needed in this world, and we are going to be pushed to recognize that layer where we come in and only we can come in.

Harrington: You see it when a doctor asks a question of an algorithm and when a lay person asks. A lay person can go down a weird path because maybe they didn’t get a word quite right or they brought in a mistake. When you ask about something in cardiology, you’re asking it from the position of a cardiologist. You get different answers.

Rosenbaum: You get very different answers. The algorithms give me a headache.

Harrington: That’s a good place to stop. We’re going to end it on that. I want to thank you for joining me.

Rosenbaum: Thank you so much. It’s so fun. It’s my favorite stuff to talk about.

Harrington: My guest has been my friend Lisa Rosenbaum, who is a cardiologist. She is the national correspondent for The New England Journal of Medicine. She’s at the Smith Center, at the Beth Israel Deaconess. She does a lot, and I’m really happy that she was here to talk to us about this subject of trust in medicine, trust with doctors, and trust in public health. I suspect this is a show and a topic that we’ll continue to come back to in the year ahead. Thanks for joining us.

Rosenbaum: Thank you.

This article is part of an ongoing, year-long Medscape series examining the decline of public trust in medicine. More content is  available here.

Robert A. Harrington, MD, is the Stephen and Suzanne Weiss Dean of Weill Cornell Medicine and provost for medical affairs of Cornell University, as well as a former president of the American Heart Association. He cares deeply about the generation of evidence to guide clinical practice. When not focusing on medicine, Harrington dreams of being a radio commentator for the Boston Red Sox.

© 2026  WebMD, LLC

Any views expressed above are the author's own and do not necessarily reflect the views of WebMD or Medscape.

Cite this: Trust in Medicine: Built By Humans, Lost at Scale - Medscape - Apr 27, 2026.