In a recent episode of Radio Advisory, host Rachel (Rae) Advisory sat down with Aaron Carroll, a pediatric physician and the chief health officer for Indiana University. Carroll is also a popular New York Times contributor, the editor-in-chief of the health policy blog the Incidental Economist, and the host of the YouTube channel "Healthcare Triage," as well as the author or co-author of several books, including "The Bad Food Bible."

This transcript has been lightly edited for length and clarity.

Cheat sheets: Evidence-based medicine 101

Question: Medical myths were obviously a problem long before Covid-19, but I'm curious, what feels different about this moment?

Aaron Carroll: I think the stakes, to be very honest with you, are the biggest difference. People who are buying into misinformation or myths at the moment are not just putting others at risk; they're putting their own lives at risk.

And most of the time when I'm talking about medical myths, it's small-ball. It's things which might make a slight difference at the edge, or might make a tiny quality of life difference. Or even, if we're talking nutrition, it might make a broad years- or decades-long difference. But right now, buying into the wrong stuff could have an impact on mortality, and like, in the very short term. It's a whole different game.

Resource library: How health care organizations can navigate the post-pandemic world

Q: Is there a moment where you started noticing more of this misinformation creeping in, either in your own practice right as a physician or in your broader career? Is there a moment where you went, "Hmm, I'm getting a lot more questions from people that just don't make sense or completely rooted in misinformation or maybe even disinformation?"

Carroll: I think things felt like they got tense when the country started the lockdown last year. Up until that point in January and February, the pandemic was something "over there." It wasn't even affecting us. And in March it was still hard to raise alarm bells; it wasn't a big deal. But by the time we got to April, when it felt like a lot of the country was locking down and people were taking it seriously, that's when I think we just starting to see pushback, because people's lives and livelihood were being really affected.

And if you weren't in the health care system, you did not see Covid-19 every day back then, so it was hard to understand why we were doing that. And I feel that's when things started to pick up. It's when you just started seeing protests around lockdowns or protests around masking. That was when it felt like things were getting worse.

Q: I also feel like in the world of myths and misinformation, there's just some particular vulnerability in the medical and health care space, because it is so complex and misunderstood even by the folks within it, that it just becomes really hard to battle.

Carroll: People in general have no appreciation of how much uncertainty there is in medicine. One of the dirty secrets that we don't tell anyone is how much we're just making it up.

The number of things for which there's rock-solid, randomized controlled trial evidence is really small. A lot of the times we're going with best guess, best practice. And we sometimes get it wrong, but we speak with the same level of authority no matter what that level of evidence is. And so this felt like a time when a lot of people were all of a sudden exposed to how much uncertainty we often have to deal with in medicine, but it was playing out right before their eyes, and people freaked out.

Q: Yeah, they didn't know how to deal with it. In their mind it's, "You're changing the goal posts on me. See, you don't know what you're talking about, so why should I trust you this time?" But they're not realizing this is inherent to the way that we study and ultimately practice medicine.

Carroll: I remember being on a podcast in, I want to say April or May, I can't remember exactly, but it was about masks. And the host was like, "How can you live with this level of uncertainty?" And I'm like, "This is every day."

I'm totally comfortable with this. I'm always playing small odds in one way or the other and understanding that even the best treatments have a number needed to treat of like one in 100, one in 1,000. Everything is incremental. And there's often a fair amount of uncertainty.

So at the beginning, when we were talking about masks, it was focused on masks to protect you, meaning N95s, that were in short supply. And we needed to hoard them for those in the health care system who were at highest risk. And so I remember even saying or tweeting one point, like, if you're wearing a mask like an N95 at this point, you're wasting a mask.

And then, months later, it was like, "OK, no, no, no, no. Now we know the coronavirus is airborne. Everyone should wear a mask."

And people are like, "Well, you said…" And I'm like, well, different masks in a different circumstance. We're learning as we go.

Q: You've pointed out before that one of the biggest problems is that the very people with the least understanding of science tend to be the ones that oppose it the most. That's why large-scale efforts to educate the public tend to fail. We know that bombarding people with facts, figures, data is not going to be that effective. Do we have any understanding of what does work to nudge behavior?

Carroll: I mean, there are some. Unfortunately it's hard. Obviously, if messages come from trusted voices in the community, they work better—but that's often hard to do because the same people want to be the answer for everything, and that doesn't work.

I also think, and this is more personal, it requires time and effort to truly get to understand where the lesion is. Where's the misinformation or misunderstanding come from? What's the concern? How do I address it?

But my biggest gripe is that the answers are often complicated. When someone wants to say, "Do masks work?", I'm like, that's going to take 10 minutes for me to answer. I cannot put it in a soundbite. And most people, unfortunately, consume their news from cable news, where if you're lucky, you get to say three sentences. And then someone else is going to say three contradictory sentences, and then they'll go, "People disagree," when really it's a nuanced, long answer required.

And there just isn't a lot of space for that in today's media—with the exception being podcasts, which is why it's one of the few things I'll say yes to, because there's an actual chance to have the long-form discussion where you might actually get into some of the nuances of the answer, as opposed to a quick hit on a panel.

Q: Exactly. And that's one of the reasons why I think the physician-to-patient or clinician-to-patient relationship is so important, because we see that, generally speaking, people still trust their doctor. And in ideal circumstances, which of course aren't always there, there is a moment for trust-building in the physician office or through telehealth or in a portal message. What is your advice for how clinicians can in the moment try to get these messages to stick?

Carroll: Well, again, I think it's important to try to figure out where the problem is. Some of it is just misinformation where there's no negative intent. Some people think it costs a lot of money still to get vaccinated—it's free, but they just think it isn't. And so just making sure they understand that—for some people, it's literally a logistical barrier. It takes activation, energy, and time that they don't feel they have. If we can just figure a way to get the vaccine to them, they might get it.

Some of it is mistrust in the health care system, which has to be combated with long-term building of trust. And some of it is that they've just literally heard lies, and those have to be carefully and thoughtfully countered in a respectful and compassionate manner.

You're right, though, that this is something physicians should be able to do, because they should have that kind of relationship where they can probe and get the answers they want.

Of course, office visits get shrunk and shrunk in terms of time. That's the problem.

Q: Yeah, that's a problem that I'm hearing. It's not that clinicians don't think that's their role or that they don't want to do it. It's that they're saying, "Hold on, I'm this overworked, I'm this understaffed. You've pushed me to be transactional in all of these different ways through telehealth. I don't have time to build a trusted relationship, let alone spend time unpacking this information in the moment." What advice do you have for that pushback?

Carroll: Make the time. I know that that's a flip answer, but we're the last line of defense here. Look, I'm a pediatrician. So for a long time, it's been difficult convincing some parents to vaccinate their children. This is not new, certainly for pediatricians who have dealt with myths and misinformation about vaccines for decades.

So this is part of what we're going to need to do for Covid as well. I don't think we've relied as much on the health care system to distribute Covid-19 vaccines, to be very honest. They're not often delivered in the doctor's office the way most vaccines are. And so it's very different. And we've perhaps missed that opportunity where, if we were making this part of the regular doctor visit, maybe we could get a few more people or at least a decent number of people vaccinated.

Q: And maybe don't assume that you're going to change somebody's mind all at one time. I read this wonderful piece that was talking about a patient with HIV, who related very strongly to his physician who was a fellow black homosexual male. And he spent the better part of a year at the end of each visit when they were doing regular checkup to say, "What about vaccination? What about Covid-19?" And it took time. But the moment that the patient said, "Doc, I did it," made it all worth it for this physician.

Carroll: I would agree. And I think that physician should be used to that. If you've ever tried to counsel someone on diet and exercise, it doesn't happen in the one visit. I mean, the way I've been talking, I've been assuming an established relationship that now, if you already have the level of trust, you can build. But if you're seeing a new patient for the first time, of course, very little is going to be successful in that first contact. It's just the beginning. And we just have to take the long road on this. It's a marathon, not a sprint.

Q: You brought up diet and exercise. I wonder if that means there's actually something that we can learn from old-school patient activation here. In population health, we assess patient activation because we want to know, should we intervene? And if so, when? Which of course means sometimes we don't. Do you think that there's some application here of choosing when, how, when to ignore medical misinformation, even when it shows up at our practice?

Carroll: Well, I mean, I will gauge sometimes—and granted, not as much in clinical practice, but more when I'm doing more public health—I will gauge sometimes how entrenched someone is. There are some people I'm like, they're so angry or antagonistic about it, then if I truly try to go deep and argue with them, I'm just going to entrench them further. So sometimes it's worth just backing off because, if I'm not the right person and I'm not the trusted individual, then if I argue, I'm just convincing them they're right and I'm wrong.

Q: What's the red flag for you, when you go, "Oh, I I'm actually might be doing more harm than good. I might be entrenching them?"

Carroll: When people leap from argument to argument to argument to argument, and then they start circling around again. Where it's, if I have an answer to everything they say, and they just keep leaping into arguments, I'm like, OK, this is not going to work.

But if people have a concern or a block, and I can address that and we can go in depth into that, then I feel like there's more progress that's likely going to be made. And especially if I feel like I can answer this in a way that maybe will stick and convince.

I mean, you can sometimes tell when people are being thoughtful about it or whether they're just using an excuse. I know plenty of people who are like, "Well, I won't get it because it's not FDA-authorized." And the day the Pfizer vaccine got authorized, then it was, well now they authorized it too fast. They didn't. That was an excuse, that was not a reason. And that's fine. Just, if you'd said that, we could have saved both of us some time.

Other people have genuine concerns, and I can explain why it feels fast for the vaccines have been developed, but that doesn't mean that safety got skipped. Then so many people can be convinced.

And so sometimes it's also opening it up, asking them if they have questions, seeing what kind of questions or concerns they have, answering the first few of them, and then getting a feel for, "Is this someone that we're going to be able to make progress with today?" Or is this a, "Let's just establish some relationship and trust and move along the next time?" Part of that is being a clinician and establishing relationships with patients.

Q: Yeah, absolutely. I think if we're going to address misinformation, we have to understand how it spreads. And you mentioned one way that it spreads, which is through the cable news networks. But a lot comes from online information and online discourse. But what strikes me is that I'm also seeing more clinicians, more researchers, more scientists online. You've obviously been doing this for a very long time. Do you believe that everyday clinician should be moving their guidance online and maybe even into social media platforms?

Carroll: It depends how engaged and involved you're willing to be. I think that the problem is that with social media, especially with things like Twitter, is that people think there's a magic tweet which will convince everyone that they were right and everyone else was wrong. And that never ever, ever happens. Most of the time you are preaching to the choir in that your engagement is going to be mostly followers who already agree with you or people who just retweet it. And then you just get like a mob of people who violently disagree with you. I think very few people are ever convinced by anything on Twitter.

So I've always seen Twitter as a tool. I use it to drive people to content that I think might make a difference. So columns I've written, videos I've made, other things other people have written, thoughtful articles by really good journalists or data that might sway someone.

But I always am amazed that if I have something that maybe went viral, I'm like, this made no difference. You don't understand this, no one was convinced by this. It made me feel better for five minutes.

Q: But are people convinced in the opposite way? I'm thinking there are a lot of videos containing misinformation that have been shared on new platforms like TikTok. And I see nurses, I even see physicians that are using their own medical background almost as armor to spread mis- and disinformation. Is it making a negative impact when clinicians are doing this?

Carroll: I mean, granted, there's people that absolutely believe that the answer is yes. But this is where I'm taking the long view on this. Anti-vaccination sentiment has existed as long as there have been vaccines. I mean, we did a series on vaccination at Healthcare Triage. It is not as if we needed social media to have a massive worldwide misinformation backlash against MMR; it did not need social media for that.

Now, does it make it faster and easier? I imagine it does. But I don't know how much of it is actually to blame versus—it's easy to point and say, "Well, this must be what it is."

I don't know. Was anyone expecting the vaccination in the United States to go much more smoothly than it has? I mean, we don't ever get more than this number in flu shots. I think I saw from the CDC that, right now, something fewer than 20% of young adults are vaccinated against HPV. If we don't mandate vaccines, people don't take vaccines. That's how it goes.

Now all the vaccines with very high levels of vaccination are mandated, and organizations and the schools that have mandated the Covid vaccine achieve very high levels of vaccination. When we don't mandate them, it doesn't happen. Blaming it on social media may feel convenient, but I don't know that that's really the cause.

Q: Do you believe that health care organizations, medical boards, professional boards, are they doing enough to enforce standards on physicians, on nurses who are spreading harmful messages?

Carroll: Well, they're just starting to threaten to do something. And so they really haven't done much.

Having said that, it's hard to police this stuff. It is very easy for physicians to couch themselves in specific patient information or uncertainty or levels of evidence, because again, we deal with uncertainty so often.

I see all the time where patients are like, "I know this is what we're supposed to do in this situation," but where my patient is different. And there's a lot of acceptance from both patients and physicians for that kind of attitude.

Q: In a lot of different ways. "My patient is different from a safety perspective."

Carroll: Hate guidelines, hate protocols, hate anything, because my patient is different and I know better. And that has also existed long before Covid.

So policing this is—I don't want to say a slippery slope, because I hate the word. But if they're going to start with this, there are lots of other areas where we also could say, well, this isn't right either, and that's not right and that's not right and that's not right.

We just don't do that. Unless things get really egregious. And maybe right now, we're at "really egregious," but I'm sure it's hard for the organizations and licensing boards to want to wade into that.

Q: We're talking about combating misinformation between the physician and the layperson. But one interesting trend is there's just a lot more online communication between clinicians. Clinicians are using open online platforms to actually debate with each other. Does that online communication quicken the pace of translating new research, new ideas into clinical practice? Or is there a downside?

Carroll: I think it's both. I think it probably does, but again, this is where I think it's important to understand that it's still probably a minority of clinicians engaging in this space. And so while it seems like it's huge and pervasive, it's still mostly a smallish number of massively exposed people. And that goes across the board.

I think in general more transparency is better. I think the public understanding that there is uncertainty in a lot of what we do, and being able to ask open and honest questions of their clinicians—I think that's massively important. So I think that's great. And I don't think it's bad for doctors or any other clinicians to be on social media or to have a presence or to answer questions. I think that's great, but I do worry that not everything that's said is true, and people often hang their hat on credentials as if that's the metric by which we should trust. And that's a problem.

Q: Or, let's be honest: People, even experts, can see different things in data can come to slightly different conclusions. And that again could have a downstream impact to real people who are going, "Oh, they don't know what they're talking about."

Carroll: The most angry professionals have gotten at me might have been two years ago, when there was a series of randomized controlled trials in Annals of Internal Medicine that looked at, "What's the real danger of meat?" And that the evidence is not great.

So I wrote an editorial on it. And I would argue I was taking a reasonable take of, "Let's assess the evidence." And clinicians lost their minds because—whatever side you fall on the meat wars—it's going to kill you. Or people have this anger and vitriol.

And I would be like, this is the issue. We don't know, but both sides are convinced they absolutely do know. And the other side's lying. And I could see how for the general public, that could be massively confusing.

Q: And it comes back to your comment about policing. We talked about medical boards policing in a very specific, strong way, but is there a role where you do want clinicians online to be policing each other and saying, "Hey, maybe we shouldn't do this publicly," or, "You are wrong," or "You are spreading misinformation"?

Carroll: The issue was less that it was public than how angry it got. I think it was good to have, honestly, a discussion of how questionable the evidence is in some of these cases.

I think people understanding that there is some gray in health care and we're all doing our best to understand it better—if people understood that and saw that play out, that might increase trust. I think people viewing us screaming and yelling at each other like children will only decrease trust.

And so it's the way we do it sometimes, not just that we do.

Q: I think you are spot-on. If there's one thing we know that works, it's the trusted relationships that patients have with their clinical team. We need to figure out, how do we use the media? How do we use the internet? How do we use the existing relationships we have to keep building that trust, which might mean being transparent about what we don't know, saying that this is a gray area.

Carroll: Yeah, I say "I don't know" all the time. I don't understand why people are so afraid of that. And sometimes that means "I don't know, I've got to go look that up," and sometimes it means "I don't know because we don't know."

Even when we talk about things like masks—people talk about masks with a really fairly large amount of surety. And I'm like, OK, there are situations, and we have a knowledge base about when masks might be useful. But then there's times where it's like, yeah, the absolute value or the benefit is probably getting small.

I mean, if you're talking about, "Should I wear a mask if I'm sitting outside with someone 20 feet away?" But other people are like, "No, masks are always needed." It's like, OK, now we've got to be able to talk about the nuance here.

And we've got to be able to do so carefully and do so honestly, and dispassionately and not assume the worst in each other. But especially on social media, too many of these discussions become just yelling at each other.

Q: And this is where good digital citizenship becomes really, really important. You use all sorts of platforms to communicate with your peers, with the public. What advice do you have? For other clinicians who might be thinking about getting a little bit deeper into their online presence, how should they be practicing good digital citizenship?

Carroll: For me, at the beginning, I tried to ground almost everything I said with evidence. Even when we started the blog in 2009, 2010, it was not that I wanted to come and tell you my opinion. It was that I wanted to explain, "Here's the reason I believe this. And here's all the evidence."

And if you disagree, there's a comment section, and let's talk about it. But it wasn't "trust me." It was "let me explain why."

And I like to think that that's what my columns are too, that they're full of links to research and I'm explaining why this study matters and why this is so and what evidence and what caveats exist and how I get to this opinion—not "I just believe it because." Then we can debate the rationale behind it, as opposed to just having a yelling match as to what we each believe.

Q: I love that. "Let me explain" instead of, "just trust me implicitly; everything that comes out of my mouth should be chapter and verse for you." Instead it's, "Let me explain." I love that.

Carroll: And it is building trust. I agree with you. It's like, I don't expect if I show up with one blog post that people are going to believe me.

And in the beginning, no one came to the blog. But over time, people said, "OK, these guys are rational and they're explaining it and they get it." And journalists started to pay attention, and it built an audience.

Healthcare Triage is the same way. It takes time to build that level of trust, and you don't ever want to squander it. So I try to be very careful, but as I said before, it takes time. I think people often want to show up in social media and think, "Let me get viral as quickly as I can."

Q: Which is dangerous.

Carroll: You can do that. It can be done, but that's never been my goal. It's more, I want to build a level of trust. And then—that is one of the things I will say I like about social media: I can follow journalists that I trust, as opposed to just reading outlets.

And so even during the pandemic—Ed Yong, The Atlantic, Amanda Mull, or Olga Khazan. Or it's STAT News like Helen Branswell, or Matt Harper. Or I grant that I've colleagues in the New York Times that I really follow, but I follow individuals and journalists that I've learned to trust, as opposed to, I just read the New York Times.

Q: Yeah. How do you handle the trolls?

Carroll: Mostly two different ways. If they're horrible people, I ignore it. If somebody sends me an email and they took the time to write, if they at all seem reasonable, I will sometimes answer them and surprise them.

And nine times out of 10, you'd be surprised. People respond by like, "Oh, now I feel terrible. It didn't occur to me that, like, you're a human being and you might actually read this and respond."

Q: Because you see it as an opportunity to build trust.

Carroll: Yeah. And so sometimes you will break through, but I mean, clearly if somebody is just being terrible, I just ignore it.

Q: Yeah. Well, this has been unbelievably helpful for me also as somebody with her own kind of social media presence. And I know that our listeners and the clinicians who are listening to this podcast will find it valuable as well.

At the end of my episodes, I always want to give our guests the platform and the chance to just speak directly to our audience. So when it comes to the world of medical misinformation and disinformation, is there one thing that you want health care leaders of all kinds to focus on or act on right now?

Carroll: The biggest thing is, don't miss an opportunity to connect with patients. I know everybody is busy, and I know that it's really hard, and this has been an incredibly stressful year and a half, it's ongoing. But it is amazing to me that, in poll after poll, doctors remain the most trusted source of information.

Above anyone you see on TV, above any politician, above any "expert," people trust their doctors. And we should make use of that. And take it out of duty if you can. Connect with patients. You'll probably do more to convince someone to get vaccinated or do the right thing and all the other messaging.

Q: And if you're an administrator, make sure that your clinicians have the protected time to do that, because I agree this is an untapped resource that we need to use going forward.

Carroll: Yeah, absolutely.