We're All Cherry Picking COVID Data

[music]

Brian Lehrer: It's The Brian Lehrer Show on WNYC. Good morning again, everyone. With us now, Dr. Dhruv Khullar, medical contributor to The New Yorker and a practicing physician and professor of medicine at the Weill Cornell Medical College. His latest article in this week's New Yorker seeks to explain what the sometimes confusing Omicron numbers really mean, and also speculates on moving from the pandemic phase to the endemic phase of COVID-19, which Omicron might be ushering in.

His article is called Do the Omicron Numbers Mean What We Think They Mean? Dr. Khullar, really appreciate you making this radio house call today. Welcome to WNYC.

Dr. Dhruv Khullar: Thanks so much for having me, Brian.

Brian Lehrer: You open your article with something that at first seems to have nothing to do with COVID or medicine at all, and that's something you call, the Texas Sharpshooter Fallacy. Would you explain it?

Dr. Dhruv Khullar: Sure. The Texas Sharpshooter Fallacy really describes this story of a Texas man who takes a rifle to the side of his barn and he just basically shoots all over the side of the wall at random. Then, he finds the densest clusters of holes and then he paints a bulls-eye around each one of those clusters, so it makes it looks like he actually hit the bull's eye every time. Then, someone comes by later and looks at this marksmanship and says, "Wow, what a sharpshooter here."

The idea here is that we sometimes narrative eyes data after the fact we see something that happens in the world, and then we try to back into an explanation, and that carries dangers. When we're looking at the Omicron numbers, I think there's a lot of confusion in part because we don't have a full appreciation of the context in terms of how to read these numbers. That's what I was trying to get at in the articles, try to give people a little bit more context. When they see things like cases, hospitalizations, and deaths, what does that actually mean? Are those actually still reliable markers?

Brian Lehrer: For example, you quote Dr. Fauci saying, "Case counts shouldn't be the main thing we look at to understand the impact of Omicron because there are so many mild cases out there, but rather hospitalizations," but you write that even the number of hospitalizations may not be quite what it seems. Do you want to explain that?

Dr. Dhruv Khullar: Yes, absolutely. One of the issues that we've seen, we are recording record levels of hospitalizations with this variant Omicron, more than 150,000 hospitalizations in the United States currently. That is a significant number. It's placing a lot of strain on the health care system, but it's also not clear that all those hospitalizations are due to the virus. There's a difference between hospitalizations that are caused by COVID, and hospitalizations in which people are actually coming to the hospital for a different reason, but they happen to test positive for the virus.

What we've seen in this wave and with this variant is that it does cause somewhat less severe disease in people. I think, if you're unvaccinated, certainly can still land you in the hospital for a really bad pneumonia, but people who are vaccinated or have had infections in the past, a lot of those peoples are coming to the hospital for a totally unrelated condition; they broke their hip, or they're having [inaudible 00:03:30], or they're in the hospital for cancer surgery, and they happen to test positive for [inaudible 00:03:36] those hospitalizations together.

It's not entirely clear what percent of the hospitalizations in the US right now are due to what we call incidental COVID, but by some estimates could be pretty substantial 30%, 40%, 50%.

Brian Lehrer: Yet, so many hospitals are now being overwhelmed because the number of patients coming in, as well as the number of staff out because of COVID, that as the National Guard is being called in at some places. Can you separate those two things? Is Omicron driving a lot more hospitalizations, not just staff being unavailable?

Dr. Dhruv Khullar: It's a really interesting paradox in that in any individual person, particularly if you've been vaccinated, and definitely if you've had a booster shot, Omicron is less severe than delta or prior variants. The issue is that it's so much more contagious that even if it's a little bit less severe, it infects just so many people that is still overwhelming hospitals. You're right, that overwhelming in the hospitals is due to both staffs being out because a lot of staff has gotten the virus and most of them have mild infections, but they still have to be out and isolating, and just the sheer number of people that are getting infected.

If a variant half is virulent, half is damaging, but four or five times as infectious, obviously you can do the math there, you're going to end up with a lot more hospitalization. That seems to be the place that we're in right now during this surge.

Brian Lehrer: You note that by some informed estimates, the chances that an older vaccinated person will die from Omicron is no greater than that they'll die from the seasonal flu. Is that your experience at Weill Cornell? We're going to spend, I think, a fair amount of the rest of our time together talking about the implications of that because that would have huge implications for people's personal behavior choices as well as policy decisions.

Is that your experience at Weill Cornell, or do you think the data is clear enough on this yet that there's no greater risk that an older vaccinated person will die from Omicron than that they'll die from seasonal flu?

Dr. Dhruv Khullar: It's complicated. It is true that there are some estimates that if an older person is vaccinated, and particularly if they're boosted, that they face about the same risk of death from the coronavirus as compared to the flu, that is obviously very different for different people. If you are older and you have a set of chronic conditions, that place you at a high risk, that can be problematic if you are not vaccinated or not fully vaccinated, that can place you at a higher risk as well. That is on average.

If you're an older person who's been fully vaccinated, fully boosted, your individual risk seems to be about the same level of the few phases according to seasonal flu. The issue again is that you are more likely to get infected because Omicron is so much more infectious than prior various but also more infectious than the traditional flu.

Brian Lehrer: Listeners, your questions welcome about understanding the Omicron numbers and about the difference between a pandemic phase and an endemic phase of COVID that we might be approaching and the implications of that for personal choices and public policy. 212-433-WNYC, 212-433-9692 for Dr. Dhruv Khullar from Weill Cornell and New Yorker medical contributor. Maybe you saw his article this week, 212-433-9692. You can tweet a question @BrianLehrer.

Just to look at some of the most recent numbers in the context of what you're putting out there in this article, Dr. Dhruv Khullar. According to The New York Times COVID Tracker today, more than 2,000 people just yesterday, were reported to have died with COVID in the US. That daily average is up by 45% from two weeks ago. More than 200 died with COVID in New York State alone, a 30% increase from two weeks ago. We were getting close to zero deaths in New York State last summer. Do the death numbers tell you anything different from the hospitalization numbers?

Dr. Dhruv Khullar: The death numbers are tragic. A lot of those deaths, unfortunately, are preventable at this stage in the pandemic with vaccination and other precautions. It's not surprising that the death numbers have increased a month or six weeks ago, we're recording around 1,000 deaths a day. Now, that's up to 2,000, obviously, and that's really problematic. The reason for that is deaths often lag by three or four weeks of infections. We saw this huge surge in infections over the past month, and so the hospitalizations and the death numbers lag.

I think one thing that is important to note is that deaths have risen by 50% or 60% over the past couple of weeks, what we've seen four or five, six times the number of cases. There is some level of dissociation, by which I mean in the past, it was much more likely that each infection would send someone to the hospital and maybe lead to a really poor outcome like death.

Now, that seems to be decoupled a little bit in part because of vaccination, in part because of prior natural infections as well, but certainly, a time in which there's still 2,000 deaths a day from this virus is not one that we've reached the endemic phase yet, although I hope that that happens in the coming weeks and months.

Brian Lehrer: Well, if the hospitalization numbers overstate COVID issues in a certain way because many of those being hospitalized came in for other things and then are found to have asymptomatic COVID as you were describing before, do you think the death stats also include a lot of people who die with COVID based on positive tests, but not from COVID?

Dr. Dhruv Khullar: It's hard to tease that out fully in part because these aren't totally independent buckets. People when and for it sounds like an easy dichotomy. At the same time, people with chronic conditions, people can get a COVID infection that doesn't quite cause the death and it doesn't quite cause the hospitalization but it contributes and so how do you think about that? Someone who is in the hospital for another reason, that they're doing okay, but they get an infection and it tips them over. It's quite complicated actually.

I think that the death numbers are probably more accurate than the hospitalization numbers and certainly than the case numbers. I do think that the majority of those deaths are actually COVID deaths but there is some wiggle room as you described.

Brian Lehrer: Yes, and I guess that distinction was smaller before Omicron as well. If you died with COVID, you almost certainly died from COVID in the past. I ask this because Donald Trump back in 2020 and others were already saying the COVID deaths are overstated because of comorbidities that people had, his way of minimizing what the response to the virus should be. That was a lie, right?

Dr. Dhruv Khullar: That's correct. No matter however you tease this out and you say, "People had chronic conditions and that's why they're dying, it's not because of the virus." One way we know that that's actually not true is that the overall number of deaths in the United States increased substantially over the past two years. Even if you say, "A lot of those people were really sick and were going to die anyway," it wouldn't explain why there was a million more deaths over the past two years than you otherwise would have had. Clearly, there's something going on there with the virus.

Brian Lehrer: At the end of your article, you suggest that because of all the immunity the population has developing from vaccination or infection or both, we're probably headed soon for an endemic phase of COVID as opposed to the pandemic phase we've been in. People are hearing this more and more. What does that really mean?

Dr. Dhruv Khullar: Yes, this is a great question, Brian. No one can predict the future. I think every expert has had their share of missteps during the pandemic, but this is what people mean when they think we're reaching the endemic phase of the virus. The idea is that the vast majority of people in America will have some type of immunity by the end of this surge. That could be because you've been vaccinated and boosted, that could be because you've had an infection once or even twice. All those things count as giving both you, immunity, and also the population some level of immunity.

I think one thing that people often think about when they think about a new virus or a new pathogen is that the severity of illness is a feature of that bug. When in reality, it's actually a dance between our immune system and the pathogen. Even a pretty mild cold virus for us, if it's introduced into a new population on an island that doesn't have any immunity, it could be devastating. By the same token, a previously deadly pathogen in a population that now has built up immunity is no longer as deadly.

The idea is that in the coming weeks and months, we're going to get to a place where the virus still exists, it still circulates, it probably still infects people but it doesn't have these severe society stopping effects anymore and we enter into an endemic phase, something like what we had with the flu every year. It's not disruptive to society. Some significant number of people still get infected. Thousands of people may even die every year. I think people don't fully recognize that the flu kills 40,000 to 50,000 people a year as well. At the same time, most people can go about their daily lives.

Brian Lehrer: What about personal choices in that regard? Would the immunocompromized need to take different precautions for the rest of their lives than they took for seasonal flu before COVID?

Dr. Dhruv Khullar: Yes, it's hard to say exactly where things are going to shake out. People who have significant medical conditions, people who are on chemotherapy, people who have some level of immunocompromise, they were at significant risk before the pandemic and they will be at significant risk after the pandemic as well, where that risk, how much higher that risk is after COVID compared to before COVID, it's not entirely clear where that'll end up.

We can say that right now, a time where there's nearly a million cases a day and there's 150,000 Americans in the hospital, it's a substantially higher risk, but where that nets out in the coming months, I think it's hard to say, but taking precautions at this time is certainly I think advisable.

Brian Lehrer: When and if we get to the endemic phase, do you think the responsibility of public policy fades to whatever it was before just relative to seasonal flu? We'd still have a situation as I think you're describing where the virus is still circulating, in fact, kind of permanently and some people are still more susceptible to hospitalization or deaths than others. I'm curious if you think there would be any responsibility left for public policy?

Dr. Dhruv Khullar: Yes, I think there is a responsibility that we have as a society, as a public health establishment, as just neighbors and community members to one another to try to keep each other safe. These, over time, start to become questions that are not necessarily science-based but they're value-based. At the end of my article, I described when we reach this endemic phase, we're going to have to decide as a society either through politics or through our community what level of disease are we willing to accept.

We're not going to get rid of this virus but we're going to have to really have hard conversations about how high is too high and what level of restrictions would we accept to drive that number down. We have to have a conversation about what the purpose of restrictions are. It's not going to be to eliminate risk completely but to keep people reasonably safe and we should have clear metrics that we go by to ensure that restrictions aren't endless but they really have an end date that is driven by the evidence and by data.

Then, answering this broad existential question of what do we owe one another, how much do we deserve to engage in this conversation around, what can we do for one another to what extent should we restrict our behavior so that people who are in tougher positions are safer? Those are all questions that we've been dancing around I think for the past few months and years.

When we're at the surge, we're not going to be able to really think through those clearly but I think this surge will come to an end. Probably, we'll be in a much better place a month from now and we're really going to have to start thinking through these questions.

Brian Lehrer: Chris in Manhattan, you're on WNYC with Dr. Dhruv Khullar. Hi, Chris.

Chris: Hey, Brian. How are you? Thanks for taking my call. Dr. Khullar, I just have a quick question for you. I'm really fascinated. I'm actually a middle school teacher and I was looking at a really great data site on the Times this morning that has tons of this sort of data. One of the things we were talking about was hospitalization.

I'm curious if you have any thoughts about what's at the heart of the lack of distinction between being hospitalized with COVID, or being hospitalized because of COVID because those, as you said, are very different things and it seems like it could create some real hysteria. Is this just lazy recordkeeping? I'm curious about your thoughts. Thank you and I'll take the answer off the air. Thanks.

Brian Lehrer: Thanks, Chris.

Dr. Dhruv Khullar: Yes, it's a great question. One of the reasons is that in the past, as I said, with Alpha and Delta and these other variants, there wasn't a lot to think through here. Actually, if you had Delta and you were hospitalized, almost certainly it was because you had Delta. Omicron is different, I think for two reasons. One is that it's not as effective at replicating in the lungs. It stays up in the upper airways more so it doesn't send as many people to the hospital, to begin with.

Also, as I said, we have a lot more population immunity both through vaccination and infection. This is a new-ish problem that we're confronting that we weren't confronting before in the pandemic. Then, you add onto that the fact that it's not always really clear the extent to which COVID is contributing to their hospitalization. Someone might come in with heart failure or a different type of pneumonia and COVID positive. Is it along for the ride or is it really driving that hospitalization? It's not a hundred percent clear. It's not lazy recordkeeping, it's more that this is a new problem that we're dealing with and we need to develop a unified framework as well as a data collection practices that allow us to distinguish one type of hospitalization from another.

Brian Lehrer: Mo in Lynbrook, you're on WNYC with Dr. Khullar. Hi, Mo.

Mo: Hey, how you doing? Doctor, I have a question regarding the numbers. I had an Omicron over the holidays and then I got the flu on top of it and I called my doctor. My question is, are those numbers that information that's going to my primary physician, where does that go? Do they have the responsibility to tell a bigger entity of a person being sick? Do you know what I mean?

Dr. Dhruv Khullar: Yes. Now thanks for that question, Mo. I'm sorry that you had to deal with both the flu and the coronavirus at the same time. Those cases are all reported. When you are tested in a hospital, or a clinic, or a lab, those are all reported to the state and then those are reported out. Those numbers are we can take a lot of confidence in those. There's other numbers, and this is another reason that the overall statistics are getting a little bit hard to read, that come from rapid at-home tests.

As you might know a lot of people are taking these rapid at-home tests now, the antigen tests. There's no good way to collect that right now across the country. Some localities have tried to say if you take these tests report them here, but that's not required and it's not done everywhere certainly. We don't know how many rapid tests are being taken every day. We don't know what percent are coming back positive.

Even the 800,000 new cases a day number that has been out there for the past week or two, that's probably an under count because that's not including any of the rapid at-home tests that people are using.

Brian Lehrer: Which I guess would mean that the percentage of cases winding up in hospitalization is even lower than what's being reported in the official statistics which is not to minimize the number of cases that there are out there or the seriousness of them, but the percentage of hospitalizations would be even lower since everybody seems to be taking rapid at-home tests these days before they do one thing or another or after one kind of exposure or another. Probably rivaling the number of people are going to testing facilities or maybe not equal, but it's becoming a significant percentage. There's so much unreported COVID asymptomatic, right?

Dr. Dhruv Khullar: That's exactly right. We always had this problem where we didn't have enough tests and so there were a lot of infections that they never turned into reported cases. Now obviously, there's a lot of tests that people are taking at home. You're right the percentage of cases that result in hospitalization is probably lower than the official statistics indicate. That being said, of course, the absolute number of of hospitalizations is still very high and that is causing the strain on the healthcare system.

Brian: Or maybe the percentage of hospitalizations again though a tremendous amount of people getting hospitalized and dying throughout the pandemic. The percentage maybe was even lower than we ever knew because there might have been a lot of asymptomatic COVID out there that never or got diagnosed because people weren't using at-home tests as much yet. It's just a hypothetical, right?

Dr. Dhruv Khullar: That's correct.

Brian: Doesn't change anything. Question from a listener on Twitter. "If you get the Omicron variant, are you now immune to the Delta variant?"

Dr. Dhruv Khullar: That's a good question. It does seem to be a case that Omicron gives you very broad immunity going forward. If you got the Delta variant that didn't mean to a large extent that you were immune to the Omicron variant, but the reverse seems to be true. Now, that you have the Omicron variant, people seem to have pretty good immunity against the Delta variant, which is now counting for a smaller and smaller number of cases in the United States.

Of course, we don't know what comes next. We all hope that we're kind of getting to the end of this and that the future variance will have harder and harder time kind of getting around in multiple layers of immunity that we've developed over the past couple of years. We don't know that for sure, but we do that Omicron gives you pretty good protection against Delta.

Brian: Related question I guess also on Twitter listener asks, "What does the doctor think is the likelihood of a new variant arising once Omicron is under control?"

Dr. Dhruv Khullar: I think the likelihood of a new variant is almost 100%. The question is, will that variant cause big issues? We know that the virus is constantly mutating. There will be new variants that is going to happen. A lot of the variants may not cause significant infection. They may not be able to get around preexisting immunity. They may not lead to a lot of hospitalization at death. In that case, the variants are not as important. That is our hope. We don't know that to be the case 100%. Of course.

No one can predict the future, but as I said, I think Omicron has provided-- it's been a tough way to get there, but a very broad base of immunity that we hope will hold up against future variants.

Brian: Gregory in North Bergen, you're on WNYC. Hi, Gregory.

Gregory: Hi, how are you? Great program. My father is 101. He tested positive a couple of days ago. He's doing more of his pretty well, but this morning, he's on oxygen level went down to 94 from 97. We're thinking to seek monoclonal therapy. Is there any adverse reaction to looking for the therapy for a 101-year old person?

Dr. Dhruv Khullar: I hesitate to give medical advice without having seen someone but I think certainly what you're describing, it would make sense for him to get monoclonal therapy if his oxygen levels are dipping to go to a hospital. There are these new COVID pills, these antiviral pills which are very effective for particular this one called Paxlovid from Pfizer. Supply is quite limited in New York, but it is possible to get those pills which are even easier to take obviously than the infusion that you need for a monoclonal, but certainly, if he's having trouble breathing and his oxygen levels are dropping, I would seek medical attention immediately.

Brian Lehrer: Gregory, I hope that's helpful and I hope your father is okay. I guess what you just said brings up something else that you write about in your article, which is the ethically daunting challenge of having to decide in some cases who gets the treatments at all if there aren't enough to go around like that new pill is in short supply. Is there anything you can say about what kinds of guidelines, risk factors, or potential years of life lost, or anything, you or others in the profession used to make those decisions that we hope never to have to make?

Dr. Dhruv Khullar: It's a really tough set of ethical questions that come up when there's a real scarcity of resources whether it's medications, or ICU beds, or ventilators, and we hope to avoid those types of decisions at all costs. In terms of the COVID pills, those are I think generally being given to people who have at least one or two risk factors for progressing to severe disease. If you're a relatively young person who's been healthy, given the scarcity of those pills at this time, those are not given in those circumstances.

Some hospitals and some states are having to enact what you're alluding to which is called crisis standards of care. In those cases, basically, decisions are having to be made about who is most likely to benefit from a scarce resource, and those are really heart-wrenching decisions that doctors are having to make.

Fortunately, I think for many doctors, it's been pretty rare that they've had to engage in those decisions, but as the healthcare system is overwhelmed, that's the type of thing that people are having to consider and is one reason that we really need to do what we can to prevent hospitals from having a huge influx of patients at the same time.

Brian Lehrer: Last call, doctor, and I'll say that at least three listeners at the same time are chiming in to ask this question. This is one that I get every day now that we do a COVID segment even though I think you're going to say there's no way to know the answer. Dave in Brooklyn, you're on WNYC with Dr. Dhruv Khullar. Hi, Dave.

Dave: Oh, thank you. The one thing which bothers me in this COVID discussion is no one says anything about long COVID. We know that even in mild cases that 10% to 20% of people will develop the long COVID and we're talking about potentially means of people in this country being disabled. When it becomes endemic, would that could be added maybe tens of thousands more each year? The COVID is not the first coronavirus to infect people. We know in other ones, the original SARS, there are people who were disabled since they got SARS decades ago before.

Brian Lehrer: Dave, thank you. He didn't ask the exact question I thought he was going to ask. Maybe the one he actually asked you can answer, Dr. Khullar.

Dr. Dhruv Khullar: Long COVID is still something that we're trying to learn a lot about. As you said, there are a lot of cases of COVID in the country and a substantial portion of those times. People have lingering effects of the virus. A lot of people describe things along the lines of fatigue, trouble breathing, and decreased ability to exercise. The NIH, a few months ago, invested a billion dollars in research laboratories around the country to try to understand this condition further. We know that other viruses can lead to lingering symptoms as well.

We hope that some of this research sheds light on the impact that other viruses have also had on people's lives over the years. I think we'll know a lot more in the next year or two on the real long-term effects of COVID.

Brian Lehrer: Well, two follow-ups then. First, the one I thought he was going to ask is, "Is Omicron producing long COVID?" I assume the answer is by definition, Omicron is so new, we can't know that yet, but is that your answer?

Dr. Dhruv Khullar: Yes, I think that's right. I think we don't know the long COVID profile of each of the different variants. Omicron, obviously, really started to spread about, six weeks, eight weeks ago. It's hard to say exactly. We do know that there's good evidence that people who have been vaccinated and then get infected, they're much less likely to continue to endorse symptoms weeks and months after. There's a lot we don't know but that is a way to reduce your chances of having those lingering effects.

Brian Lehrer: Last follow-up on this, listener tweets. "Does long-COVID necessitate the rewriting of the Americans with Disabilities Act?"

Dr. Dhruv Khullar: It's another really good question. I think as we learn more about the condition, we need to think about how we ensure that people who are still struggling with the after-effects of infection, how they have the support that they need. It may be tapping into the Americans with Disabilities Act or may be other mechanisms that states and localities choose to use to support people through that process but I think we'll know a lot more in the coming months.

Brian Lehrer: Dr. Dhruv Khullar, physician and professor of medicine at Weill Cornell and medical contributor to The New Yorker. His latest article is called Do the Omicron Numbers Mean What We Think They Mean? So informative, doctor. Thank you for answering my questions and listeners' questions. We really, really appreciate it.

Dr. Dhruv Khullar: Thanks so much for having me.

 

Copyright © 2022 New York Public Radio. All rights reserved. Visit our website terms of use at www.wnyc.org for further information.

New York Public Radio transcripts are created on a rush deadline, often by contractors. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of New York Public Radio’s programming is the audio record.