You’re listening to a press conference from the Harvard T.H. Chan School of Public Health with Barry Bloom, the Joan L. And Julius H. Jacobson Research Professor of Public Health, and William Hanage, associate professor of epidemiology and a faculty member of the Center for Communicable Disease Dynamics. This call was recorded at 11:30 a.m. Eastern Time on Thursday, August 6th.
MODERATOR: Dr. Bloom, do you have any opening comments?
BARRY BLOOM: No, I look forward to questions. It’s an exciting time. I would also say today is an anniversary of the dropping of the atomic bomb on Hiroshima, and we should have in mind the misuse of science as well as the use of science.
MODERATOR: Thank you, Dr. Bloom. And Dr. Hanage, do you have any opening remarks?
BILL HANAGE: I certainly agree with what Barry was saying about the anniversary of the dropping of the atomic bomb on Hiroshima. I will say briefly as an opening for what’s going on in the pandemic, while cases seem to have peaked across the South, slow and limited testing may obscure the full picture, and we should not expect a long plateau, assuming things not change. But we are also seeing some clear upticks in a number of other states, including some of the Midwest, but also New Jersey and Massachusetts, albeit from a very low level, which experienced strong surges in the spring. This suggests that we have to be pretty cautious in how we go forward if we wish to maintain low levels of community transmission. And with that, I’ll turn it over. I look forward to the questions.
MODERATOR: Thank you, Dr. Hanage. Looks like our first question.
Q: Hey, thanks for doing this. I had a question about vaccines. The president keeps saying there will be a vaccine very soon, and says something about this fall. Just wanted to get your thoughts on given the political realities of this year. November 3rd being an important day. How concerned are you that there’ll be political pressure to cut corners or speed up the timeline? And feel free to differentiate between widespread distribution and emergency use authorization. Thank you.
BARRY BLOOM: Thanks for a really great question that has really absorbed many of us in the scientific community. It is likely that phase three studies of safety and efficacy, with about 30,000 people enrolled in each with a placebo control, will be completed for sure by the end of the year. There is a race, as you know, to try to get there early. There has been a concern that it would be used without full analysis of safety data and efficacy data to say the epidemic is so bad that we’ll just give it an emergency use authorization and we can get it out by November 3rd. I know of no specific plans that anybody has thought about doing that within the administration, but certainly that was a concern. So let me break that concern in several pieces.
The first piece is that the trial designs are really thoughtfully done, which gave a great relief to many of the scientific people who were involved in vaccines. The way you do a trial design is to figure out how to get statistical power as to whether there is a statistically significant difference between the vaccine recipients and those with a placebo. That requires knowing what the incidence of the infection is and disease is within communities where the vaccine is being tested. For example, China had a study done of a whole virus killed vaccine that they had to stop because they ran out of cases and they couldn’t get statistical power. So the design created by a committee with membership of the NIH, headed by the director of NIH, headed by the vaccine program with people from FDA, as well as NIH, is a really rigorous, thoughtful standard, critical design to get statistical power to the best that one can estimate what the prevalence will be at any time while the trial is ongoing to tell us if the vaccine is 50 percent or better protected. Looking at thirty thousand people over the time of that trial and what the adverse effects are and making an evaluation, whether they are serious enough to recommend that they not be given permission to distribute the vaccine.
The one thing that is not involved in the planning of a vaccine trial this time. It is not a variable. The variable is statistical significance. If you can run, instead of 10,000 people in a year, 30,000 people in three months, you get the same statistical power. And what you gain, in the way this is being done, is presumably a rigorous analysis of the efficacy and safety in 30,000 people over that period of time. What you don’t get in that timeframe is any indication of longer term adverse effects. So two points on adverse effects. As far as I know, there is no vaccine, in fact no medical intervention, of which I’m aware that doesn’t have adverse effects. The adverse effects seen in the phase one and two studies for the vaccines moving forward are standard. The same as a flu vaccine, which is kind of standard sore arm fever, muscle pain, weakness, almost all of which dissipates in 24 to 48 hours. What we worry about is uncommon or rare adverse effects, and they often don’t show up until you have very large numbers of people. So the scientific community, as you know, has written letters and spoken to the press. We want to be sure that all the data are analyzed and characterized and that the public is informed of what we know for each of the vaccines of what the efficacy is in fifteen to thirty thousand people that will get the vaccine. And what we know about adverse effects over that period of time. That’s been the concern, and I know of no evidence that that will not be followed. However, there are two concerns that cause those letters to be written. The FDA has within its authority, emergency use authorization, which has several committees within the FDA required to review it. But in the case of Hydroxychloroquine, somehow, mysteriously, with no documentation or evidential basis, the director of the FDA gave emergency use approval and that caused a great deal of confusion both in the medical community and the public and some distrust of the whole enterprise, packed in with a distrust of what government does in any case.
The second case, which I found equally troubling, is when a whole slew of antibody tests came in over a hundred. The FDA simply took the companies, mostly from China’s analysis of their specificity and sensitivity and never checked on them. And as you know, they had to rescind the emergency use for approvals for about 90 or 88 out of 100, so that only 12 were approved as being within the range that would even be credible to collect meaningful data. So with 0 for 2 on those two major issues, there is reason for skepticism that there could be an October surprise. Without public data, without full analysis, the director of FDA could simply say it’s a bad problem. I give it emergency use and we’ll collect data. That would really, I think, compromise what little trust we have in this country about A, vaccines in general, and B, the government’s controlling or regulating vaccines to protect the public interests. Those are the concerns. I’m very hopeful that they, having been published, having been in the public news, will not occur.
The second issue is, once those vaccines become available, after they have hopefully real approvals, after detailed analysis of the data by the FDA and made those public and adverse effects. The big challenge will be who gets it first, and I think an extraordinary development, unprecedented in my knowledge, has occurred in that. That is, there has been an agency that is responsible for recommendations for vaccines, a little known agency reporting to the CDC called the Advisory Committee on Immunization Practices. They look at the dosage. They look at the timing. They look at the interference of one vaccine in another. They look at new vaccines and they make recommendations of, when they are limited or a company has trouble, who would get them. It would be ordinarily the case that they would make a recommendation to the director of CDC, and then that directive would be approved, if it had to be approved by the White House, which, as far as I know, has never really been involved before. Those recommendations would be the government’s policy on release of vaccines. In this circumstance, it is extraordinary that the director of the NIH and the director of the Center for Disease Control went to the National Academy of Medicine and said, we would like an independent agency to make the recommendations on who gets vaccines when they are limited. Which groups, which tiers would get first, second and third shots set by a vaccine. I think that reflects a concern that there is a possibility to manipulate what goes on within the government. But the National Academy of Sciences, Engineering, Medicine and Sciences was created in 1865 by Abraham Lincoln. It is a fully independent agency of government and there is no possibility that I’m aware of for interference in any aspect of a NAM or NAS report. Having served on many of those, I can tell you, you get no external input whatsoever on the recommendations of a committee of scientists. So I am pretty confident that they will do their best job to do the fairest recommendations of how to use the most limited vaccines. I would present them as a dilemma and much more complicated than I will present it.
There are really two ways to think about priorities in vaccines. One is, who are the people who are most at risk that you want to save their lives. That would be, among others, health care personnel, people within nursing homes, prisoners, people who are essential workers in public contacts. I would point out in that context, prioritizing them is clearly a good and it would certainly probably prioritize saving lives. On the other hand, and Bill may comment more adequately than I can, Nursing home people are not the major sources of the epidemic. Hospital personnel are not the major sources of epidemic. In fact, most infections among hospital personnel where it’s been studied come from household contacts of the hospital workers. If you want to stop the epidemic, you want to stop the transmitters. And I’m not sure we fully understand who they are, but that might be students, college aged kids, young people, people who go to bars. I’m not sure that would be the obvious priority, but those are the dilemmas that I believe the National Academy of Medicine has to contemplate. So that would be the second part of your question.
The third is, once the vaccines are approved, how many doses do you need to provide community protection commonly and appropriately called herd immunity to get the country protected. If the estimates are that about 60 percent of the people would have to be vaccinated – it’s a rough figure making guesses on prevalence of infection a year from now, for example – we have no company at the moment that could make 200 million doses for Americans or 5.6 billion doses to get 60 percent of the world’s people protected against coronavirus too. That explains why there has been support for multiple different vaccines. A, we don’t know which will work. B, we don’t know which are safe. And C, if they were safe and effective, as of the moment, nobody can make that number of doses to cover the entire planet. And most of them don’t have the capacity at this point to make enough doses in less than a year, to protect the American people. So there will be a big push either for big companies to, as they have done, make major investments in building factories. To build a vaccine that’s new, you basically have to build a factory, and that’s a huge investment. Many of the companies, the big companies, are doing that. The small companies don’t have the resources to do that. And so the government and the BARDA agency is helping to support them, to be able to do things like contract manufacturing organizations that have the facilities and capacity to make doses of vaccines, and contract out the work plan for different companies, and in this case, biotech companies. The expectation is by the end of 2021, there will be plenty of whatever vaccines are safe and effective for the United States. Less clear how they will be distributed around the world. The WHO and Europe have worked hard to make an integrated proposal of rationally allocating vaccines globally. The United States government has not joined that effort and, as you know, has withdrawn its support for the World Health Organization. I will stop there and apologize for talking so long. But thanks for the question.
Q: Thank you.
MODERATOR: Dr. Hanage, did you have anything you would like to say?
BILL HANAGE: Thank you, Barry, I learned a great deal in that. I would add to it, amid the reasonable anxiety about critical influence over these processes, the logistics which were alluded to, right at the end of that, matter. Because even if there isn’t a vaccine or a promising candidate, which is announced at some point in the fall or before or after the election, it still matters how we go through the logistics of actually getting the health of people. So the announcement of a vaccine is not what should be in people’s minds, what should be in people’s minds is that they feel confident that an administration would be able to supply it, and actually get it out to the American people.
MODERATOR: Thank you very much Dr. Hanage. All right, next question.
Q: Hi guys. Bill, I think you might have touched on this at the beginning, but I’m curious. There’s been a drop in testing in a number of states. Can you talk more about that basically and what that means for our numbers and maybe also why we might be seeing this?
BILL HANAGE: Yeah. There has been a drop in testing and a large number of states and there are all kinds of reasons why testing rates might drop, depending on the amounts of disease which are going on. People may start making presumptive diagnoses or it may be that the quantity of pandemic activity is just something we’re seeing a bit lag time in terms of the amount of time it takes for tests to be turned around. As you probably know, we have a pretty bad rate of turnaround for a lot of tests, certainly such that they are not overly useful for contact tracing, and even in the kind of clinical cases which are being tested for here, it can take a bit of time. So I have noticed, for instance, that the numbers of tests being done in Florida, in Texas as well I think maybe Arizona, and I would check one of those, by the way, because, you know, this is something which I did a few weeks ago so they may not be consistently that, but they have been dropping virus to proportion of tests coming back, say roughly the same. Which means overall, the exact state of the pandemic in those states is uncertain. We don’t have enough of a handle on exactly what’s going on. I think it’s reasonable to think that there is some plateauing going on. However, exactly how much of it there is would require much better testing than we have at the moment. And I’ll it hand off to Barry.
BARRY BLOOM: Just to harp on my usual intervention. There’s enormous focus on tests and test numbers. I would simply reiterate that testing does not have directly any impact on epidemic persay. It is the quarantine of somebody whose positive, and the identification and isolation of contacts, which is the public health intervention that interrupts trains of transmission. And when you find out that fewer than half the people in New York can give the name of a contact within 24 or 48 hours, it doesn’t matter how many tests you do. Fifty percent of the people are not identified or not isolating and that are capable of transmitting. The numbers of tests doesn’t really tell you how effective the testing and quarantine isolation are actually working. And I worry a lot about that.
BILL HANAGE: And that’s a really excellent point, the difference between testing as something which is meaningful to public health and testing as just a way of keeping score.
MODERATOR: Do you have a follow up question?
Q: No, I’m good. Thank you very much. I appreciate it, guys.
MODERATOR: Thank you. Next question.
Q: Thank you. It’s probably for Dr. Hanage, but to what can we attribute the drop in daily new cases in the US? They’ve been going down in recent days, and I just wondered if you think that’s due to the lower numbers of tests or could it be mask mandates in some states, any idea what’s behind those drops?
BILL HANAGE: You can attribute it to any number of things. The actual amount of testing is just one thing. The other possibility is that there are people making different kinds of contacts over time. So there is a genuine plateau going on because there’s a lot of attention. And people started to attempt to take action about maybe two to three weeks ago. That was when we started seeing a lot of this. And around now is when you expect to see that. However, I’m somewhat loath to really suggest that it’s making a huge impact. One thing which gives me a little bit more optimism about it is it seems rates of hospitalization do appear to be slowing somewhat. And that means, I think, that something real may be happening. But as with every stage of this pandemic, we are struggling to identify exactly what exactly the causes are of the dynamics that we’re seeing on the ground. So masks may contribute to it, yes, they may. But also, I would just suggest that the fact that people are more likely staying home and that they’re more aware of the risks of large gatherings, that can play an important role as well.
Q: Appreciate it. Thank you.
MODERATOR: Next question.
Q: Thanks for taking my question. I’m writing a story on whether it’s safe for people to resume public transportation. Most people haven’t been taking subways or busses, and many people are concerned about the spread of COVID-19. But, several of the largest systems. D.C., New York City and Chicago, are taking safety precautions, requiring passengers and staff to wear masks, distributing hand sanitizer and even air circulation rates are somewhat higher than indoor restaurants. So I just wonder if you can comment on the combination of these measures and then I’ll have a follow up about reducing capacity.
BILL HANAGE: It’s an excellent question. I would say that at the moment, we probably do not have the absolute activity to be able to specifically tie down risk. This is because of the fact that in most places where this is happening, community prevalence right now is comparatively rare, and we are just seeing people start moving back into those locations, into public transit. It’s certainly safer than it would have been. And it’s certainly helpful that these steps have been taken. In combination with other actions, it is likely to make it substantially safer. But again, we need to distinguish between a couple of things here, which is, you know, I think everybody on this call knows this, but it’s worth reiterating that there is individual risk which is never nil, but is much lower for some ages and others. Then there is also the community risk. If we find that these are associated with more transmission or if we find that in any of these places, we start getting a big uptick in terms of the amount of transmission, then we may have to be very careful and consider perhaps taking the foot off the gas somewhat, because it does appear that it doesn’t take a lot of this virus to be able to transmit and reach significant levels in the community. But again, the individual personal risk of these things, that’s another matter entirely. But everywhere. I would keep a close eye on the current prevalence.
Q: Just a quick follow up. Chicago’s transit system has implemented a digital dashboard to monitor capacity, and they’re even not letting a lot of people on if they reach it, or the bus drivers are allowed to skip stops. That they seem to be the only city that I’ve noticed doing this. I’m wondering if more cities, again, the big lines, New York, DC, do you see that the larger systems at least should be implementing that?
BILL HANAGE: I’m unaware of exactly what’s happening in all the cities, but it sounds like an extremely sensible idea, depending on how set the capacity. As I’ve commented before, the virus is characterized along with just like other coronaviruses we know of, like the original SARS and MERS, by having a tendency to transmit primarily through clusters. A lot of people just do not transmit at all or transmit to any one other person. But then a few of them transmit to a lot. And obviously the density of people on public transport would affect the number of people who could be transmitted in such a so-called super spreading event. So it seems like a really good idea to me, depending on where you set it. If you set it so there’s a lot of crowding, then obviously it’s a less than ideal than if you set it for less crowding.
Q: Thank you very much.
BARRY BLOOM: I could just chip in. I hate to do what I’m about to do, but I’ll do it. An anecdote which I found actually quite meaningful. There is some skepticism about the effectiveness of face masks. And there’s just a lot of data coming out to indicate one, as we know, they’re not perfect. In terms of protecting someone in your company from getting infected from you, they’re pretty good. About two thirds effective if they were decent, ordinary face, mask, cloth, face, mask, skepticism has been. Are they protected to you of someone else who does not wear or does wear face masks? And the anecdote was in MMWR are a couple of weeks ago that was quite striking of a beauty salon in Florida where two hairstylists both went to work feeling mildly ill. Both wore masks, styled the hair of 139 women, I think, all of whom both the stylists and the clients wore masks. And after the stylists were found to be sick with COVID, they tracked down the clients and not a single one became COVID infected. It’s an anecdote. One can’t generalize, but it does say if everybody was a face mask, infection as it is, either all or none, either you’re infected or you’re not. And at least in my field, TB, we talk about the force of infection. How many virus particles, how many bacterial particles are you exposed to? And even if you are exposed to a small number, you may not get sick. And if you don’t have a face mask, then you’ll be exposed to more. So that’s the anecdote for the day.
Q: Thank you.
MODERATOR: Next question.
Q: Thank you. So just returning to the vaccine question, I’m intrigued by the FDA saying they’ll accept a 50 percent effective vaccine. My interest is, it’s one thing if it prevents 50 percent of infections. It’s a little bit different, perhaps if it prevents 50 percent of serious disease. So can you talk a little bit about, depending on what kind of vaccine we get, what people should expect from it in terms of herd immunity, in terms of I think a lot of people think of the vaccine as the solution, and as soon as we have it, we’re all going to go back to normal. So that’s sort of my question, because it seems to be an indication that the vaccine is probably not going to be really, really effective and in case it doesn’t prevent infection. That seems sort of important.
BARRY BLOOM: Yeah. So what you really want out of a vaccine is to prevent transmission from a public health point of view. And if, for example, it prevents infection, then the person not infected is obviously not able to transmit, and you have interrupted at the first possible entry point, a transmitter. In the FDA guidelines that outline what they’re expecting in trials, they’re interested in the ability of the vaccines to prevent infection and disease. And you may get different answers from both. It may not prevent infection, and there would be a time when people still have virus in their upper respiratory tract that can transmit. But it would prevent them from going into intensive care units and going on ventilation. That would be a good thing, but not ideal, which would be to block infection if one possibly could do that. So they will collect that information on that.
Second point. The 50 percent endpoint, I would guess, has to do with is it in the economic interest of any company to make a vaccine that is less than 50 percent effective, knowing that there are 100 other candidates out there? I would say that from a practical point of view, a vaccine that is less than 50 percent effective would not be of interest, given the possibility that would be others better. I would point out to you that on average, flu vaccines that we have every year are on the order of 50 percent effective. Sixteen thousand people died last year of influenza. To cut any potentially lethal infectious disease by 50 percent is not trivial. The expectation that we have a vaccine is like smallpox and measles to be almost 100 percent effective and take that for granted. This is not a realistic expectation. So the hope is the vaccines will be significantly better than that. I’m very dubious they will be in the range of 94 percent protection like measles, although I would point out one company is making a recombinant measles with the coronavirus spike protein in the hopes that it would be a one shot vaccine that might be as effective as measles.
BILL HANAGE: I’m just going to add a little bit to that. So the naive calculation that you would do in order to estimate the proportion of the population that needs to be vaccinated with a completely perfect, 100 percent effective vaccine that completely prevents a person becoming infected in the first place, given the parameters we think are relevant for SARS-CoV-2, the proportion of the population would need to be around 50 percent. Maybe a little bit lower of more complicated things because, you know, the naive calculation is just that, naive. Now, obviously, getting a vaccine to that many people is difficult. But if the vaccine does not prevent against somebody becoming infected and being able to transmit, then you’d have to vaccinate far more of the population, basically all of it, in order to obtain the benefit because there is no herd effect as a result. Now, if it is 50 percent effective at preventing infection, so 50 percent vaccine efficacy. That means, again, you would have to vaccinate far more of the population in order to achieve that desired amount of herd immunity and to reduce the effective reproduction number of the virus below one. So essentially, a 50 percent effective vaccine, like Barry said, I wouldn’t sneeze of it, as it were. However, it’s also something which is not going to be as important and effective in terms of the public health impact of something which is which is better. I don’t expect it to be 100 percent effective. However, we should hope to be able to get a vaccine which is more effective than 50 percent and which is able to prevent, crucially infection rather than just the most severe consequences of being infected.
BARRY BLOOM: I could just chip in one more point, a point that I find frustrates me greatly. The term herd immunity is a very poor misnomer.
BILL HANAGE: Yeah, absolutely.
BARRY BLOOM: The reason is that it defines community protection. If enough people in a room are protected such that a naive person never immunized walks in. Herd immunity means that that person has a low probability of getting infected. It does not mean that that person is immune. In fact, that person is fully susceptible. So it’s a concept that is confusing. And community protection says that the more people who are immune in the community, the lower the chance that anybody else entering that community with out vaccination, the lower their chance of getting infected, I think in pointing out that herd immunity does not guarantee that people who don’t get vaccinated are protected.
Q: So do you expect herd to come into play with a vaccine? And then you do expect a vaccine to prevent infection?
BILL HANAGE: I would hope the vaccine can prevent infection. I mean, a vaccine that stops the most severe consequences of disease would still be a good thing, but it wouldn’t have a large public health impact. And if it only prevents against the most severe consequences of disease, herd immunity is irrelevant. If it can prevent infection, then it will have an effect upon the level of population level immunity, shall we say. But the amounts to which it contributes to that will depend upon its efficacy.
Q: OK. Thank you.
BILL HANAGE: If it’s 50 percent effective, then you need to vaccinate twice as many people to have the same effect.
MODERATOR: Next question.
Q: Hi. I’m wondering if you have any thoughts on vaccine distribution. It looks like Congress is poised to leave town without distributing any more money towards vaccine manufacturing, kill finish or a distribution plan. There is no national distribution plan in sight. It seems like CDC is working with individual states. Is that in keeping with prior pandemics or is that a concern for you?
BARRY BLOOM: Everything is a concern for me that didn’t used to be a concern. But let me go back to the ACIP again. A greatly under understood and under appreciated organization that has managed to provide vaccines for years and years, all the childhood vaccines in every district in the United States, quietly and effectively. The only problem being not the distribution system and oversight of the CDC, at least traditionally, it’s the unwillingness of the public in some cases to accept the vaccines, which is, to be absolutely honest. My biggest concern right now is not that the vaccines won’t have sufficient efficacy to make it worthwhile distributing them, some of them, at least, it’s that people won’t take them. And so ACIP works through the Association of Immunization Managers. I could ask all of you online, have you ever heard of the AIM? These are the workers called public health workers in every state that organized the distribution of vaccines, report on every school and the number of children in every school, in every state, to states, and then the CDC on how many kids in that age cohort in that school, in that district, are vaccinated. An enormously important source of information on understanding, for example, places where measles vaccine is not taken up, where you put if you’re a public health person, keep your eyeballs to see whether you’re running an outbreak there.
So what I’m trying to suggest is that we actually have a terrific mechanism, or have had, underfunded, under recognized, understaffed, that has done a remarkable job of getting childhood vaccines out. We’re also responsible for adult vaccines, which we have very few. One of them is a second shot of measles, as you know, a booster shot. And the other one is human papillomavirus vaccine, HPV, to prevent cervical cancer and head and neck cancer in boys. They would have to tool up and get resources to be able to organize if it was nursing homes or prisons. Not so difficult for prisons because you have a captive population, pardon the pun, but nonetheless, the agency that could do it if they have the funding and personnel to do it. I’m less worried about the distribution than the acceptance.
BILL HANAGE: I think I will only add that I made the comment earlier about the important logistics of getting a vaccine out. And it’s something which we’re gonna be thinking about a lot in the coming months. I would urge lawmakers to take it absolutely seriously because it even if we got a vaccine, being able to use it well, is one of the most important things, we have to do something with it. It’s not just getting a vaccine. It’s using it and using it appropriately.
BARRY BLOOM: Just a small comment. Again, the focus is on vaccines. Vaccines don’t prevent anything. Vaccination does.
BILL HANAGE: Good soundbite, Barry.
BARRY BLOOM: Thank you.
MODERATOR: Do you have any follow up?
Q: I do, actually. So I talked to the Association of Immunization Managers and they’ve sort of stressed that there is this an existing infrastructure, but they’re concerned about the involvement of the military through operation warp speed and how that could harm public trust in the process. Is that similar to what you’re saying? Are you saying that this existing infrastructure through CDC is capable of scaling up to the degree that would be necessary to get out enough vaccines for COVID-19? And do you agree that the involvement of DOD could hurt public trust?
BARRY BLOOM: In the distribution of vaccines, the supply chain issues are really complicated. That is associated with a cold chain where some vaccines have to be kept at four degrees. It is not clear, but it looks like RNA vaccines may have to be kept frozen. Those are things that, for the vaccines we know about, the Association of Immunization Managers and CDC have managed to handle well. I would say there is a complementary role for the DOD to fly large numbers of vaccines, fast, to every distribution point that is necessary. I don’t see the role of the military in going into local school districts and actually handing it to the vaccinators or physicians assistants and doctors offices or emergency rooms. I would hope they could be organized in a complementary way. And I actually have some experience in overseas stuff with Ebola and the military in their effectiveness at getting stuff from one place to another safely and quickly, in an organized way. They could be helpful here, provided that their role is defined. And my worry is whether the immunization managers and the CDC have the resources to take on 200 million vaccines from five different producers and get them out quickly enough.
Q: Thank you.
MODERATOR: We have about ten, twelve minutes left to go, and a lot more questions to go. So if we could try and get going a bit faster, if possible.
Q: Hi. Thank you for taking my question. So this is speaking to the genetics of the virus and a bit to vaccine potency in the future. My question is, can you guys speak to the latest research that was published August 4th in Science that suggests exposure to common cold coronaviruses can teach the immune system to recognize SARS-CoV-2? And what are the implications of that?
BILL HANAGE: And that’s a really good question. Barry, do you want to start or should I?
BARRY BLOOM: The answer is it is intriguing at a number of levels. But let me start with the simplest case. So we have four common coronaviruses that are involved in causing common colds. One of the first questions is, do they provide new antibodies to those viruses that seem to keep them coming and as mild infections on a regular basis, but do not eliminate them from the population? Could those antibodies cross react with the coronavirus and protect us? And almost all the studies, with some rare exceptions, have shown that antibodies, the surface proteins of any of those viruses are not protective, at least in test tubes against infection by SARS-CoV-2.
What this new work indicates is, however, that there are T cells that see common determinants on different antigens that are engaged and primed by seasonal coronaviruses, and that there may be something called immunological memory that would enable people to make a more rapid response. I have no way of knowing to what extent that is possible. We don’t have data on it. There are sure enough, people that have had common cold viruses that end up on ventilation and dying, to indicate that I wouldn’t put my life on T cells against a common cold virus keeping me out of ICU. But there is a possibility that they do provide some advantage. There is also a possibility that they could work in the opposite direction, that they’re primed to produce the wrong cytokines. And then when the virus gets out of control in some people, for whatever reason, they contribute to the cytokine storm because they’re already primed to make cytokines that are inflammatory and there’s no control on that. So the answer is there is T cell memory, and I don’t believe we have a good idea of what the all of that, if any, is in either protecting or enhancing the seriousness of this disease.
BILL HANAGE: I will echo that, and I will add, it’s not often that I necessarily quote journalists on this, but Ed Yong, The Atlantic, who is an excellent writer, wrote a brilliant piece yesterday in which he made the comment, “Immunology is where intuition goes to die.” I mean, this is the really crucial thing is that it’s a really important finding. But what it means, we don’t know. I mean, it could plausibly have some role in explaining why some people appear to be very mildly affected by disease, while others have it much more seriously, but at the moment, we just don’t know. One thing I would like to push back on is I have heard some places suggesting that this can explain why different regions have had different experiences with the pandemic so far, with some countries having a lot of death and lots of disease, others having less. And that is just fundamentally, completely impossible because you’d have to suggest that for some reason, different amounts of people are being infected by different amounts of cold viruses and so on. And we don’t have any reason to really think that, say, people in Germany are being infected with different coronaviruses, causing the common cold, than people in Britain. So it’s extremely important observation, its implications are not at all clear.
Q: Thank you.
MODERATOR: Did you have a follow up? Are you all set?
Q: No, I’m set. Thank you very much.
MODERATOR: Okay. Next question.
Q: Hey, thanks so much. I have one more tough question and thanks for your previous answer. I’ll try to be quick. In Arizona, tests have dropped, but things are also getting slightly better and we’re seeing fewer hospitalizations. So how important is continued widespread testing and why? And kind of what would the ideal for that look like?
BILL HANAGE: Firstly, I’m really glad that things are getting a little better in Arizona. I think the important thing here is to look at it from the lens that Barry put out. Are you doing testing in order to keeping score and monitoring the situation? Or are you doing testing as a way which is going to actually intervene and it’s actually gonna be a public health intervention? So more testing will enable you to get a better idea of the amount of community transmission which is going on and whether or not you’re seeing local upticks in particular places, as we’ve found, regrettably, in a lot of places across the United States. We know that this does not restrict itself to metropolitan areas, such as small towns as well. Testing in those small towns needs to be good enough to capture introductions, large clusters, and medium to large amounts of local transmission as well. Unfortunately, the question of what proportion of tests coming back positive is correct is a really complicated part because it depends on why you’re doing the tests and who you are testing.
I mean, just to take the obvious example. If some of these test results are from testing the baseball team every week, then we expect most of them to be coming back negative. And so it can end up inflating the negative tests over and above what they should be. So I would tend to say, you know, the thing about testing is to be clear about what you’re doing it for. And if you’re going to do it in order to test the amounts of transmission in your community, you’re going to need to have more testing, not less. But that in and of itself is not going to help you. It’s only going to let you know that something bad is happening.
MODERATOR: Do you have a follow up?
Q: No, that’s great. Thank you so much.
MODERATOR: Next question.
Q: Hey, how’s it going? Thanks for thanks for taking the call. This is a pretty specific question for Miami-Dade, but there is a model that I’m sure you both have heard of by Youyang Gu, that basically extrapolates out from deaths. It shows that Miami-Dade County has more than 34 percent of its population already infected by SARS-CoV-2, which would be higher than the New York City area. I know you all haven’t necessarily looked under the hood there, but it just seemed like a pretty astounding estimate. I spoke to Dr. Gu about this recently, and he basically said the number jumped out to him as well. But that’s what his modeling shows based off the deaths. And so I just kind of wanted to get that in context and get your thoughts on how much should we read into a model like that and how skeptical should we be that 34 percent of the population in Miami-Dade County, which has about 3 million people, have already been infected with SARS-CoV-2.
BILL HANAGE: How many deaths have been there so far? Because that might be able to make it see if it’s plausible or not.
Q: Yeah. Let me let me pull it up real quick. I believe it’s 6,140 last time I checked.
BILL HANAGE: So 6,140 deaths in Miami-Dade alone? If you multiply that by around two hundred and then you ask whether or not that’s that similar proportion of the county, then it seems like it might be on the nose. With Youyang Gu’s model, it’s an interesting one because, actually, it’s an SEIR model, so it’s a mechanistic model, which actually accounts for transmission. But it estimates parameters using machine learning. So all of these models have complicated parameters. Like how likely you are to transmit, how likely you are to die, and so on. And it estimates using machine learning. Now, models are only as good as our assumptions, and machine learning is only as good as its data set. So while it is a model, I actually look at a lot and I take quite seriously, I would expect it to occasionally throw some interesting things. On the other hand, not many deaths in a county is consistent with a relatively large number of people being infected. So I’d just be interested to know what the arithmetic shows.
Q: Yeah. And so I apologize. I was thinking of statewide deaths, so currently in Miami-Dade, it’s 1,724 currently and it projects 3,392 by November 1st.
BILL HANAGE: And that’s the number of people who are are immune or who have been exposed already.
Q: So those were deaths.
BILL HANAGE: No, of the 38 percent.
Q: Well, it actually it looks like it changed a little bit since I last looked at it, but it now says 32.8 percent as of August 5th.
BILL HANAGE: In the interest of time, it might be easier if I just correspond with you after this because this will take some calculation.
Q: That would be great. Thanks a lot. I appreciate it.
BILL HANAGE: Yes.
MODERATOR: Next question.
Q: Thanks so much for doing this and taking my question. My question is about the safety of reopening for schools. And I don’t know if there can be an overall answer to that question. If those schools are safe to return for the entire nation, that may be unfair, but I am in Texas, specifically, I am covering Harris County, which should probably pop up in your head as one of the hot spots. We also have a hot spot in the valley near the border. So maybe Texas focused, I mean, do we have the right data or what does the data show regarding the risk of reopening schools here in Texas?
BILL HANAGE: I would say in Texas, schools should not be reopened, at the moment. And I want to emphasize that schools should be a priority in any plan. I’m trying to step away from the phrase reopening, by the way, and try to introduce the phrase pandemic management, because I think it’s in the more responsible way in communicating what’s actually going on. It’s a mistake to think you can have everything, but schools are really important. Now in terms of explaining what happens in schools, and this is relevant to other people on the call, it’s very clear that we need to think about age groups differently. So adolescents, college students, they seem to transmit the virus a lot. We have that by direct observation of clusters. We also have it by looking at serology, meaning the presence of antibodies from the spring surge, which show that this age group tends to get infected quite a lot. We don’t observe it very much because of the fact that we weren’t testing very much at the time. But when we did look at major surges in Florida, for instance, we found that the age group was being infected quite a lot.
Now, the contrast with that is a younger age group. And that’s a very, very complicated question to which there are no really coherent, consistent answers on the moment. And the reason for that is twofold. Firstly, almost everywhere schools were shut at that age group, the elementary schools, quite early on. So we don’t actually have much data on how much transmission is gonna be happening within them. Even if they have been going on within them, because that age group is much less likely to show severe symptoms of disease, they have very much less likely to be detected. Now, I think it is reasonable to suggest they probably are somewhat less likely to be infected. This is partially as a result of the actual amounts of the receptor that the virus binds to among the younger children. I think it’s reasonable that they’re less likely to be infected and somewhat less likely to transmit, but not to the extent that we can rule out the potential for schools in that age group to actually be sources of infection.
Now, as a caveat on that, the fact that there are places in the world that have managed to open schools with very little consequence, if any, for the overall course of the pandemic. And I would look at Denmark as a great example. The thing about Denmark is that Denmark has never had a huge amount of disease in the first place. And if there’s not a large amount of community transmission going on, then the probability of it being increased to school and one of these clusters of infection starting, that is correspondingly, extremely low. Whereas, if you’re in a place where there is a relatively large amount of transmission going on, the probability of the virus entering a school is actually quite high. Once the virus is able to get into the school, if it does transmit, and that’s an opportunity for it to potentially get into a number of other households. So it’s a complex question. The science is evolving as I speak, and that’s a bit of a moving target. But I feel very strongly that if community transmission is at the level that you’re describing, it is not safe to open schools. And I’ll let Barry say what he thinks.
BARRY BLOOM: I think that’s a really good answer, and I sometimes listen to lots of experts who have very strong views on this, for which it is not much of an evidential base. There are anecdotes and analogies. So Denmark had a good record. Sweden also had their schools open and kids did fine. The epidemic was no different than in Denmark, except that there was a five to eight times higher death rate among people in the elderly population. And one interpretation is that grandma and grandpa catch pneumonia, pneumococcal pneumonia from their grandchildren, maybe that’s a source of infection in Sweden. And to pursue Bill’s point, in a country like Israel, which has an ongoing, fairly high transmission rate, when they opened the schools, it was a nightmare and there is no question the schools contributed. So I would take Bill’s advice as the best advice. It depends on the context in the district in which you’re making a decision. You can’t make it on a national basis. If transmission is high, kids will contribute to transmission and be involved. If it’s low, you have a good chance, particularly for younger kids, not to make a big contribution, but you really have to get better data than we have.
BILL HANAGE: Yes, I will echo what Barry said about Sweden. Sweden actually did close high schools and universities, but it kept elementary school schools for young kids, what they call compulsory schools, open. And as Barry says, the per capita mortality rates in Sweden is enormously higher than it is in Denmark.
Q: A quick follow up question, if I may. So right now, Texas in some districts have set a special date for in-person instruction and some are going back like next week. So is it fair to set a in person start instruction by, say, September X? Based on the fact that we don’t know what the virus is going to do by September X?
BILL HANAGE: I wish people would stop pretending they know what the virus is going to do. I wish they would. I mean, come on. You know, I find it not enough going out beyond a few weeks. The pandemic is not going to play ball. The only impact that you can have on the pandemic is the actions we take in order to preserve public health. That’s it. The virus is not going to read that and sort of think, oh, wow, it better back off because they need schools reopen. No, what you have to do is you have to figure out how to allow an amount of transmission within the community that you’re comfortable with. That might be very little. If you want to be say, New Zealand or Australia, and you take the corresponding action, whereas essentially, if you are going to do something on a date at some point in the future, you have to be prepared for the situation to change. This is something which we have seen again and again and again and again. And I really profoundly wish that people would stop expecting the virus to cooperate.
MODERATOR: OK, I’m going to break in here real quick. So we’ve run over already, but we still have questions. I don’t know if you want to stay on longer or if you need to go.
BILL HANAGE: I don’t have anything immediate. But I have got a lot of people waiting on me since I started my Zoom calls many hours ago. But OK, we’ll try and be quick.
MODERATOR: OK. Because I’m going to ask you this one, it’s from an independent journalist working on a story about a group of educators who are pushing for officials to consider an alternative back to school scenario in which the youngest grades, K through three, return to school in some form of outdoor learning in locations where that’s appropriate and positivity rates are low. The concept is to be able to phase in older grades once infection rates are lower, but perhaps until there’s a vaccine. This would be an alternative to returning to enclosed classrooms and or as a means of extending school space, six feet of separation. Does this concept seem safe or feasible? Other doctors have mentioned that it might be an option in communities where PCR positivity rates are around one out of five hundred and she’d appreciate hearing any input on that concept in metric.
BILL HANAGE: That’s a very interesting idea. It’s the kind of creative idea that is likely going to be needed if we’re going to be moving forward. I hesitate to say that it would work everywhere for reasons similar to those which I was just saying. However, I do think that’s an intriguing idea. And remember, like I said, the PCR positivity test rates reflects a whole lot of things from the kind of people who are being tested to whether or not you’ve been investigating nursing homes recently. So you should be careful to not get hung up upon that. However, if you want to talk about allowing in-person instruction for younger children, I think they’re less likely to transmit overall, but then much less likely to transmit if you’re going to be doing it outside. So I think that innovative concepts, innovative plans, are something which is going to be really helpful. And I will point out actually there are records of people doing outdoor teaching in the 1918, 1919 influenza pandemic.
MODERATOR: OK, great. Thank you. Next question.
Q: Thank you so much, everyone. Quickly, there seems to be an increasing call for a new national testing strategy of moving from the flow insensitive PCR test to the fast and frequent antigen tests. Understanding all your points about testing, and it is more than just testing, we need isolation and quarantine, but is this a strategy that you would recommend to get out of our current backlog and problems that we’re having, particularly here in California.
BARRY BLOOM: So, it is clear to me that the PCR test, which has high specificity and sensitivity, once the numbers get beyond the numbers where you can do tracking cases and contacts, even if it were turnaround times in 48 hours, it’s not going to keep up with this epidemic. The antigen tests are easy to do. They’re fast and they’re not very sensitive. The good ones, maybe about 50 percent of the small do it yourself kind. You can go and send samples to places that have higher specificity, sensitivity. But there, in my view, not the ultimate answer. And so a lot of us are pinning hopes on new molecular tests that would be doing at home tests. And I think what I have to emphasize. What one needs is not a test to know if you’re positive or negative. In college students, in high school students, in school students, in a business, you need to test people every other day, every three days, once a week. And you can’t do that. PCR is 100 dollars a test. You can’t do it. So we don’t have a public health acceptable test at this point. The NIH put out an urgent call for new ideas called RADX for rapid something or other accelerated diagnostics. They have already funded a couple concepts from biotech companies. They are reviewing more. I am hopeful that some of them will be good enough to be able to be approved by the end of the summer. And that would make a transformational effect in people able and businesses able for a buck or two a test, per person, for people to know whether they have been infected or not, and then self isolate. And without that, I think we are in deep trouble unless the numbers come down by ordinary public health needs.
BILL HANAGE: I think that is a really smart comment. I will add to it in a couple of ways. My colleague, Mike Mina, has been eloquent about the potential benefits of a high specificity, low sensitivity test, meaning that if it comes back positive, you definitely have COVID. If it comes back negative, you might not of have it, because of the fact that it would have the ability to be informative. And if it’s cheap enough, as Barry says, you can do it for a few bucks a day. It’s transformative for business. I want to point out one other thing about this, because it holds a place in my heart, and we’ve written some of a paper on it. It’s a particularly useful strategy if you have a virus which, like this, tends to transmit in clusters because even if it has low sensitivity, if that has been a cluster of transmission and a large number of people have become infected in a workplace, that’s something where it is extremely unlikely that all of the tests on them will come back with a false negative. Then if you have a positive, you’re then informed that something’s been in place there, a transmission case or potentially a transmission event, and once you have that, you can actually start taking directed action to take all of those individuals, isolate this period of time, and we’ll come back and testing with an alternative and better test.
Finally, a point which I think would be profoundly useful, which you alluded to, a national testing strategy would look completely different to what we have now. When people like me are sitting there trying to figure out, talking to people like Ben about what’s going on in Florida, I’m thinking, hang on, what exactly is Florida counting? Then it’s, oh, you don’t count people who are non Florida residents. What’s with that? And then trying to figure out how that will alter the numbers. It would be a profoundly helpful thing to people who are trying to understand the epidemiology of the pandemic. But the bit in which Barry is much interested in, the bit about making a public health impact, yes. Those tests, those kind of innovations will be really, really helpful.
Q: Perfect. Thank you so much. Just so I’m clear, it is my understanding with the antigen test, some of which have been distributed in nursing homes, that do have like a 15, 20 minute turnaround, can that meet the criteria for some of the innovations that you’re describing?
BILL HANAGE: Yes. That kind of a 15, 20 minutes turnaround is profoundly better than those sort of seven to nine days that we’re seeing from some commercial PCR testing facilities.
Q: Perfect. Thanks for your help.
BILL HANAGE: Thank you.
MODERATOR: I think this is our last question.
Q: Hi. Thank you so much for doing this and staying on a bit longer. There’s a study published in JAMA today of 300 COVID patients in South Korea. It found that asymptomatic patients had similar viral loads compared to those who did have symptoms. Does this line up with what we already have known about COVID or have suspected? And what are the implications for the response in the US?
BILL HANAGE: I think it lines up completely and I’ll be honest, I haven’t read the article yet. But if what you’re telling me is correct, it’s completely plausible, especially given what we know about the fact that there is a significant amount of transmission from people who are currently asymptomatic. And I’m not sure I haven’t read the study. I’m not sure if these are people who were continuously asymptomatic or if they developed some mild symptoms at some point. But, we have known for some time that unwitting transmission is possible from people who are currently not displaying symptoms, or displaying symptoms which would not be associated with COVID infection. Because, you know, this is a virus that transmits by the respiratory route, but its effects are systemic and it has a very, very wide range of presentations. The implications, what we do in the United States and in the world is that interventions which are focused on treating everybody as if they could be infected, all the most important ones, because to be honest, based on symptoms at any time, anybody could be infected. And that relates to the sort of non pharmaceutical interventions that we’ve been talking about all of this time before we start getting into vaccines.
Q: Thank you.
MODERATOR: One really quick follow up question. She’d like to know what your experts believe should be the transmission rate in order to safely open schools in a community.
BILL HANAGE: I don’t think you can actually put a simple number on that because of the fact that each community has different amounts of testing that are being done in each place are different. I would in doing so and thinking about that, and again, remember, it’s different ages are important here. I think that it’s very, very difficult to think about opening high schools if there’s the risk of introduction. But what you could do is you could ask your local community and ask yourself, well, what risk are we prepared to take? And then look at an estimate which will be different from the actual testing rate, remember, because the actual incidents within a community, meaning the number of people who are infected at a given time point, is not necessarily the same as the people who are getting tested. That’s only an estimate. Then, ask what and perhaps take when somebody comes in here with the virus? And then think that through. I’m sorry, I can’t be more specific, but I’d be concerned if I were to be more specific, it would be taken off to places where it’s not really appropriate to do so.
MODERATOR: Thank you. Dr. Bloom, did you have any final thoughts before we go?
BARRY BLOOM: No, I thank you all for very good questions and it’s always a pleasure to hang out with Bill.
BILL HANAGE: Likewise. I like our sort of sartorial contrasts.
MODERATOR: Dr. Hanage, do you have any final comments?
BILL HANAGE: No. It’s a pleasure to hang out with you and Barry and everyone on the call and I will do this again in a few weeks.
This concludes the August 6th press conference.
You’re listening to a press conference from the Harvard T.H. Chan School of Public Health with Barry Bloom, the Joan L. And Julius H. Jacobson Research Professor of Public Health, and William Hanage, associate professor of epidemiology and a faculty member of the Center for Communicable Disease Dynamics. This call was recorded at 11:30 a.m. Eastern Time on Thursday, August 6th.
Transcript
MODERATOR: Dr. Bloom, do you have any opening comments?
BARRY BLOOM: No, I look forward to questions. It’s an exciting time. I would also say today is an anniversary of the dropping of the atomic bomb on Hiroshima, and we should have in mind the misuse of science as well as the use of science.
MODERATOR: Thank you, Dr. Bloom. And Dr. Hanage, do you have any opening remarks?
BILL HANAGE: I certainly agree with what Barry was saying about the anniversary of the dropping of the atomic bomb on Hiroshima. I will say briefly as an opening for what’s going on in the pandemic, while cases seem to have peaked across the South, slow and limited testing may obscure the full picture, and we should not expect a long plateau, assuming things not change. But we are also seeing some clear upticks in a number of other states, including some of the Midwest, but also New Jersey and Massachusetts, albeit from a very low level, which experienced strong surges in the spring. This suggests that we have to be pretty cautious in how we go forward if we wish to maintain low levels of community transmission. And with that, I’ll turn it over. I look forward to the questions.
MODERATOR: Thank you, Dr. Hanage. Looks like our first question.
Q: Hey, thanks for doing this. I had a question about vaccines. The president keeps saying there will be a vaccine very soon, and says something about this fall. Just wanted to get your thoughts on given the political realities of this year. November 3rd being an important day. How concerned are you that there’ll be political pressure to cut corners or speed up the timeline? And feel free to differentiate between widespread distribution and emergency use authorization. Thank you.
BARRY BLOOM: Thanks for a really great question that has really absorbed many of us in the scientific community. It is likely that phase three studies of safety and efficacy, with about 30,000 people enrolled in each with a placebo control, will be completed for sure by the end of the year. There is a race, as you know, to try to get there early. There has been a concern that it would be used without full analysis of safety data and efficacy data to say the epidemic is so bad that we’ll just give it an emergency use authorization and we can get it out by November 3rd. I know of no specific plans that anybody has thought about doing that within the administration, but certainly that was a concern. So let me break that concern in several pieces.
The first piece is that the trial designs are really thoughtfully done, which gave a great relief to many of the scientific people who were involved in vaccines. The way you do a trial design is to figure out how to get statistical power as to whether there is a statistically significant difference between the vaccine recipients and those with a placebo. That requires knowing what the incidence of the infection is and disease is within communities where the vaccine is being tested. For example, China had a study done of a whole virus killed vaccine that they had to stop because they ran out of cases and they couldn’t get statistical power. So the design created by a committee with membership of the NIH, headed by the director of NIH, headed by the vaccine program with people from FDA, as well as NIH, is a really rigorous, thoughtful standard, critical design to get statistical power to the best that one can estimate what the prevalence will be at any time while the trial is ongoing to tell us if the vaccine is 50 percent or better protected. Looking at thirty thousand people over the time of that trial and what the adverse effects are and making an evaluation, whether they are serious enough to recommend that they not be given permission to distribute the vaccine.
The one thing that is not involved in the planning of a vaccine trial this time. It is not a variable. The variable is statistical significance. If you can run, instead of 10,000 people in a year, 30,000 people in three months, you get the same statistical power. And what you gain, in the way this is being done, is presumably a rigorous analysis of the efficacy and safety in 30,000 people over that period of time. What you don’t get in that timeframe is any indication of longer term adverse effects. So two points on adverse effects. As far as I know, there is no vaccine, in fact no medical intervention, of which I’m aware that doesn’t have adverse effects. The adverse effects seen in the phase one and two studies for the vaccines moving forward are standard. The same as a flu vaccine, which is kind of standard sore arm fever, muscle pain, weakness, almost all of which dissipates in 24 to 48 hours. What we worry about is uncommon or rare adverse effects, and they often don’t show up until you have very large numbers of people. So the scientific community, as you know, has written letters and spoken to the press. We want to be sure that all the data are analyzed and characterized and that the public is informed of what we know for each of the vaccines of what the efficacy is in fifteen to thirty thousand people that will get the vaccine. And what we know about adverse effects over that period of time. That’s been the concern, and I know of no evidence that that will not be followed. However, there are two concerns that cause those letters to be written. The FDA has within its authority, emergency use authorization, which has several committees within the FDA required to review it. But in the case of Hydroxychloroquine, somehow, mysteriously, with no documentation or evidential basis, the director of the FDA gave emergency use approval and that caused a great deal of confusion both in the medical community and the public and some distrust of the whole enterprise, packed in with a distrust of what government does in any case.
The second case, which I found equally troubling, is when a whole slew of antibody tests came in over a hundred. The FDA simply took the companies, mostly from China’s analysis of their specificity and sensitivity and never checked on them. And as you know, they had to rescind the emergency use for approvals for about 90 or 88 out of 100, so that only 12 were approved as being within the range that would even be credible to collect meaningful data. So with 0 for 2 on those two major issues, there is reason for skepticism that there could be an October surprise. Without public data, without full analysis, the director of FDA could simply say it’s a bad problem. I give it emergency use and we’ll collect data. That would really, I think, compromise what little trust we have in this country about A, vaccines in general, and B, the government’s controlling or regulating vaccines to protect the public interests. Those are the concerns. I’m very hopeful that they, having been published, having been in the public news, will not occur.
The second issue is, once those vaccines become available, after they have hopefully real approvals, after detailed analysis of the data by the FDA and made those public and adverse effects. The big challenge will be who gets it first, and I think an extraordinary development, unprecedented in my knowledge, has occurred in that. That is, there has been an agency that is responsible for recommendations for vaccines, a little known agency reporting to the CDC called the Advisory Committee on Immunization Practices. They look at the dosage. They look at the timing. They look at the interference of one vaccine in another. They look at new vaccines and they make recommendations of, when they are limited or a company has trouble, who would get them. It would be ordinarily the case that they would make a recommendation to the director of CDC, and then that directive would be approved, if it had to be approved by the White House, which, as far as I know, has never really been involved before. Those recommendations would be the government’s policy on release of vaccines. In this circumstance, it is extraordinary that the director of the NIH and the director of the Center for Disease Control went to the National Academy of Medicine and said, we would like an independent agency to make the recommendations on who gets vaccines when they are limited. Which groups, which tiers would get first, second and third shots set by a vaccine. I think that reflects a concern that there is a possibility to manipulate what goes on within the government. But the National Academy of Sciences, Engineering, Medicine and Sciences was created in 1865 by Abraham Lincoln. It is a fully independent agency of government and there is no possibility that I’m aware of for interference in any aspect of a NAM or NAS report. Having served on many of those, I can tell you, you get no external input whatsoever on the recommendations of a committee of scientists. So I am pretty confident that they will do their best job to do the fairest recommendations of how to use the most limited vaccines. I would present them as a dilemma and much more complicated than I will present it.
There are really two ways to think about priorities in vaccines. One is, who are the people who are most at risk that you want to save their lives. That would be, among others, health care personnel, people within nursing homes, prisoners, people who are essential workers in public contacts. I would point out in that context, prioritizing them is clearly a good and it would certainly probably prioritize saving lives. On the other hand, and Bill may comment more adequately than I can, Nursing home people are not the major sources of the epidemic. Hospital personnel are not the major sources of epidemic. In fact, most infections among hospital personnel where it’s been studied come from household contacts of the hospital workers. If you want to stop the epidemic, you want to stop the transmitters. And I’m not sure we fully understand who they are, but that might be students, college aged kids, young people, people who go to bars. I’m not sure that would be the obvious priority, but those are the dilemmas that I believe the National Academy of Medicine has to contemplate. So that would be the second part of your question.
The third is, once the vaccines are approved, how many doses do you need to provide community protection commonly and appropriately called herd immunity to get the country protected. If the estimates are that about 60 percent of the people would have to be vaccinated – it’s a rough figure making guesses on prevalence of infection a year from now, for example – we have no company at the moment that could make 200 million doses for Americans or 5.6 billion doses to get 60 percent of the world’s people protected against coronavirus too. That explains why there has been support for multiple different vaccines. A, we don’t know which will work. B, we don’t know which are safe. And C, if they were safe and effective, as of the moment, nobody can make that number of doses to cover the entire planet. And most of them don’t have the capacity at this point to make enough doses in less than a year, to protect the American people. So there will be a big push either for big companies to, as they have done, make major investments in building factories. To build a vaccine that’s new, you basically have to build a factory, and that’s a huge investment. Many of the companies, the big companies, are doing that. The small companies don’t have the resources to do that. And so the government and the BARDA agency is helping to support them, to be able to do things like contract manufacturing organizations that have the facilities and capacity to make doses of vaccines, and contract out the work plan for different companies, and in this case, biotech companies. The expectation is by the end of 2021, there will be plenty of whatever vaccines are safe and effective for the United States. Less clear how they will be distributed around the world. The WHO and Europe have worked hard to make an integrated proposal of rationally allocating vaccines globally. The United States government has not joined that effort and, as you know, has withdrawn its support for the World Health Organization. I will stop there and apologize for talking so long. But thanks for the question.
Q: Thank you.
MODERATOR: Dr. Hanage, did you have anything you would like to say?
BILL HANAGE: Thank you, Barry, I learned a great deal in that. I would add to it, amid the reasonable anxiety about critical influence over these processes, the logistics which were alluded to, right at the end of that, matter. Because even if there isn’t a vaccine or a promising candidate, which is announced at some point in the fall or before or after the election, it still matters how we go through the logistics of actually getting the health of people. So the announcement of a vaccine is not what should be in people’s minds, what should be in people’s minds is that they feel confident that an administration would be able to supply it, and actually get it out to the American people.
MODERATOR: Thank you very much Dr. Hanage. All right, next question.
Q: Hi guys. Bill, I think you might have touched on this at the beginning, but I’m curious. There’s been a drop in testing in a number of states. Can you talk more about that basically and what that means for our numbers and maybe also why we might be seeing this?
BILL HANAGE: Yeah. There has been a drop in testing and a large number of states and there are all kinds of reasons why testing rates might drop, depending on the amounts of disease which are going on. People may start making presumptive diagnoses or it may be that the quantity of pandemic activity is just something we’re seeing a bit lag time in terms of the amount of time it takes for tests to be turned around. As you probably know, we have a pretty bad rate of turnaround for a lot of tests, certainly such that they are not overly useful for contact tracing, and even in the kind of clinical cases which are being tested for here, it can take a bit of time. So I have noticed, for instance, that the numbers of tests being done in Florida, in Texas as well I think maybe Arizona, and I would check one of those, by the way, because, you know, this is something which I did a few weeks ago so they may not be consistently that, but they have been dropping virus to proportion of tests coming back, say roughly the same. Which means overall, the exact state of the pandemic in those states is uncertain. We don’t have enough of a handle on exactly what’s going on. I think it’s reasonable to think that there is some plateauing going on. However, exactly how much of it there is would require much better testing than we have at the moment. And I’ll it hand off to Barry.
BARRY BLOOM: Just to harp on my usual intervention. There’s enormous focus on tests and test numbers. I would simply reiterate that testing does not have directly any impact on epidemic persay. It is the quarantine of somebody whose positive, and the identification and isolation of contacts, which is the public health intervention that interrupts trains of transmission. And when you find out that fewer than half the people in New York can give the name of a contact within 24 or 48 hours, it doesn’t matter how many tests you do. Fifty percent of the people are not identified or not isolating and that are capable of transmitting. The numbers of tests doesn’t really tell you how effective the testing and quarantine isolation are actually working. And I worry a lot about that.
BILL HANAGE: And that’s a really excellent point, the difference between testing as something which is meaningful to public health and testing as just a way of keeping score.
MODERATOR: Do you have a follow up question?
Q: No, I’m good. Thank you very much. I appreciate it, guys.
MODERATOR: Thank you. Next question.
Q: Thank you. It’s probably for Dr. Hanage, but to what can we attribute the drop in daily new cases in the US? They’ve been going down in recent days, and I just wondered if you think that’s due to the lower numbers of tests or could it be mask mandates in some states, any idea what’s behind those drops?
BILL HANAGE: You can attribute it to any number of things. The actual amount of testing is just one thing. The other possibility is that there are people making different kinds of contacts over time. So there is a genuine plateau going on because there’s a lot of attention. And people started to attempt to take action about maybe two to three weeks ago. That was when we started seeing a lot of this. And around now is when you expect to see that. However, I’m somewhat loath to really suggest that it’s making a huge impact. One thing which gives me a little bit more optimism about it is it seems rates of hospitalization do appear to be slowing somewhat. And that means, I think, that something real may be happening. But as with every stage of this pandemic, we are struggling to identify exactly what exactly the causes are of the dynamics that we’re seeing on the ground. So masks may contribute to it, yes, they may. But also, I would just suggest that the fact that people are more likely staying home and that they’re more aware of the risks of large gatherings, that can play an important role as well.
Q: Appreciate it. Thank you.
MODERATOR: Next question.
Q: Thanks for taking my question. I’m writing a story on whether it’s safe for people to resume public transportation. Most people haven’t been taking subways or busses, and many people are concerned about the spread of COVID-19. But, several of the largest systems. D.C., New York City and Chicago, are taking safety precautions, requiring passengers and staff to wear masks, distributing hand sanitizer and even air circulation rates are somewhat higher than indoor restaurants. So I just wonder if you can comment on the combination of these measures and then I’ll have a follow up about reducing capacity.
BILL HANAGE: It’s an excellent question. I would say that at the moment, we probably do not have the absolute activity to be able to specifically tie down risk. This is because of the fact that in most places where this is happening, community prevalence right now is comparatively rare, and we are just seeing people start moving back into those locations, into public transit. It’s certainly safer than it would have been. And it’s certainly helpful that these steps have been taken. In combination with other actions, it is likely to make it substantially safer. But again, we need to distinguish between a couple of things here, which is, you know, I think everybody on this call knows this, but it’s worth reiterating that there is individual risk which is never nil, but is much lower for some ages and others. Then there is also the community risk. If we find that these are associated with more transmission or if we find that in any of these places, we start getting a big uptick in terms of the amount of transmission, then we may have to be very careful and consider perhaps taking the foot off the gas somewhat, because it does appear that it doesn’t take a lot of this virus to be able to transmit and reach significant levels in the community. But again, the individual personal risk of these things, that’s another matter entirely. But everywhere. I would keep a close eye on the current prevalence.
Q: Just a quick follow up. Chicago’s transit system has implemented a digital dashboard to monitor capacity, and they’re even not letting a lot of people on if they reach it, or the bus drivers are allowed to skip stops. That they seem to be the only city that I’ve noticed doing this. I’m wondering if more cities, again, the big lines, New York, DC, do you see that the larger systems at least should be implementing that?
BILL HANAGE: I’m unaware of exactly what’s happening in all the cities, but it sounds like an extremely sensible idea, depending on how set the capacity. As I’ve commented before, the virus is characterized along with just like other coronaviruses we know of, like the original SARS and MERS, by having a tendency to transmit primarily through clusters. A lot of people just do not transmit at all or transmit to any one other person. But then a few of them transmit to a lot. And obviously the density of people on public transport would affect the number of people who could be transmitted in such a so-called super spreading event. So it seems like a really good idea to me, depending on where you set it. If you set it so there’s a lot of crowding, then obviously it’s a less than ideal than if you set it for less crowding.
Q: Thank you very much.
BARRY BLOOM: I could just chip in. I hate to do what I’m about to do, but I’ll do it. An anecdote which I found actually quite meaningful. There is some skepticism about the effectiveness of face masks. And there’s just a lot of data coming out to indicate one, as we know, they’re not perfect. In terms of protecting someone in your company from getting infected from you, they’re pretty good. About two thirds effective if they were decent, ordinary face, mask, cloth, face, mask, skepticism has been. Are they protected to you of someone else who does not wear or does wear face masks? And the anecdote was in MMWR are a couple of weeks ago that was quite striking of a beauty salon in Florida where two hairstylists both went to work feeling mildly ill. Both wore masks, styled the hair of 139 women, I think, all of whom both the stylists and the clients wore masks. And after the stylists were found to be sick with COVID, they tracked down the clients and not a single one became COVID infected. It’s an anecdote. One can’t generalize, but it does say if everybody was a face mask, infection as it is, either all or none, either you’re infected or you’re not. And at least in my field, TB, we talk about the force of infection. How many virus particles, how many bacterial particles are you exposed to? And even if you are exposed to a small number, you may not get sick. And if you don’t have a face mask, then you’ll be exposed to more. So that’s the anecdote for the day.
Q: Thank you.
MODERATOR: Next question.
Q: Thank you. So just returning to the vaccine question, I’m intrigued by the FDA saying they’ll accept a 50 percent effective vaccine. My interest is, it’s one thing if it prevents 50 percent of infections. It’s a little bit different, perhaps if it prevents 50 percent of serious disease. So can you talk a little bit about, depending on what kind of vaccine we get, what people should expect from it in terms of herd immunity, in terms of I think a lot of people think of the vaccine as the solution, and as soon as we have it, we’re all going to go back to normal. So that’s sort of my question, because it seems to be an indication that the vaccine is probably not going to be really, really effective and in case it doesn’t prevent infection. That seems sort of important.
BARRY BLOOM: Yeah. So what you really want out of a vaccine is to prevent transmission from a public health point of view. And if, for example, it prevents infection, then the person not infected is obviously not able to transmit, and you have interrupted at the first possible entry point, a transmitter. In the FDA guidelines that outline what they’re expecting in trials, they’re interested in the ability of the vaccines to prevent infection and disease. And you may get different answers from both. It may not prevent infection, and there would be a time when people still have virus in their upper respiratory tract that can transmit. But it would prevent them from going into intensive care units and going on ventilation. That would be a good thing, but not ideal, which would be to block infection if one possibly could do that. So they will collect that information on that.
Second point. The 50 percent endpoint, I would guess, has to do with is it in the economic interest of any company to make a vaccine that is less than 50 percent effective, knowing that there are 100 other candidates out there? I would say that from a practical point of view, a vaccine that is less than 50 percent effective would not be of interest, given the possibility that would be others better. I would point out to you that on average, flu vaccines that we have every year are on the order of 50 percent effective. Sixteen thousand people died last year of influenza. To cut any potentially lethal infectious disease by 50 percent is not trivial. The expectation that we have a vaccine is like smallpox and measles to be almost 100 percent effective and take that for granted. This is not a realistic expectation. So the hope is the vaccines will be significantly better than that. I’m very dubious they will be in the range of 94 percent protection like measles, although I would point out one company is making a recombinant measles with the coronavirus spike protein in the hopes that it would be a one shot vaccine that might be as effective as measles.
BILL HANAGE: I’m just going to add a little bit to that. So the naive calculation that you would do in order to estimate the proportion of the population that needs to be vaccinated with a completely perfect, 100 percent effective vaccine that completely prevents a person becoming infected in the first place, given the parameters we think are relevant for SARS-CoV-2, the proportion of the population would need to be around 50 percent. Maybe a little bit lower of more complicated things because, you know, the naive calculation is just that, naive. Now, obviously, getting a vaccine to that many people is difficult. But if the vaccine does not prevent against somebody becoming infected and being able to transmit, then you’d have to vaccinate far more of the population, basically all of it, in order to obtain the benefit because there is no herd effect as a result. Now, if it is 50 percent effective at preventing infection, so 50 percent vaccine efficacy. That means, again, you would have to vaccinate far more of the population in order to achieve that desired amount of herd immunity and to reduce the effective reproduction number of the virus below one. So essentially, a 50 percent effective vaccine, like Barry said, I wouldn’t sneeze of it, as it were. However, it’s also something which is not going to be as important and effective in terms of the public health impact of something which is which is better. I don’t expect it to be 100 percent effective. However, we should hope to be able to get a vaccine which is more effective than 50 percent and which is able to prevent, crucially infection rather than just the most severe consequences of being infected.
BARRY BLOOM: I could just chip in one more point, a point that I find frustrates me greatly. The term herd immunity is a very poor misnomer.
BILL HANAGE: Yeah, absolutely.
BARRY BLOOM: The reason is that it defines community protection. If enough people in a room are protected such that a naive person never immunized walks in. Herd immunity means that that person has a low probability of getting infected. It does not mean that that person is immune. In fact, that person is fully susceptible. So it’s a concept that is confusing. And community protection says that the more people who are immune in the community, the lower the chance that anybody else entering that community with out vaccination, the lower their chance of getting infected, I think in pointing out that herd immunity does not guarantee that people who don’t get vaccinated are protected.
Q: So do you expect herd to come into play with a vaccine? And then you do expect a vaccine to prevent infection?
BILL HANAGE: I would hope the vaccine can prevent infection. I mean, a vaccine that stops the most severe consequences of disease would still be a good thing, but it wouldn’t have a large public health impact. And if it only prevents against the most severe consequences of disease, herd immunity is irrelevant. If it can prevent infection, then it will have an effect upon the level of population level immunity, shall we say. But the amounts to which it contributes to that will depend upon its efficacy.
Q: OK. Thank you.
BILL HANAGE: If it’s 50 percent effective, then you need to vaccinate twice as many people to have the same effect.
MODERATOR: Next question.
Q: Hi. I’m wondering if you have any thoughts on vaccine distribution. It looks like Congress is poised to leave town without distributing any more money towards vaccine manufacturing, kill finish or a distribution plan. There is no national distribution plan in sight. It seems like CDC is working with individual states. Is that in keeping with prior pandemics or is that a concern for you?
BARRY BLOOM: Everything is a concern for me that didn’t used to be a concern. But let me go back to the ACIP again. A greatly under understood and under appreciated organization that has managed to provide vaccines for years and years, all the childhood vaccines in every district in the United States, quietly and effectively. The only problem being not the distribution system and oversight of the CDC, at least traditionally, it’s the unwillingness of the public in some cases to accept the vaccines, which is, to be absolutely honest. My biggest concern right now is not that the vaccines won’t have sufficient efficacy to make it worthwhile distributing them, some of them, at least, it’s that people won’t take them. And so ACIP works through the Association of Immunization Managers. I could ask all of you online, have you ever heard of the AIM? These are the workers called public health workers in every state that organized the distribution of vaccines, report on every school and the number of children in every school, in every state, to states, and then the CDC on how many kids in that age cohort in that school, in that district, are vaccinated. An enormously important source of information on understanding, for example, places where measles vaccine is not taken up, where you put if you’re a public health person, keep your eyeballs to see whether you’re running an outbreak there.
So what I’m trying to suggest is that we actually have a terrific mechanism, or have had, underfunded, under recognized, understaffed, that has done a remarkable job of getting childhood vaccines out. We’re also responsible for adult vaccines, which we have very few. One of them is a second shot of measles, as you know, a booster shot. And the other one is human papillomavirus vaccine, HPV, to prevent cervical cancer and head and neck cancer in boys. They would have to tool up and get resources to be able to organize if it was nursing homes or prisons. Not so difficult for prisons because you have a captive population, pardon the pun, but nonetheless, the agency that could do it if they have the funding and personnel to do it. I’m less worried about the distribution than the acceptance.
BILL HANAGE: I think I will only add that I made the comment earlier about the important logistics of getting a vaccine out. And it’s something which we’re gonna be thinking about a lot in the coming months. I would urge lawmakers to take it absolutely seriously because it even if we got a vaccine, being able to use it well, is one of the most important things, we have to do something with it. It’s not just getting a vaccine. It’s using it and using it appropriately.
BARRY BLOOM: Just a small comment. Again, the focus is on vaccines. Vaccines don’t prevent anything. Vaccination does.
BILL HANAGE: Good soundbite, Barry.
BARRY BLOOM: Thank you.
MODERATOR: Do you have any follow up?
Q: I do, actually. So I talked to the Association of Immunization Managers and they’ve sort of stressed that there is this an existing infrastructure, but they’re concerned about the involvement of the military through operation warp speed and how that could harm public trust in the process. Is that similar to what you’re saying? Are you saying that this existing infrastructure through CDC is capable of scaling up to the degree that would be necessary to get out enough vaccines for COVID-19? And do you agree that the involvement of DOD could hurt public trust?
BARRY BLOOM: In the distribution of vaccines, the supply chain issues are really complicated. That is associated with a cold chain where some vaccines have to be kept at four degrees. It is not clear, but it looks like RNA vaccines may have to be kept frozen. Those are things that, for the vaccines we know about, the Association of Immunization Managers and CDC have managed to handle well. I would say there is a complementary role for the DOD to fly large numbers of vaccines, fast, to every distribution point that is necessary. I don’t see the role of the military in going into local school districts and actually handing it to the vaccinators or physicians assistants and doctors offices or emergency rooms. I would hope they could be organized in a complementary way. And I actually have some experience in overseas stuff with Ebola and the military in their effectiveness at getting stuff from one place to another safely and quickly, in an organized way. They could be helpful here, provided that their role is defined. And my worry is whether the immunization managers and the CDC have the resources to take on 200 million vaccines from five different producers and get them out quickly enough.
Q: Thank you.
MODERATOR: We have about ten, twelve minutes left to go, and a lot more questions to go. So if we could try and get going a bit faster, if possible.
Q: Hi. Thank you for taking my question. So this is speaking to the genetics of the virus and a bit to vaccine potency in the future. My question is, can you guys speak to the latest research that was published August 4th in Science that suggests exposure to common cold coronaviruses can teach the immune system to recognize SARS-CoV-2? And what are the implications of that?
BILL HANAGE: And that’s a really good question. Barry, do you want to start or should I?
BARRY BLOOM: The answer is it is intriguing at a number of levels. But let me start with the simplest case. So we have four common coronaviruses that are involved in causing common colds. One of the first questions is, do they provide new antibodies to those viruses that seem to keep them coming and as mild infections on a regular basis, but do not eliminate them from the population? Could those antibodies cross react with the coronavirus and protect us? And almost all the studies, with some rare exceptions, have shown that antibodies, the surface proteins of any of those viruses are not protective, at least in test tubes against infection by SARS-CoV-2.
What this new work indicates is, however, that there are T cells that see common determinants on different antigens that are engaged and primed by seasonal coronaviruses, and that there may be something called immunological memory that would enable people to make a more rapid response. I have no way of knowing to what extent that is possible. We don’t have data on it. There are sure enough, people that have had common cold viruses that end up on ventilation and dying, to indicate that I wouldn’t put my life on T cells against a common cold virus keeping me out of ICU. But there is a possibility that they do provide some advantage. There is also a possibility that they could work in the opposite direction, that they’re primed to produce the wrong cytokines. And then when the virus gets out of control in some people, for whatever reason, they contribute to the cytokine storm because they’re already primed to make cytokines that are inflammatory and there’s no control on that. So the answer is there is T cell memory, and I don’t believe we have a good idea of what the all of that, if any, is in either protecting or enhancing the seriousness of this disease.
BILL HANAGE: I will echo that, and I will add, it’s not often that I necessarily quote journalists on this, but Ed Yong, The Atlantic, who is an excellent writer, wrote a brilliant piece yesterday in which he made the comment, “Immunology is where intuition goes to die.” I mean, this is the really crucial thing is that it’s a really important finding. But what it means, we don’t know. I mean, it could plausibly have some role in explaining why some people appear to be very mildly affected by disease, while others have it much more seriously, but at the moment, we just don’t know. One thing I would like to push back on is I have heard some places suggesting that this can explain why different regions have had different experiences with the pandemic so far, with some countries having a lot of death and lots of disease, others having less. And that is just fundamentally, completely impossible because you’d have to suggest that for some reason, different amounts of people are being infected by different amounts of cold viruses and so on. And we don’t have any reason to really think that, say, people in Germany are being infected with different coronaviruses, causing the common cold, than people in Britain. So it’s extremely important observation, its implications are not at all clear.
Q: Thank you.
MODERATOR: Did you have a follow up? Are you all set?
Q: No, I’m set. Thank you very much.
MODERATOR: Okay. Next question.
Q: Hey, thanks so much. I have one more tough question and thanks for your previous answer. I’ll try to be quick. In Arizona, tests have dropped, but things are also getting slightly better and we’re seeing fewer hospitalizations. So how important is continued widespread testing and why? And kind of what would the ideal for that look like?
BILL HANAGE: Firstly, I’m really glad that things are getting a little better in Arizona. I think the important thing here is to look at it from the lens that Barry put out. Are you doing testing in order to keeping score and monitoring the situation? Or are you doing testing as a way which is going to actually intervene and it’s actually gonna be a public health intervention? So more testing will enable you to get a better idea of the amount of community transmission which is going on and whether or not you’re seeing local upticks in particular places, as we’ve found, regrettably, in a lot of places across the United States. We know that this does not restrict itself to metropolitan areas, such as small towns as well. Testing in those small towns needs to be good enough to capture introductions, large clusters, and medium to large amounts of local transmission as well. Unfortunately, the question of what proportion of tests coming back positive is correct is a really complicated part because it depends on why you’re doing the tests and who you are testing.
I mean, just to take the obvious example. If some of these test results are from testing the baseball team every week, then we expect most of them to be coming back negative. And so it can end up inflating the negative tests over and above what they should be. So I would tend to say, you know, the thing about testing is to be clear about what you’re doing it for. And if you’re going to do it in order to test the amounts of transmission in your community, you’re going to need to have more testing, not less. But that in and of itself is not going to help you. It’s only going to let you know that something bad is happening.
MODERATOR: Do you have a follow up?
Q: No, that’s great. Thank you so much.
MODERATOR: Next question.
Q: Hey, how’s it going? Thanks for thanks for taking the call. This is a pretty specific question for Miami-Dade, but there is a model that I’m sure you both have heard of by Youyang Gu, that basically extrapolates out from deaths. It shows that Miami-Dade County has more than 34 percent of its population already infected by SARS-CoV-2, which would be higher than the New York City area. I know you all haven’t necessarily looked under the hood there, but it just seemed like a pretty astounding estimate. I spoke to Dr. Gu about this recently, and he basically said the number jumped out to him as well. But that’s what his modeling shows based off the deaths. And so I just kind of wanted to get that in context and get your thoughts on how much should we read into a model like that and how skeptical should we be that 34 percent of the population in Miami-Dade County, which has about 3 million people, have already been infected with SARS-CoV-2.
BILL HANAGE: How many deaths have been there so far? Because that might be able to make it see if it’s plausible or not.
Q: Yeah. Let me let me pull it up real quick. I believe it’s 6,140 last time I checked.
BILL HANAGE: So 6,140 deaths in Miami-Dade alone? If you multiply that by around two hundred and then you ask whether or not that’s that similar proportion of the county, then it seems like it might be on the nose. With Youyang Gu’s model, it’s an interesting one because, actually, it’s an SEIR model, so it’s a mechanistic model, which actually accounts for transmission. But it estimates parameters using machine learning. So all of these models have complicated parameters. Like how likely you are to transmit, how likely you are to die, and so on. And it estimates using machine learning. Now, models are only as good as our assumptions, and machine learning is only as good as its data set. So while it is a model, I actually look at a lot and I take quite seriously, I would expect it to occasionally throw some interesting things. On the other hand, not many deaths in a county is consistent with a relatively large number of people being infected. So I’d just be interested to know what the arithmetic shows.
Q: Yeah. And so I apologize. I was thinking of statewide deaths, so currently in Miami-Dade, it’s 1,724 currently and it projects 3,392 by November 1st.
BILL HANAGE: And that’s the number of people who are are immune or who have been exposed already.
Q: So those were deaths.
BILL HANAGE: No, of the 38 percent.
Q: Well, it actually it looks like it changed a little bit since I last looked at it, but it now says 32.8 percent as of August 5th.
BILL HANAGE: In the interest of time, it might be easier if I just correspond with you after this because this will take some calculation.
Q: That would be great. Thanks a lot. I appreciate it.
BILL HANAGE: Yes.
MODERATOR: Next question.
Q: Thanks so much for doing this and taking my question. My question is about the safety of reopening for schools. And I don’t know if there can be an overall answer to that question. If those schools are safe to return for the entire nation, that may be unfair, but I am in Texas, specifically, I am covering Harris County, which should probably pop up in your head as one of the hot spots. We also have a hot spot in the valley near the border. So maybe Texas focused, I mean, do we have the right data or what does the data show regarding the risk of reopening schools here in Texas?
BILL HANAGE: I would say in Texas, schools should not be reopened, at the moment. And I want to emphasize that schools should be a priority in any plan. I’m trying to step away from the phrase reopening, by the way, and try to introduce the phrase pandemic management, because I think it’s in the more responsible way in communicating what’s actually going on. It’s a mistake to think you can have everything, but schools are really important. Now in terms of explaining what happens in schools, and this is relevant to other people on the call, it’s very clear that we need to think about age groups differently. So adolescents, college students, they seem to transmit the virus a lot. We have that by direct observation of clusters. We also have it by looking at serology, meaning the presence of antibodies from the spring surge, which show that this age group tends to get infected quite a lot. We don’t observe it very much because of the fact that we weren’t testing very much at the time. But when we did look at major surges in Florida, for instance, we found that the age group was being infected quite a lot.
Now, the contrast with that is a younger age group. And that’s a very, very complicated question to which there are no really coherent, consistent answers on the moment. And the reason for that is twofold. Firstly, almost everywhere schools were shut at that age group, the elementary schools, quite early on. So we don’t actually have much data on how much transmission is gonna be happening within them. Even if they have been going on within them, because that age group is much less likely to show severe symptoms of disease, they have very much less likely to be detected. Now, I think it is reasonable to suggest they probably are somewhat less likely to be infected. This is partially as a result of the actual amounts of the receptor that the virus binds to among the younger children. I think it’s reasonable that they’re less likely to be infected and somewhat less likely to transmit, but not to the extent that we can rule out the potential for schools in that age group to actually be sources of infection.
Now, as a caveat on that, the fact that there are places in the world that have managed to open schools with very little consequence, if any, for the overall course of the pandemic. And I would look at Denmark as a great example. The thing about Denmark is that Denmark has never had a huge amount of disease in the first place. And if there’s not a large amount of community transmission going on, then the probability of it being increased to school and one of these clusters of infection starting, that is correspondingly, extremely low. Whereas, if you’re in a place where there is a relatively large amount of transmission going on, the probability of the virus entering a school is actually quite high. Once the virus is able to get into the school, if it does transmit, and that’s an opportunity for it to potentially get into a number of other households. So it’s a complex question. The science is evolving as I speak, and that’s a bit of a moving target. But I feel very strongly that if community transmission is at the level that you’re describing, it is not safe to open schools. And I’ll let Barry say what he thinks.
BARRY BLOOM: I think that’s a really good answer, and I sometimes listen to lots of experts who have very strong views on this, for which it is not much of an evidential base. There are anecdotes and analogies. So Denmark had a good record. Sweden also had their schools open and kids did fine. The epidemic was no different than in Denmark, except that there was a five to eight times higher death rate among people in the elderly population. And one interpretation is that grandma and grandpa catch pneumonia, pneumococcal pneumonia from their grandchildren, maybe that’s a source of infection in Sweden. And to pursue Bill’s point, in a country like Israel, which has an ongoing, fairly high transmission rate, when they opened the schools, it was a nightmare and there is no question the schools contributed. So I would take Bill’s advice as the best advice. It depends on the context in the district in which you’re making a decision. You can’t make it on a national basis. If transmission is high, kids will contribute to transmission and be involved. If it’s low, you have a good chance, particularly for younger kids, not to make a big contribution, but you really have to get better data than we have.
BILL HANAGE: Yes, I will echo what Barry said about Sweden. Sweden actually did close high schools and universities, but it kept elementary school schools for young kids, what they call compulsory schools, open. And as Barry says, the per capita mortality rates in Sweden is enormously higher than it is in Denmark.
Q: A quick follow up question, if I may. So right now, Texas in some districts have set a special date for in-person instruction and some are going back like next week. So is it fair to set a in person start instruction by, say, September X? Based on the fact that we don’t know what the virus is going to do by September X?
BILL HANAGE: I wish people would stop pretending they know what the virus is going to do. I wish they would. I mean, come on. You know, I find it not enough going out beyond a few weeks. The pandemic is not going to play ball. The only impact that you can have on the pandemic is the actions we take in order to preserve public health. That’s it. The virus is not going to read that and sort of think, oh, wow, it better back off because they need schools reopen. No, what you have to do is you have to figure out how to allow an amount of transmission within the community that you’re comfortable with. That might be very little. If you want to be say, New Zealand or Australia, and you take the corresponding action, whereas essentially, if you are going to do something on a date at some point in the future, you have to be prepared for the situation to change. This is something which we have seen again and again and again and again. And I really profoundly wish that people would stop expecting the virus to cooperate.
MODERATOR: OK, I’m going to break in here real quick. So we’ve run over already, but we still have questions. I don’t know if you want to stay on longer or if you need to go.
BILL HANAGE: I don’t have anything immediate. But I have got a lot of people waiting on me since I started my Zoom calls many hours ago. But OK, we’ll try and be quick.
MODERATOR: OK. Because I’m going to ask you this one, it’s from an independent journalist working on a story about a group of educators who are pushing for officials to consider an alternative back to school scenario in which the youngest grades, K through three, return to school in some form of outdoor learning in locations where that’s appropriate and positivity rates are low. The concept is to be able to phase in older grades once infection rates are lower, but perhaps until there’s a vaccine. This would be an alternative to returning to enclosed classrooms and or as a means of extending school space, six feet of separation. Does this concept seem safe or feasible? Other doctors have mentioned that it might be an option in communities where PCR positivity rates are around one out of five hundred and she’d appreciate hearing any input on that concept in metric.
BILL HANAGE: That’s a very interesting idea. It’s the kind of creative idea that is likely going to be needed if we’re going to be moving forward. I hesitate to say that it would work everywhere for reasons similar to those which I was just saying. However, I do think that’s an intriguing idea. And remember, like I said, the PCR positivity test rates reflects a whole lot of things from the kind of people who are being tested to whether or not you’ve been investigating nursing homes recently. So you should be careful to not get hung up upon that. However, if you want to talk about allowing in-person instruction for younger children, I think they’re less likely to transmit overall, but then much less likely to transmit if you’re going to be doing it outside. So I think that innovative concepts, innovative plans, are something which is going to be really helpful. And I will point out actually there are records of people doing outdoor teaching in the 1918, 1919 influenza pandemic.
MODERATOR: OK, great. Thank you. Next question.
Q: Thank you so much, everyone. Quickly, there seems to be an increasing call for a new national testing strategy of moving from the flow insensitive PCR test to the fast and frequent antigen tests. Understanding all your points about testing, and it is more than just testing, we need isolation and quarantine, but is this a strategy that you would recommend to get out of our current backlog and problems that we’re having, particularly here in California.
BARRY BLOOM: So, it is clear to me that the PCR test, which has high specificity and sensitivity, once the numbers get beyond the numbers where you can do tracking cases and contacts, even if it were turnaround times in 48 hours, it’s not going to keep up with this epidemic. The antigen tests are easy to do. They’re fast and they’re not very sensitive. The good ones, maybe about 50 percent of the small do it yourself kind. You can go and send samples to places that have higher specificity, sensitivity. But there, in my view, not the ultimate answer. And so a lot of us are pinning hopes on new molecular tests that would be doing at home tests. And I think what I have to emphasize. What one needs is not a test to know if you’re positive or negative. In college students, in high school students, in school students, in a business, you need to test people every other day, every three days, once a week. And you can’t do that. PCR is 100 dollars a test. You can’t do it. So we don’t have a public health acceptable test at this point. The NIH put out an urgent call for new ideas called RADX for rapid something or other accelerated diagnostics. They have already funded a couple concepts from biotech companies. They are reviewing more. I am hopeful that some of them will be good enough to be able to be approved by the end of the summer. And that would make a transformational effect in people able and businesses able for a buck or two a test, per person, for people to know whether they have been infected or not, and then self isolate. And without that, I think we are in deep trouble unless the numbers come down by ordinary public health needs.
BILL HANAGE: I think that is a really smart comment. I will add to it in a couple of ways. My colleague, Mike Mina, has been eloquent about the potential benefits of a high specificity, low sensitivity test, meaning that if it comes back positive, you definitely have COVID. If it comes back negative, you might not of have it, because of the fact that it would have the ability to be informative. And if it’s cheap enough, as Barry says, you can do it for a few bucks a day. It’s transformative for business. I want to point out one other thing about this, because it holds a place in my heart, and we’ve written some of a paper on it. It’s a particularly useful strategy if you have a virus which, like this, tends to transmit in clusters because even if it has low sensitivity, if that has been a cluster of transmission and a large number of people have become infected in a workplace, that’s something where it is extremely unlikely that all of the tests on them will come back with a false negative. Then if you have a positive, you’re then informed that something’s been in place there, a transmission case or potentially a transmission event, and once you have that, you can actually start taking directed action to take all of those individuals, isolate this period of time, and we’ll come back and testing with an alternative and better test.
Finally, a point which I think would be profoundly useful, which you alluded to, a national testing strategy would look completely different to what we have now. When people like me are sitting there trying to figure out, talking to people like Ben about what’s going on in Florida, I’m thinking, hang on, what exactly is Florida counting? Then it’s, oh, you don’t count people who are non Florida residents. What’s with that? And then trying to figure out how that will alter the numbers. It would be a profoundly helpful thing to people who are trying to understand the epidemiology of the pandemic. But the bit in which Barry is much interested in, the bit about making a public health impact, yes. Those tests, those kind of innovations will be really, really helpful.
Q: Perfect. Thank you so much. Just so I’m clear, it is my understanding with the antigen test, some of which have been distributed in nursing homes, that do have like a 15, 20 minute turnaround, can that meet the criteria for some of the innovations that you’re describing?
BILL HANAGE: Yes. That kind of a 15, 20 minutes turnaround is profoundly better than those sort of seven to nine days that we’re seeing from some commercial PCR testing facilities.
Q: Perfect. Thanks for your help.
BILL HANAGE: Thank you.
MODERATOR: I think this is our last question.
Q: Hi. Thank you so much for doing this and staying on a bit longer. There’s a study published in JAMA today of 300 COVID patients in South Korea. It found that asymptomatic patients had similar viral loads compared to those who did have symptoms. Does this line up with what we already have known about COVID or have suspected? And what are the implications for the response in the US?
BILL HANAGE: I think it lines up completely and I’ll be honest, I haven’t read the article yet. But if what you’re telling me is correct, it’s completely plausible, especially given what we know about the fact that there is a significant amount of transmission from people who are currently asymptomatic. And I’m not sure I haven’t read the study. I’m not sure if these are people who were continuously asymptomatic or if they developed some mild symptoms at some point. But, we have known for some time that unwitting transmission is possible from people who are currently not displaying symptoms, or displaying symptoms which would not be associated with COVID infection. Because, you know, this is a virus that transmits by the respiratory route, but its effects are systemic and it has a very, very wide range of presentations. The implications, what we do in the United States and in the world is that interventions which are focused on treating everybody as if they could be infected, all the most important ones, because to be honest, based on symptoms at any time, anybody could be infected. And that relates to the sort of non pharmaceutical interventions that we’ve been talking about all of this time before we start getting into vaccines.
Q: Thank you.
MODERATOR: One really quick follow up question. She’d like to know what your experts believe should be the transmission rate in order to safely open schools in a community.
BILL HANAGE: I don’t think you can actually put a simple number on that because of the fact that each community has different amounts of testing that are being done in each place are different. I would in doing so and thinking about that, and again, remember, it’s different ages are important here. I think that it’s very, very difficult to think about opening high schools if there’s the risk of introduction. But what you could do is you could ask your local community and ask yourself, well, what risk are we prepared to take? And then look at an estimate which will be different from the actual testing rate, remember, because the actual incidents within a community, meaning the number of people who are infected at a given time point, is not necessarily the same as the people who are getting tested. That’s only an estimate. Then, ask what and perhaps take when somebody comes in here with the virus? And then think that through. I’m sorry, I can’t be more specific, but I’d be concerned if I were to be more specific, it would be taken off to places where it’s not really appropriate to do so.
MODERATOR: Thank you. Dr. Bloom, did you have any final thoughts before we go?
BARRY BLOOM: No, I thank you all for very good questions and it’s always a pleasure to hang out with Bill.
BILL HANAGE: Likewise. I like our sort of sartorial contrasts.
MODERATOR: Dr. Hanage, do you have any final comments?
BILL HANAGE: No. It’s a pleasure to hang out with you and Barry and everyone on the call and I will do this again in a few weeks.
This concludes the August 6th press conference.