Podcast: Breakthrough Infections and Lonely Puppies

Things are starting to look up, at least in the U.S., but we’re looking ahead at potential future worries.

A man wearing glasses, a surgical mask, and rubber gloves examines laboratory samples in small vials.
Molly Darlington / Reuters / The Atlantic

While COVID-19 case counts in the United States continue to drop, you might still be reading worrisome headlines about variants and “breakthrough” infections. Fortunately, The Atlantic staff writer Katherine Wu explains to James Hamblin and Maeve Higgins why these shouldn’t alarm us just yet. And staff writer Sarah Zhang drops in to help figure out how to keep pandemic puppies from being too anxious as people return to pre-pandemic routines.

What follows is a transcript of their conversation, edited and condensed for clarity:

Maeve Higgins: People are ready to party, but if people are still getting COVID-19 after the vaccine, even if it’s not that often, how worried should we be?

Katherine Wu: I think there’s two ways to answer that question. Collectively, we should not be super worried. I think on a population level, these so-called breakthrough infections where people are getting infected with the virus and very occasionally getting sick despite being fully vaccinated—it’s so, so, so rare. And even the people who are picking up the virus don’t seem to be getting as sick; in short, the vaccines are doing their job. That said, I don’t want to downplay how concerning this can be for that individual person who does get infected or sick, or the worries of the people around them. It’s really tough to talk about these breakthrough infections because we do want to track them and pay attention to them on an individual level. But broadly, I do not see anything that is unexpected, to be totally honest.

James Hamblin: Can you catch us up just on the basic numbers? How many cases have been reported in the U.S., and how are we defining cases?

Wu: There’s kind of two numbers that I can tell you at this point. One is going to be a number that is only current through the end of April, and that’s back when the CDC was tracking all breakthrough infections that were reported to them, regardless of severity. So if you test positive for this virus, and your health department reports it to CDC, they’re going to count it even if you’re asymptomatic. That was in kind of like the 10,000 range.

But I really hesitate to do math here, because it’s super tempting to be like, “Oh, there were about 10,000 of these cases. And by that point, like, I don’t know, 100 million people were vaccinated. So let’s just divide.” That’s super tricky because we know that not all 100 million of those vaccinated Americans were exposed to the virus, so they didn’t all have the opportunity to get infected. We can’t just say, like, “Oh, this is exactly how effective the vaccines are.” And the way that the CDC is sort of tracking that number, if you look on their data tracker, they actually add people to that “fully vaccinated” column the day they get their second shot.

And we know that full vaccination in terms of, like, how immune, [how] protected, you are, doesn’t really kick in until a couple of weeks later. So it’s really hard to do that kind of math, but that is still pretty good odds in terms of the big picture. Ten thousand people by the end of April. We also do know that’s an undercount, given that there are probably going to be … a decent more number of asymptomatic infections after people are vaccinated. And again, that is the vaccine doing its job. It’s keeping you from getting sick.

At the start of May, we started to transition into this different set of numbers as the CDC stopped tracking breakthrough cases that were not associated with hospitalization or death. So now they’re only tabulating on their website cases where people ended up in the hospital or ended up dying and were also positive for the coronavirus. That doesn’t always mean the coronavirus caused their sickness or death, but they tested positive and they also happened to be in the hospital. So it’s tough.

And when the CDC made the switch, it was kind of controversial, because people were like, “Well, how are we going to get the full range of data here? How are we going to know if there’s, like, a variant that is more consistently making people sick if we don’t have anything to compare that to?” But as you can imagine, the numbers have really dropped since the CDC did this. And so now it’s current through May 24, 2021. There have been 2,454 hospitalized or fatal vaccine-breakthrough cases reported to CDC, where the person was also positive for the coronavirus.

Hamblin: I mean, do you have a sense that there are many people in there who had, you know, say, a heart attack or a car accident and died because of that and are included in that number? Is there any way to know how much of that is actually attributable to a serious case of COVID-19?

Wu: So, to the CDC’s credit, they do actually put some little asterisks on this little spreadsheet here. Five-hundred-forty of those 2,000-plus cases were actually reported as asymptomatic. So we know that not everything in that bucket that we just described is, like, someone dying of COVID-19, which I think is an important distinction to make, because I have seen some people talking about this on Twitter or in different news outlets and saying like, “Oh, this is the percentage of breakthrough cases where COVID-19 is killing people.” And that’s not quite accurate. I mean, it may have had something to do with it. I can’t know the internal workings of every person’s body that is in this list, but I think it’s safe to say that sometimes infections just happen at a really unfortunate time. So it’s really hard to draw firm conclusions based on just this number alone.

Hamblin: Yeah. Do you know why they made that switch toward tracking in this new way? So, not tracking just every single person who’s tested positive but only the hospitalization or death cases?

Wu: Yeah, it’s a really good question. And it’s a question that I think stirred a lot of debate in the past few weeks as this became more public knowledge. Basically, the CDC justified this by saying, “We are keeping track of the cases”—quote unquote—“of most public health and clinical concern,” though that also felt a little weird to me because, again, COVID-19 is not necessarily the direct cause of hospitalization or death in these cases. I suspect there’s also kind of a pragmatism at work here, just because it’s really hard to cast a net wide enough to say confidently that we are really getting a good sense of all the symptomatic cases or—God forbid—we’re trying to get a sense of every single infection, asymptomatic or not, that’s out there. You know, hospitals and other places through which people pass when they’re really sick—they’re going to keep pretty good medical records. It’s probably going to be easier to figure out if a person has been fully vaccinated or not. So to tell if they qualify for a breakthrough case, it’s just easier to to track.

Higgins: Yeah, but the line in your piece that really struck me was the goal of vaccination isn’t eradication, but a détente in which humans and viruses coexist with the risk of disease at a tolerable low. That helped me to think about it in a more practical way, I guess.

Wu: Yeah. And I think also thinking long-term here, I could see this kind of being a more sustainable way to track breakthrough infections, just because labs across the country have just been slammed with, like, “Please sequence everything; please test everything,” for so long now. It’s really tough, and that wouldn’t necessarily be the most sustainable way to go forward. But I also do worry that this was soon; this happened even before everyone in our country, much less the world, was eligible for a vaccine and had access to a vaccine. So “What are we missing by putting our blinders on?,” I think, is a huge question here.

Hamblin: Right. Are we picking up on any trends as to who is prone to breakthrough cases that are significant, any relationship to how long ago people were vaccinated or to age or chronic conditions, or is it too soon to see any patterns?

Wu: Yeah, I think the short answer is that it’s too soon to see any huge patterns, though a couple slight and maybe unsurprising ones have been picked up. The first is that the majority of the breakthrough cases that are documented as being related to hospitalization or death, they are occurring in people who are over the age of 65. And we did kind of expect that vaccines might not be quite as efficacious in people who are older, just because their immune systems are a little sleepier.

But apart from that, it’s not huge. I think the other thing that people are really on the lookout for is: Are we seeing that particular version of the virus that has specific mutations? Is it consistently eluding the vaccines? And mostly the answer seems to be no. But it’s really, really hard to tell, because less than 10 percent of these reported breakthroughs have actually been whole-genome sequenced, which means we can read the entire virus’s genome from start to finish. I think that’s something that a lot of people are concerned about with relation to what breakthroughs are we tracking and how many should we be tracking at once.

Higgins: You know, I’m still in Ireland, and they just released information about the first person who’s definitely gotten the virus twice in the space of eight months. So is it like that? Is it the same thing where, like, your body was able to cope with it and then you get a lesser version of it? Because she didn’t get it as bad the second time.

Wu: Hmm. First off, that’s very good to hear. That hopefully means her body built up some pretty decent defenses that maybe weren’t perfect against the second version of the virus, if it was a different variant. If we were to see that a majority of people who are vaccinated and getting infected are getting infected with a variant, and the proportion of those people who are getting this variant exceeds the proportion of people in the population who are getting this variant who are just unvaccinated, yeah, I would maybe start to be a little concerned, but it’s not the end of the world.

I would also want to look at how severely are these people getting sick, because you’re right, it is kind of similar with both natural infection and a vaccine. The body sees this invader or something that looks a lot like it, and it prepares a bunch of defenses and squirrels them away. And maybe it kind of learned the wrong version of the virus. But it can still tell a few things. I think about it as like a mug shot. You take a mug shot of a criminal, and he comes back, but he has grown a mustache and you feel a little bit confused. But for the most part, it’s still like, Okay, I still kind of know what’s going on, and I’m still going to take care of this. So I’m still mostly okay with it.

Higgins: I’d know those eyes anywhere. Even if you are wearing glasses with fake eyebrows and a fake mustache attached.

Wu: The virus is like, How did you know?

Hamblin: So it sounds like you’re not extremely concerned by what you’re seeing in terms of breakthrough cases at this point. And there’s nothing that we’ve learned that should change the overall messaging that most people have gotten about vaccines being extremely effective and how life should basically be able to go on as pretty close to normal, as long as you and the people you’re spending your time around are vaccinated, correct?

Wu: Yeah, I think that’s right. And it’s probably worth it to point out that for months now there have been all these headlines about like, “Oh, these scientists tested this variant in the lab. And all of these antibodies were like, Oh, crap, what’s going on? I don’t recognize this thing.” And there were really terrifying numbers about, like, 40-fold reduced efficacy. In a vacuum, it is true that some of those antibodies were not doing as good a job against the virus, but those were single antibodies. The immune system is so complicated. It has so many different arms and branches.

Basically, the immune system is not putting all of its defensive eggs in one basket. And what’s been really encouraging is that when people really zoom out and don’t just look at what’s happening in a laboratory petri dish, they’re seeing the vaccines are still really effective against variants. It’s another reason why we shouldn’t obsess too much over only antibodies. Even though they’re great, they’re not the whole picture.

Hamblin: So what else I’ve been concerned about is my puppy, Moses … I’m not the only one, it turns out, who has gotten gotten a pet during the pandemic, who now is extremely attached. He can’t be not in the same room as me or my wife or he goes crazy. And that’s not just me.

Higgins: I’ve read it too. And that’s why we’re going to talk to Sarah. So will you stay with us, Katie? Because we’re going to talk to your other colleague, with Sarah Zhang, because she wrote about this.

Wu: I would love to. And if I cry about the puppies, you just have to bear with me.

Higgins: Okay. If your weeping gets too loud, we’re going to mute you.

Wu: Don’t minimize my pain, Maeve. (Both laugh.)

Higgins: Sarah, are you there?

Sarah Zhang: Hi. Yes, I am here.

Higgins: Thank you so much for joining us.

Hamblin: And you have a cat with you.

Zhang: I have two cats, though I would say they’re maybe not super relevant for this story, because I actually believe one of my cats cannot wait for me to go back to the office. And in fact, he has this horrible habit of meowing at one of us, either me or my partner, nonstop from 10 a.m. to 2 p.m. every day until he finally decides to go to sleep. And a couple of days ago, my partner was out of town for a few days, and he was totally fine. So I think he just wants the humans to leave.

Wu: That is not my experience. So I have three cats because I’m crazy. Two of them are deeply attached to me. Like, I’m actively worried about what’s going to happen when both my partner and I leave—like, they crawl in my lap; they will tap me on the shoulder and ask me to pick them up.

Higgins: Wait, how do they tap you? Are they so tall? (All laugh.)

Wu: They are actually giant cats.

Higgins: And they walk on their back legs.

Wu: I said cats, but I meant jaguars.

Hamblin: Wow, that’s cool.

Wu: No, they are normal-sized cats, and they get up behind me on the couch, or if I am deploying bad work habits and lying down while I work, they will come up behind me and access my shoulder.

Zhang: They just want you to have good posture.

Wu: Oh, they’re saying, Stop slouching. Insight!

Higgins: But what did you learn, Katie? I mean, Jim, you probably have more questions because you’re worried about Moses.

Hamblin: Cats, to me, they’re fickle and sometimes want things that you don’t always know. But the puppies, they kind of just wear everything on their sleeves. And Moses, he follows me into the bathroom. He can’t be alone at all, and I’m worried about not being with him all the time. So, Sarah, you wrote an excellent story about this. And I’m wondering what I should do to wean my one-year-old puppy from constantly needing my presence.

Zhang: Well, you’re not alone. Literally. I talked to many dog owners who are in the same boat. And, you know, one trainer I spoke to said that she had never in her life talked to someone before who had literally never left their pet until this pandemic. And by “never,” she meant, like, not even to go get the mail or to take out the trash or get groceries. Literally the dog is with you at your side looking at you all the time. There are dogs with like real separation anxiety, where you cannot even leave them for a second before they start howling. And I spoke to one woman who had a dog, who, you know, she took a walk, and she could hear him barking and howling from a block away.

And so what she had to do, literally, was go through this training where first she wouldn’t even leave him. She would just kind of do the things that you would do before you went out the door. So she picked up her keys and then put them down and put on her coat, put them down.

And because this is COVID, she would put on her mask and take it down, until he got used to that and stopped reacting to that. And then it was literally, like, leave for a few seconds, a minute, a few seconds, a minute. And she said literally five minutes was like, We’re having a party here. She lives in Oregon. So she’s working in her garage in the cold for minutes or hours at a time while her dog is acclimating to this new work-apart reality.

Hamblin: Gotcha. So you’re getting the dog used to it, sort of training it in small increments so they’re not traumatic.

Zhang: Yeah, exactly. And I think one thing that I thought was really interesting is that apparently dogs are smart enough to realize that you’re leaving them for longer and longer periods of time, and they start dreading that it’s going to get even longer and longer. What you actually [should] do is do longer and then shorter and then longer and then shorter, so that they can’t feel like they can predict what you’re about to do.

Hamblin: Yeah, that makes total sense. That’s really helpful.