You’re listening to a press conference from the Harvard T.H. Chan School of Public Health with Stephen Kissler, research fellow of the Department of Immunology and Infectious Diseases. This call was recorded at 11:30 a.m. Eastern Time on Friday, August 14th.
Transcript
MODERATOR: Dr. Kissler, do you have any opening remarks?
STEPHEN KISSLER: Yeah. So I think a lot of different things going on right now, of course, the opening of schools is on the forefront of many of our minds. And really, what can be done, what should be done and what are the different considerations there? Cases seem to be varying in different places across the United States. Many of the states where we’ve seen substantial rises seem to have been able to turn some things around, whereas others are starting to potentially emerge, including here in Massachusetts, we’ve had potentially some small rises in cases as well. So we can definitely speak to that. And then some news from Russia, of course, with the vaccine that they’ve implemented for the population. Probably not something I can speak about in a whole lot of detail, but definitely something we could discuss if anyone is interested to some extent. I’m happy to take questions on those or other things. And I think I’ll just turn it over to everyone who’s here, see what you’re interested in. Thanks.
MODERATOR: All right. Looks like we have a couple of questions already. So first question.
Q: Hi, Dr. Kissler. How are you?
STEPHEN KISSLER: I’m doing well.
Q: Good. So I’m looking at the COVID-19 vaccine trials. That’s the context for my question. And I was surprised to hear a Dr. Fauci, the other day, say that the FDA standard for efficacy is 50 percent or greater. It kind of got me thinking a little bit more about the implications of that. So if you could comment on that in the context first, of other vaccines such as influenza, and then any implications for herd immunity or along those lines?
STEPHEN KISSLER: There’s a lot to unpack there. So maybe I can start by talking about the ways that we think about vaccine efficacy. I think the efficacy that Dr. Fauci was referring to was efficacy in preventing illness, in preventing clinical illness. I imagine that part of the evidence base that he was using to come up with that 50 percent number was the influenza vaccine, which varies in efficacy from year to year, depending on how close the match is between the strains that are included in the vaccine and the strain that’s actually circulating. And that efficacy speaking a bit off the top of my head, but rough ballpark, can range anywhere from about 40 to 80 percent. So 50 percent sort of falls within that ballpark and I think is a realistic efficacy goal. Certainly, we would like to have something that has a higher efficacy than that. But there’s evidence that vaccines that have on the order of 50 percent efficacy are still very useful as public health tools. So I think that that’s a sensible mark, basically, for something that would be helpful, while also being realistic. So basically, not being too stringent in approving the vaccine. Now, of course, there are a couple of different levels of efficacy.
First of all, efficacy can vary depending on who’s getting the vaccine. For example, we know for sure that for a lot of vaccines, the efficacy reduces as age increases. And so that’s one of the things that’s roped into this, is that 50 percent efficacy will be an average. That includes probably a lower efficacy in older age groups, but a correspondingly higher efficacy at younger age groups. Again, depending on the vaccine platform. So there’s some variation, though, that’s worth paying attention to. And also, there are differences in efficacy. So we normally think about efficacy in preventing illness. But there’s also a very important efficacy in terms of preventing the spread of disease, preventing transmission. It’s possible to prevent illness without necessarily doing much to slow down transmission. So if we have a vaccine that does that, that won’t really get us much closer to herd immunity. It will only sort of prevent illness and people who might be infected, which is a great thing. But if we actually want to use a vaccine to get us towards herd immunity, we also need a vaccine that has efficacy towards blocking transmission. I believe that many of the vaccine trials that we’re using now are trying to measure both of those types of efficacy, so we’ll just have to see. But that’s normally how we’re thinking about it. But that said, I think that benchmark for 50 percent clinical efficacy is both a realistic one and one that would absolutely be helpful in the real world.
Q: Right. That’s the FDA standard I read. Thank you.
STEPHEN KISSLER: Thanks.
MODERATOR: Next question.
Q: Thank you Nicole and Dr. Kissler. I’m wondering what you think the relationship is between the current rise in child cases that we’re seeing throughout the country in testing? Children were largely kept out of school, especially in Massachusetts once the outbreak began. So are more children now contracting the virus or are more children now just being tested for the virus?
STEPHEN KISSLER: So my understanding is that it is it’s a little bit of both. So I think that as schools are considering going back, some schools are encouraging their students to be tested. I think some parents are also probably voluntarily having their kids tested. So I think that part of it is just sort of a behavioral increase in testing. But I would also imagine that as schools begin to open, as activities begin to increase prior to the school term and now even some schools are beginning to open up, that that is probably contributing to an increase in cases among kids. So I think it’s both. As far as I can tell, I’m encouraged by the fact that there is an increase in testing among kids because that will just help us maintain surveillance. But I do think the story is probably an element in both.
Q: Sort of a slightly different topic, but I was wondering, still on children in school, if you had any thoughts on whether this hybrid model that most school districts in Massachusetts are opting for is all that much safer than a full return? Why or why not?
STEPHEN KISSLER: Yeah, so in my mind, that depends very much on the age group of the students that we’re talking about, let me start with that. One of the things that could basically compromise the effectiveness of a hybrid model in a younger age groups, is that younger age groups are probably going to need childcare when they’re not at school. So by adopting a hybrid model, you may end up with a scenario where children are interacting with one group of students within their school, and then interacting with a different group of students at their daycares. And that’s not great because you really want to limit the number of unique individuals who a child is interacting with, limit spread. Now, on the other hand, for older age groups where older kids might not need external childcare, I think that a hybrid model could make a lot of sense because it would de-densify the school, basically limit the number of people they’re interacting with and the opportunities for transmission. So I could see a hybrid model potentially working very well for older age groups, for younger age groups it really depends on what the kids are doing when they’re not at school.
Q: Thank you.
MODERATOR: Next question.
Q: Hi. Thanks for doing this call. I had a question about contact tracing. It seems like there’s a lot of states that are either not reporting the same information or the information varies from state to state with what they’re tracking and even what they’re reporting. Now, I’m wondering if you can talk a little bit about what the implications of that are, and if that’s going to be problematic in terms of being able to find out where the cases are and where they’re coming from?
STEPHEN KISSLER: Yes. So I will say up front that I’m not intimately familiar with the different types of contact tracing related data that the different states are publishing. I’ve more been following sort of raw testing numbers and case counts in those sorts of things. That said, I think that, certainly, contact tracing is really important for understanding where transmission is happening, and that will probably vary from state to state and actually probably even more profoundly, just from neighborhood to neighborhood. Basically, what types of activities, what types of settings are more important to spread, I imagine will vary quite a bit from place to place. Much of the research that’s come out looking at geographic disparities in COVID prevalence basically suggests that there can be huge differences in infection rates between very, very proximal neighborhoods. And that suggests just different dynamic spread between the two. So I think that the fact that different states and even different communities are doing contact tracing differently and publishing their their findings differently is not necessarily a huge cause for alarm, given that the interventions that we’re going to need to adopt will also be similarly very local. It just matters that they’re collecting the sort of information that’s relevant to their local community fundamentally. And I’m not entirely sure how to assess that, whether that’s actually happening. But I do hope that one of the most important things, I think, in this outbreak moving forward is to have very locally tailored responses. So my hope is that the local public health agencies are asking these sorts of questions and doing these sorts of follow ups that are relevant to their specific communities. And as long as that’s happening, I have faith that contact tracing will help reduce the spread, and will also, importantly, give us information about where spread is happening.
Q: Thanks. And just to follow up then, are there places right now where the spread is just too great for any type of contact tracing to be effective?
STEPHEN KISSLER: Yes. I mean, it depends what your intention for the contact tracing is. So contact tracing will always help to prevent some number of transmissions for sure. But I think that many parts of the country, especially outside of the Northeast, currently do, simply have too many cases to use contact tracing as essentially the primary public health measure to control cases. It’s just not enough. We just don’t have enough resources, and in a lot of these places, enough contact tracers to follow up on all of the case. So if you’re using it as a fundamentally transmission limiting strategy, we have too many cases in most parts of the country for that. But if you’re using it to just gather information about where transmission is happening, then I think it can still be effective. So it’s not something that we should categorically not be doing. But it’s also not something that we can rely on solely to save us from ongoing transmission for sure.
Q: Thanks.
MODERATOR: Next question.
Q: My question is about the weather. We heard, I think, from the Children’s Hospital of Philadelphia this week that Boston will likely be a hotspot when it gets colder this fall. How much do we really know about the factors of the virus and weather? Is it just people congregating inside more? Is it really that warmer weather and sunlight kills the virus or knocks down transmission? Or are we just making some assumptions and science doesn’t know how much the weather is a factor?
STEPHEN KISSLER: Yeah. So for specifically for SARS-CoV-2, we still don’t know very well. There have been a number of studies that have tried to address exactly this question and they seem to be mixed. Some of them suggest that there might be influence from weather and some of them are just sort of inconclusive. So that said, we do know that a lot of respiratory illnesses and especially the common coronaviruses, the causes that tend to cause the common cold, do follow a very strictly seasonal pattern with spikes in the winter in the temperate regions of the globe. For flu, we’re pretty sure that that’s not just due to human behavior. Human behavior congregating indoors probably contributes. But part of it is that actually the colder and drier weather in the winter actually sort of helps the virus to sort of spread further in the air, in a sense, to sort of linger in the atmosphere in a way that warmer, wetter weather in the summer sort of helps prohibit coronaviruses and flu share transmission routes to a large extent.
We think that droplets spread and to some extent surface based spread can be important for both. And so I think it would be reasonable to assume that both changes in behavior, but also potentially changes in the weather would contribute to the spread of SARS-CoV-2, just based on what it seems to do for flu and also probably for the related coronaviruses. But at this point, it’s hard to tell. The only other thing that I’ll note with respect to this, is that one of the things that could limit spread is if there was enough immunity in the population to prevent a spike in infections from happening. But based on serological studies, it seems like there are still plenty of people who are susceptible to SARS-CoV-2 as we move into the fall. So I think that’s congregating indoors, having schools open up, and then potentially shifts in the weather, are all things that could contribute, and it’s hard to unpick which of those is most important. All of them are pretty unsatisfying to answer your question. But I think that with all of these things together, I do expect transmission of SARS-CoV-2 to be a little bit more difficult in the winter than it was in the summer. So that’s what I would say.
Q: If I may. I’m going to ask a question about the positivity rate. There’s been some confusion or distinctions between different states about how the positivity rate is calculated. Rhode Island was using, or is still using, the total number of people who test positive over total tests. In Massachusetts until this week, was using people who test positive over just the total number of people tested. So they weren’t counting repeat tests. So which calculation is better? And why don’t we have a uniform calculation for positivity rate across the country?
STEPHEN KISSLER: Yeah, it’s a great question. I think that both have their merits. I find the positivity rate in which the number of positive tests is divided by the number of people tested to be a little bit more intuitive. This is largely because there can be an issue with false negatives for these tests, so sometimes the person who is presumed to have SARS-CoV-2 might be tested a couple of times with the first test comes back negative just to verify. And then if that test comes back positive, it doesn’t really seem to make sense to me to include all of the negative tests before because we don’t really trust them as accurate necessarily so. So I think that dividing by the total number of people who are tested seems to make a little bit more intuitive sense for me. But I think the most important thing is that to the extent we’re able, we just stay consistent, because what I think the real value of the percent positive testing is to actually identify trends in cases whether we’re increasing or decreasing. And it can give us a rough sense of sort of the rough prevalence in the population. But since different testing protocols between different states are varying so much, for example, whether or not we’re testing employees, whether or not we’re regularly testing people going to schools and sorts of things. It’s really difficult to compare these numbers across states anyway, regardless of which of those choices you’re making. So I think that the best practice to be back to publish both. If you’re not going to publish both, at least try to stay consistent with what you are publishing so that you can compare week to week. And then aside from that, I think just publishing as much information as you can about who you’re testing, when and why, is the only way that we can really interpret these data.
MODERATOR: Next question.
Q: I’ve got a couple of questions about the ways that we’re measuring the spread of the virus and counting the number of people killed by it. My first question is, do you think cities and towns have accurate estimates right now of how many people have been killed by the disease or is there potential for a significant undercounting here?
STEPHEN KISSLER: I do think that there’s potential for a significant undercounting. It’s difficult to say, though. I mean, I know that there was recently this report in The New York Times that suggests basically that the mortality over the last few months has largely followed the case counts. But that seems to be somewhat higher than the mortality that was expected, or the direct mortality that’s been measured and directly attributed to COVID. So I do think that there’s definitely a potential for undercounting. And I think that this is a common phenomenon across the surveillance of many infectious diseases. You know, the mortality counts for flu are often adjusted to account for exactly this, that deaths that are directly attributed to flu are actually probably an undercount, and you sort of have to extrapolate upward from that to really try to get a true count. But, of course, some of the mortality could also be due to other things, including people maybe hesitant to go to the doctor because of COVID. And do you count that as a COVID associated death or not? I think it really depends on what you’re trying to count there. So for that, I’m referring to things like potentially even from like heart attacks or something. And it’s really difficult to document this. I don’t have any hard data on it, but I can just sort of intuitively imagine that people might be more hesitant to go to the doctor and might not catch some severe illnesses as early as they might otherwise if there weren’t a pandemic happening. So I think both of those are in play.
Q: Thanks. Are there any other common causes of death that are related to the pandemic but aren’t directly caused by the virus? You mentioned people may be neglecting medical care they need. Anything else that you think is something we should be paying attention to?
STEPHEN KISSLER: That’s really the main one. And I mean, I could speculate on others, but it’s not an area that I really looked in too much, so I don’t want to say anything that I’m not particularly sure about. So I think it would be neglecting medical care would be the main what I’d be paying attention to now, because that’s what I’ve heard about most.
Q: And lastly, can a randomized serology survey give a community a reasonable estimate of what share of its residents have been infected?
STEPHEN KISSLER: Yes, it can. As long as it’s done responsibly. So, again, like you said, if it’s randomized and randomized sensibly, you’re really doing best effort to get a like truly a demographically, geographically, representative random sample. Again, the serological tests can suffer from, in this case, low specificity oftentimes. So you can have some false positives. And so depending on what the actual prevalence is in the population, that can sort of adjust how accurate your findings are too. So you really need to sort of make sure that you’re doing your statistical due diligence and reporting these findings as well. But that said, absolutely serological tests are probably the best way we have to identify the cumulative prevalence, the total number of people who have been infected in any given population. And I think it’s something that we ought to be doing.
Q: And last question. What are the barriers preventing communities from using randomized serology testing? I haven’t seen that many use that in Massachusetts yet.
STEPHEN KISSLER: Yeah, I mean, I think some of it is just cost and logistics. I think that there is a sense that it’s much more urgent to try to understand what’s happening precisely now rather than what might have happened in the past. And so I think that there’s sort of relatively more emphasis being placed on how do we open schools safely? How do we make sure we’re ramping up our basic virological testing capacity? And so I think part of it is just that it hasn’t quite risen to the level of priority that we would like. So I think it’s mainly just that.
Q: Thank you.
MODERATOR: Next question.
Q: Hi. Thank you so much for taking my question. My question is about the daily tests conducted nationally. We’re now averaging about 724,000 tests per day, according to the COVID tracking project, which is down from July. And the Trump administration testing tsar said yesterday, that that is the appropriate amount of testing and that we don’t need to be doing more. But estimates from the Harvard Global Health Institute and others have suggested that we need up to four million tests per day. So gut check from you. Is our testing today adequate? Do we need more or less testing than we are doing now?
STEPHEN KISSLER: I would say we absolutely need more testing, that the more testing we do, the better. We still have some blind spots as to where transmission is happening. We’re a lot better off than we were the spring. But you’re right. I was also looking at the COVID tracking project testing data just this morning, and it really does seem like there was a rise in testing, but it’s really been falling off lately. And I really have trouble understanding why that’s the case. Our testing capacity really should be increasing because frankly, in my opinion, we’re still such a long way off from having the amount of testing that we would need to identify people when they’re potentially infectious so that they can adjust their behavior and prevent from infecting other people. That’s really what we want to use testing for. It doesn’t do us an awful lot of good if we test someone only after they’ve been infectious for a while or for missing really a large number of people who are infectious. And at this point, with the positivity rates that we’re seeing in many states, we really do need to be doing more testing. It’s clear that we must be missing people who are infectious, who are not getting tests. So, yeah, unequivocally we can do more and we really should be doing more.
Q: And as a follow up, what is a good number? I’ve seen the four million from Harvard Global Health Institute and the Rockefeller Foundation that says four million tests a day just sort of get cases to zero. Others have said that’s a little ambitious. Like in your mind, it would be a good number of daily tests nationally?
STEPHEN KISSLER: So I think that one of the guidelines that I’ve heard that I think makes some good sense actually has to do with thinking about what the positivity is. And ideally, we’d be doing enough testing to keep the positivity somewhere down below around five percent. Clearly, that’s not happening in a lot of different places. So that’s roughly the benchmark that I would use because that would suggest that there’s enough comprehensive testing going on, that we’re not only picking up people after they presented the care, after they’re symptomatic. So, yeah, that’s the benchmark I would use. Absolutely.
Q: Thank you.
STEPHEN KISSLER: Thanks.
MODERATOR: All right. Next question.
Q: I was just going to follow back up since no one else has their hand raised right now, what I asked earlier about child cases, what might be some of the reasons that children are now contracting COVID more? If it is not just that it’s sort of more testing being done. Is it just that kids are out in public more? You know, obviously, Massachusetts kids aren’t back in school yet, so they aren’t in public more since the economic reopening is taking place. Or, you know, what might be some of those reasons that children are contracting it more now?
STEPHEN KISSLER: Yeah, I mean, I think it’s probably that there are just more opportunities for spread. I know that there’s often a lot of activity that needs to happen prior to a school getting opened. So before students actually go back to class, there can still be opportunities for interpersonal mixing among families who might be sending their kids back. So the school start date is not necessarily sort of like a hard marker of when the mixing that’s related to school begins happening, it probably bleeds a little bit forward in time as well. And it’s also not just the kids. It would be their parents and the parents of the people with whom they’re interacting as well and anybody who’s, helping to care for them, that sort of thing. So I know we’ve seen, definitely, increases in cases among young adults in many places. And so part of the rise in kids might also just be sort of the percolation of spreads to kids, naturally to be expected by rising case numbers in the community in general. So I think all of those things might be contributing.
Q: Thank you.
MODERATOR: Next question.
Q: If we can revisit herd diminished herd immunity concept. I was reading that up to 50 percent of Americans may not get the COVID-19 vaccine when it’s available. So what are the implications? I thought we wanted 60 to 70 percent herd immunity, so the numbers aren’t quite adding up.
STEPHEN KISSLER: Yeah, you’re right. It’s definitely concerning. My hope is that, you know, if and when we do have an effective vaccine, that the number of people who are willing to get it increases if they see other people getting it. And those people being fine, hopefully the confidence in the vaccine will increase over time. I am hopeful about that. But you’re right. I mean, if there is a vaccine refusal at that rate, then you need a proportionally more effective vaccine to sort of account for the vaccine refusal as well, if you want to bump up herd immunity, so that makes the technological challenges for the vaccine that much more difficult. It means that we might not be able to rely on a vaccine, even if it does block transmission to actually give us immunity from from SARS-CoV-2. So I think that the fact that such a low number of people seem to be willing to get the vaccine is definitely alarming and definitely something we’ll have to pay attention to.
The other thing is with vaccine refusal, it’s often concentrated in geographic pockets as well. So that 50 percent who or whatever it may be, but we’ll call it 50 percent who refuse the vaccine won’t be evenly distributed across the population. And so what that means is that we’ll probably have neighborhoods that have much higher numbers of people who are not vaccinated and don’t have that immunity. And when that happens, that sort of makes these particular communities even more susceptible to spread. That’s something we’ve seen with measles outbreaks, for example, with vaccine refusal for measles, where, if the vaccination were sort of evenly distributed across the U.S., we still wouldn’t have a problem. But there are specific communities where vaccination rates are low and those are the ones where we tend to see these resurgences of measles outbreaks. So I think we can probably expect to see the same kinds of dynamics happening with SARS-CoV-2. And then lastly, it’s where we’ve had a flu vaccine for a very long time, but that still hasn’t led to the eradication of flu either. So. And part of that is due to vaccine efficacy. Part of that is due to vaccine uptake. Part of that is just due to the fact that our immunity to flu doesn’t last forever. And it seems like all of those things might be in play for SARS-CoV-2 as well. So I think it’s also worth bearing in mind that for all of these reasons, having a vaccine does not necessarily mark an end of the pandemic or an end to our having to deal with SARS-CoV-2. It’ll help a ton. Absolutely. I’m convinced of that. But it’ll just be the beginning of the end, I think.
Q: So it sounds like we may have to live with it longer and maybe even still take some precautions.
STEPHEN KISSLER: Yes, that’s right. And I think flu offers a very good example of how this kind of thing works. The flu strains that caused pandemics over the past century did enter into regular wintertime circulation after the fact. So I think there’s a very good historical precedents for viruses that cause pandemics then entering into seasonal circulation after the fact. And I don’t see any reason to think that that wouldn’t happen for SARS-CoV-2, too, as well. So we’ll just have to see. You know, it could be that with increasing exposure to SARS-CoV-2, even if our immunity declines over time, whatever immunity we do keep will help prevent the illness from being as severe in the future. So that might sort of make our future experience with SARS-CoV-2 to a little bit less severe than what we have now. But again, it’s just too soon to say. But I do think that there’s good historical precedent for viruses like this entering into pretty regular circulation in the population. And I think that that’s a very likely scenario for SARS-CoV-2 as well.
Q: Thank you.
MODERATOR: It looks like we have a question.
Q: Hi. Thanks so much for doing this yet again. I was wondering if you could talk a little bit about distribution. Are you concerned at all, once we have a vaccine, how it’s going to get out there?
STEPHEN KISSLER: Yes, I am. And concerned on a couple of different levels. I mean, there’s the problem of international allocation and then there’s also the problem of within country allocation. So there’s a lot of modeling work being done. There’s a lot of historical work, looking at the previous distribution of vaccines, to try to understand how we can do this best. But I think it is a cause for concern because there’s so many factors that play into this beyond just the just the public health implications. There’s a lot of politics. There is, who pays for it? How do we pay for it? And so it does absolutely concern me. I mean, I’m concerned in particular that low and middle income countries might be sort of second or late to get the vaccine simply because that’s just not prioritized, just due to potentially having more difficulties to pay for it. And that’s, of course, not a good thing for the countries themselves. But then also, you know, if anywhere in the world is left unprotected, then that puts the entire world at risk. You know, we’re all very interconnected here. So we might end up with a very suboptimal allocation strategy if it’s left solely to the economics to decide who gets it. And then within countries. I guess you can imagine a similar scenario playing out, historically speaking. Usually health care workers are among the first to get the vaccine and then after that, those at highest risk from severe outcomes from the infection. And based on some of the modeling work that we and others have done, that seems like it’s probably a sensible thing to do with an upcoming SARS-CoV-2 vaccine as well. But it takes an awful lot of administrative oversight and planning to make sure that that is implemented fairly and equitably. And I worry that will actually happen. So certainly on the public health side, we’re doing our best to generate those guidelines. And I think only time will tell if they’re actually implemented.
Q: The CDC traditionally has has played that role, although I guess the military has been mentioned this time? Does CDC still have the capacity? Did we cut CDC so much that they don’t have the capacity to do this anymore?
STEPHEN KISSLER: I worry that we might have. I think that their capacity certainly has been sort of intentionally limited over the last couple of years. And I think that that’s placed us at higher risk. And to some extent, also, I think just trust in the institution has has declined, which you know so much about the allocation of vaccines. And again, going back to the vaccine uptake question that came before, are you willing to get the vaccine, really depends on trust in our institutions and in our public health authorities. And I think that that’s really been eroded, unfortunately, over the course of the pandemic and even prior to that. So, yeah, I think that we’re in a difficult spot for sure. I don’t think that the CDC is is incapable of this. I do think that they could and will play an important role in the vaccine allocation, but definitely their job has been made harder.
MODERATOR: That looks like it may have been our last question for today. Dr. Kissler, do you have any final thoughts to share with us?
STEPHEN KISSLER: No, nothing to add. Thanks for all the questions and thanks for being here today.
This concludes the August 14th press conference.