A boy waved through the window to his father, who was recovering from COVID-19 in an Osage Beach, Mo., hospital in July 2021.

Brody Barker waves to his father, Daryl, from outside his hospital room in Osage Beach, Mo., as he recovers from COVID-19, in this July 2021 photo.

Science journalist Ed Yong of The Atlantic has earned the reputation as one of, if not the, top chroniclers of the COVID-19 pandemic. He won the 2021 Pulitzer Prize for explanatory reporting with a series of in-depth pieces about the interconnection of science, politics, economics, culture and humanity in the first year of the crisis, and has continued to cover it as variants, vaccines and vaccine resistance add to the complexity. The Pulitzer judges noted that his stories last year “anticipated the course of the disease, synthesized the complex challenges the country faced, illuminated the U.S. government’s failures and provided clear and accessible context for the scientific and human challenges it posed.”

Atlantic science reporter Ed Yong

Ed Yong

Yong’s COVID stories all have a strong foundation of authority, clarity and accessibility: He makes sense of complex science by using specific examples and everyday analogies. But the knowledge he pulls into that center are wide-ranging, built from tireless interviews with doctors, nurses, hospital administrators, public health officials, patients, grieving families, scientists, and government officials. These sources give voice to the varied sides of a mind-bending maze and and rapidly changing situation fraught with emotion and plagued with public misinformation.

While the bulk of his work could best be categorized as explanatory, recent work has included more emotional work. A piece from earlier this month explores the plight of long-haulers. In mid-July, he did an intimate story about the frustration and exhaustion of health care professionals at two hospitals in southwest Missouri, which were overwhelmed with unvaccinated patients infected by the Delta variant. In between those two, he filed “How the Pandemic Now Ends,” a structural echo of his March 2020 piece on “How the Pandemic Will End” and a signature example of his explanatory work.

We wondered how Yong, working on deadline, navigates the challenges of such stories. How do you cover a story that, like the virus itself, mutates as quickly as we can make sense of it? How do you elicit specifics from health care workers, who must speak carefully to avoid violating health privacy laws? How do you connect with subjects in a time and place where face-to-face contact is dangerous or prohibited? How do you ration face-time with people who are working through exhaustion in a race to save human lives? And how do you take care of your own exhaustion through it all?

… there are so many things that we prize in journalism, and so many things we talk about as cultural professional values. But I feel like we don’t talk enough about the value of empathy, and of being kind to people. ~ Ed Yong of The Atlantic

We asked Yong about all of that, and especially about his approach to interviewing sources that range from the top scientists in the world to the hands-on nurse in a hospital ward. Our conversation has been edited for length and clarity.

Can you describe your general approach to interviewing?
I try approaching things in a fairly open way. A lot of the people in this piece were very open and unguarded in their views. And I honestly think some of that comes from my desire to really learn about what these folks are going through. I’m not doing an interview to get good quotes. And I’m not doing it to confirm something I already thought. I’m really trying to pick people who are going to teach me something new about the situation that’s unfolding. I want to tap into their expertise. But I also want to learn more about their experiences and their emotions.

There are sort of two categories of pieces that I’ve written. One is the very big, sky-high view of the pandemic, and where we are, and where we might get to. And then there are more intimate pieces focused on two groups of people: either long haulers (people with long COVID) or healthcare workers and public health experts who are experiencing burnout and other emotional difficulties in the middle of the surges.

For the latter, my interest is not only to find out what is happening around them, but also to find out what’s happening inside them. What is it like to be a healthcare worker in the middle of a COVID surge? What was it like to be a healthcare worker in the middle of the last COVID surge, when things were really starting to take off, and is it different now, in an era when vaccines are around? Does it make a difference to them that they are vaccinated? Anyone can go to a number of news sites and find the number of people who are hospitalized with COVID in, say, Missouri. But you can’t know what a healthcare worker is going through unless you actually ask them.

How do you decide whom to interview?
I try to talk to a wide range of different sources across expertise, and specifically across gender and race lines. The majority of experts quoted in media tend to be white and male. If you just try to find women or people of color to talk to, you’ll tend to get people who’ve done fewer interviews in the past; every piece I write, I try and ensure that at least half the people I’m talking to are people I’ve never interviewed before.

Every piece I write, I usually try and ensure that at least half the people I’m talking to are people I’ve never interviewed before.

You interview a lot of scientists and policymakers, and a lot of those people tend to be trained to handle interviews. Does their training make your job easier or harder? And when you interview healthcare workers on the front lines, who probably don’t have media training, do you approach them differently?
I don’t change my approach. I don’t necessarily find the people who have been media trained are any easier to talk to than people who aren’t. Sometimes it can be harder. I have been media trained, and I do interviews, and I know that sometimes I’m going to answer the question that I want to answer rather than the one that I’ve actually been asked. That is a problem when you talk to people who’ve done a lot of these. Another is that you might get answers that, if not pat answers, are things that people have said before. You ask a question and get an answer that this person has clearly given many times over, rather than one that specifically addresses the thing that you’ve just asked. So I’m not sure that media training massively helps. I do the same thing, whether I’m talking to Tony Fauci or to a nurse who’s never spoken to the media before. 

How do you get access to that second layer of people doing hands-on work with patients?Access is less of an issue than it might seem, especially under the current circumstances. You mentioned that hospital people tend to be quite guarded; they can often close ranks. I’ve talked to healthcare workers who I found on social media, who asked for their institution not to be named. I’ve arranged interviews with people who came back to me afterwards and said, you know, our press office just nixed the interview.

…what I find is that this situation is so devastating in so many hospitals, that a lot of them just want to get the word out.

But what I find is that this situation is so devastating in so many hospitals, that a lot of them just want to get the word out. People want the public to understand how difficult it is, what the cost of having an uncontrolled pandemic actually is for them. So for a lot of the hospitals that I have spoken to, I’ve just gone to their head of communication and asked to be hooked up with five doctors and nurses and respiratory therapists. Often people are incredibly willing to do that. And in most of those cases, the press person hasn’t even sat in on any of the calls. I absolutely would never agree to let people review copy, but that has never even been put on the table as something that they want. I think what a lot of hospitals really want right now is for people to understand how vulnerable the U.S. healthcare system is.

It certainly comes out in the piece out of the Missouri hospitals. You can hear the desperation in their voices and in what they say. For a story like this, how many people will you talk to?
For the three hospital pieces that I’ve done, I think I’ve spoken to probably a dozen people for each one. Time constraints are a factor, obviously. We all work to deadlines. And usually when I’m writing these kinds of pieces, it’s because things are really hitting the fan. So these pieces need to be done a little bit more quickly than some of the sweeping, big-picture looks. I try and talk to people with a range of different specialties. When I’ve gone through a press office, what I specifically asked them is, I don’t just want to talk to doctors and I don’t just want to talk to your chief of medicine or your hospital CEO. I’m happy to talk to them. But I also need to talk to people on the front lines, people who are treating patients right now. I want to talk to a mix of physicians. And I absolutely want to talk to nurses, because they do so much work on the front lines, they handle so much of the emotional burden.

With a recent piece, I really wanted to talk to at least one respiratory therapist — another group of professionals who often gets ignored in these discussions. People have called me out on that before, and wanted to rectify that. For the Missouri piece, as well as talking to people from the two big hospitals in Springfield, which was at the time the epicenter of the epidemic in Missouri, I also talked to folks in the local public health department. Given that issues about vaccinations and people who weren’t vaccinated was so critical to understanding what’s happening, I wanted to talk to folks on the frontlines of the vaccination drives and people who worked in public health who had experienced trying to get people in the local communities to take the shots.

With pieces like this, it seems easy to over-report. How do you know when you’re done?
There comes a point when you start hearing the same things again. Like if you’re asking people to describe their opinions, you start to asymptote, in terms of how much new stuff you’re getting. If I’m reporting on a bigger piece, like a big-picture look at COVID, I’m a little bit happier to over-report. I think my record for one of those magazine features was 40 people. A third of those weren’t quoted in the piece at all, and most people were quoted like a sentence or two. Everyone helped.

When I’m talking to healthcare workers in the middle of a surge, or long haulers, I try and be very sensitive of their time. The healthcare workers in the middle of a surge have, frankly, other shit to do besides talk to me. And if I’m going to call them away, or soak up half an hour of their break time, I damn well better make sure that I’m using that time wisely and respectfully. Likewise, if you’ve got long COVID, a half-hour interview might knock you out for days. So I try to be very, very sparing with time. I try to talk to exactly the right people I need to and no more.

And sometimes it depends on the type of story that I’m telling. If I’m going to make a claim that hospitals in Missouri are overwhelmed, I need to talk to enough people to make sure that it’s not just one person saying that — that it’s actually multiple sources backing each other up. If I’m going to make a claim, which I did, that not only are patients younger, but the younger patients seem to be sicker than before, I made sure that I confirmed that with five or six different people. Sometimes the gravity of the claim forces the number of interviews you need to back it up.

Twelve people isn’t going to be a representative sample, but it’s certainly going to be better than just doing a Q&A with one source.

That’s a really smart litmus test: The bigger the claim, the more people you need to interview.
If I talk to one nurse who is saying that they’re having a really bad time, but everyone else is fine, I’m going to represent that in the piece. If every single person I talked to is on the verge of a breakdown, I’m also going to represent that in the piece. Sometimes, you need to get a large enough sample size for the pieces that have more of an emotional bent to them to fairly represent the situation on the ground. Twelve people isn’t going to be a representative sample, but it’s certainly going to be better than just doing a Q&A with one source.

Do you interview them in person, on the phone, on Zoom, or a mix?
Usually on the phone. In fact, every interview I’ve done for the pandemic pieces has been on the phone. And I usually do just audio; I don’t really find video helps. Sometimes people find it uneasy to be on video. I just want to hear their voices.

Does your interviewing approach change based on that lack of in-person proximity?
I don’t think my approach has changed very much. There are obviously details that you could theoretically get in person. I had a chat with my editors about whether to go to Missouri or not. I’m fully vaccinated. I don’t have massive concerns about my own safety, given that. But I’m also not massively keen to test out the efficacy of my vaccination by heading into the COVID epicenter of the U.S. More germanely to that question, it was clear that in-person reporting wouldn’t get much more detail or color beyond what I could get over the phone. One of the hospitals made it clear that I wouldn’t be allowed inside the hospital as a visitor, let alone into a patient’s room, so I’d have to interview people out in the parking lot or something.

You can get a lot through the phone. We sometimes forget this. In-person reporting is highly prized, and reasonably so. But you can ask people to describe what it’s like. You can ask them to fill in all the sensory details that you would otherwise get in person. And when my goal is to ask them to tell me what their experiences are and, crucially, what their emotions are, I don’t need to see the sore on the nurse’s nose to know that things are bad for them.

What kind of questions do you ask to get those sensory details?
The first questions I ask are usually pretty open ended. I tell people what I want to do in the piece. Like with the Missouri hospital story, I’ll say that I know things are bad in Missouri right now; I want this piece to explore what it’s like on the ground and what it’s like for healthcare workers to go through this particular surge during the era of vaccination. I start by explaining my intentions as a baseline for our conversation. Then I throw it over to them. I tell them that before we go into any specifics, I want to know whether they had any immediate thoughts on the situation, the framework, what I just said to them. I ask them to tell me what they’re thinking. That usually gives me some pretty clear signals about what they want to talk about, and I can use that to guide the rest of the conversation.

Do you go in with prepared questions, and if so, how do you decide when to deviate from them?
I usually go into interview with a list of things I might want to ask, but I let the answers guide flow of the conversation. I’m not asking people things that either they have no expertise in, or that they’re not massively keen to talk about. A lot of the questions are essentially me saying, “Okay, let’s go back to that thing that you just said…” Then I repeat exactly what they’ve just said to me (I transcribe in real time) and then say, “Say more about that. When you said this bit, tell me more about what you’re feeling.” Or “I didn’t quite understand that.” I will type my own responses, my own reactions to what they said, in the flow of the conversation. If someone says to me, “Most the patients we see are unvaccinated.” I might ask, “When I hear you say that, I feel like it’s a tragedy, given that the last time we went through this there weren’t any vaccines. Now there are, and yet you’re sort of going through the same experience. Does that ring true for you?” What I’m trying to do is use that as a springboard for them to push off against. It’s a way of saying, “First, have I understood what it is you’re telling me correctly?” Then, “Here’s what I think. What do you think about that?”

It’s almost like you’re fact checking your own assumptions, in real time, in the middle of the conversation.

I think that’s an interesting thing to warn young reporters about. Sometimes they embed assumptions in a question. Or they move too quickly to the next question without saying, “Here’s what I’m hearing. Am I getting that right?”
Yes. There are two mistakes you can make. One is going in completely unprepared. And the other is to go in with a very formalized, concrete set of things that you then just work your way through. That will get you somewhere, but it won’t get you as far as if you had a more flexible approach. Because if you’re doing that, then you’re not really listening to the interviewee. If you’re really listening and trying to actually think about what they’re saying in real time, which is in itself a skill, then your interview should change according to what people are telling you.

I see my questions more as a list of the kinds of things that I want to cover — a list of fallback options in case we run into conversational dead air. But it’s not a checklist.

I teach a college creative nonfiction class, and many of my creative writing students have come through fiction and have never done an interview. I try to impress upon them the mental processing power required when you’re doing an interview. You’re asking a question. You’re listening. You’re typing or transcribing and thinking of the next question, but also thinking about your list of questions. Sometimes I feel like my processor isn’t fast enough for that!
It’s exhausting! I’ve been doing this for a long time now. It gets easier, but when I do these hospital pieces, they’re emotionally wrenching. Usually I’m trying to do them on quite tight deadlines — the fact that the pieces need to be done at all means that things are really bad — so I’m doing these dozen or so interviews over the course of a day and a half. Often I have seven half-hour calls in a row. I feel more tired after one of those sessions than I do after a really intense workout.

How do you recover from that? Do you have to go a couple days and do something else to mentally rejuvenate? I know I do.
The mental health aspects of this are very difficult. There are basically two types of pieces I’m writing, right? The sky-high view ones, and the emotional underground ones. The former are conceptually challenging, and the latter are emotionally challenging. And both of them are very, very, very exhausting. It is really hard to listen to people cry on the phone. That has been the reality every single time I’ve written a piece about healthcare workers or long haulers. People always cry.

It is really hard to listen to people cry on the phone.

How do you deal with that?
I give them space. I usually say something affirming. “I’m sorry. I know this is really hard.” I think that’s part of what I said about letting the interview guide you. If someone’s crying on the phone, I want to talk about that. I might ask them, “How do you get the space you need to deal with these kinds of raw emotions?” I’m not going to just ask the next question on my list.

There are so many things that we prize in journalism, and so many things we talk about as cultural professional values. But I feel like we don’t talk enough about the value of empathy, and of being kind to people. We have a certain distance we maintain from the people we’re talking to. But that doesn’t need to stop you from being a decent human to someone who’s suffering in the moment. You can still be kind and get what you need.

I did not think that I would be back at this position in the seventh month of the Year of Our Lord 2021, talking to health care workers who are breaking down again. That frustration of being in that same place again — I think that helps me empathize with people who are also in that place again. It feeds into the kinds of things that I might ask or talk to them about. I am not in the story, but I have also been through this pandemic and some of these same experiences. The recovery process is… I won’t lie, it’s been really, really hard. I don’t feel that I have a good handle on it.

When you’re talking to healthcare workers, do the health privacy restrictions of HIPAA ever come into play? Do you know what you can and can’t ask? Or do they know?
The healthcare workers I’ve spoken to have a pretty good sense of how to deal with that and I let them lead the way. Some of them talked about some of the experiences they’ve had, but they were very clear about what kinds of details they could and could not offer.

Was there any specific, interesting challenge that you faced in any of these interviews?
I don’t think so. I think people respond to who you are as an interviewer. I have rarely had problems with people being very closed. I think that’s because I try and be very open. I try to be an empathetic listener. I try to respond to what they’re actually saying. I think you get better results out of people in that way.

Often we are asked, “How do you get a source to trust you as a journalist?” But we don’t ask enough about “How should you be trustworthy to a source?”

Often we are asked, “How do you get a source to trust you as a journalist?” But we don’t ask enough about “How should you be trustworthy to a source?” I think that’s crucial. It gets you a lot further. Sometimes people don’t really want to share their emotions. Sometimes their answers are a little shorter. They’re a little unsure, or haven’t talked to the press before, they’re being a little guarded. And that’s fine! There’s almost no value to be gained from poking people. You can create opportunities for them to open up. But if I’ve asked someone, “If you’re walking through the ICU, you’re seeing the things that you’ve just told me about, how do you feel?” — and if they’re not giving me details, if they’re just saying, “It’s very frustrating,” — then I’m not going to vampirically tear open their psyche and pull out a quote.

Again, on the basis of just trying to be kind to people who are doing their best under a hard situation, you can move on, you know? Maybe in that case, for that interview, it’s more about trying to get the actual raw details of what it’s like to go through what they’re going through.

It’s interesting that you said we often talk about how to how to get people to trust us versus how to trustworthy. Can you talk a little bit more about that? How do you work on being trustworthy — as a human being, not just a journalist. How do you make an effort to earn or deserve that trust?
Again, I start interviews by telling people where I’m at and what I want to do. That sets expectations very clearly upfront. And some of it is achieved through the flexible and open act of interviewing, by showing people that you’re actually listening to what they’re saying.

When I interview long-haulers, I say up front: “I know that doing interviews can be really hard for you. So if you’re feeling fatigued in the middle of this, then you tell me that we need to stop, and we will stop. We can reschedule for a different time.” I said that to someone once, and they were completely stunned, because no one had ever said that to them before in the interviews that they’d done.

The key to convincing people that you’re trustworthy is to actually be trustworthy. You can’t fake that. Which sort of goes back to my point about trying to actually be empathetic and kind. There’s a perception among some that that’s antithetical being a journalist. You have to be distanced from your sources, you have to be like hard-nosed or whatever. But you can ask people hardball questions without being an ass about it.

Let’s take this example: One of the common things that long haulers go through is that they get disbelieved by the medical establishment. So instead of saying, “Could your symptoms be in your head?” — which is something that literally everyone asks them — you can ask, “What do you say to people who think your symptoms are in your head? Have people said that to you? How has that made you feel?” You can ask hard questions, but there are ways of doing it that actually respect the people you’re talking to.

When you’ve written your piece, I know there are fact checkers who come behind and check quotes and everything. But do you do any circling back to read a sentence to someone and say, “Did I get that right?” I’ve talked to people who will read parts of a piece to a source to make sure that they’ve used words that ring true.
Yeah, I’ll sometimes send people emails with, not the actual text, but a very gently paraphrased version of what I’ve just written: “Is this accurate?” I don’t usually check quotes; I’m pretty confident in that stuff. Sometimes it’s more about basic fact checking. You know, like, if you’ve told me some numbers around your hospital, let me just make sure that I got those correctly. If you’re making a big claim about the kinds of things we’re seeing, I will check that to make sure it’s correct.

In your pieces, you use quotes to a great degree to let the people speak for themselves. Instead of trying to describe what they’re feeling, you let them tell us, the readers, how they’re feeling. Anything else you’d like to share for people who want to get better at interviewing people in in tough situations?
The question that you literally just asked me —”Anything else?” — is great. I definitely do that.

I honestly think that the key to this is to go in openly. You will get the best results out of people if you genuinely want to find out what they’re going through. That’s the whole point of interviewing. I think we sometimes lose sight of that because of the process stuff, right? Interviewing is not just for getting quotes from people — it’s for actually understanding them. And if you approach it that way, you’ll get better quotes.


Kim Cross a freelance magazine writer and author of “What Stands in a Storm,” a New York Times best-selling account of the biggest tornado outbreak on record. She teaches narrative journalism near her home in Boise, Idaho.

Further Reading