Coronavirus (COVID-19): Press Conference with Thomas Tsai, 11/05/20

You’re listening to a press conference from the Harvard School of Public Health with Thomas Tsai, assistant professor in the Department of Health Policy and Management and a faculty member in the Harvard Global Health Institute. This call was recorded at 12:00 p.m. Eastern Time on Thursday, November 5th.


MODERATOR: Dr. Tsai, do you have any opening remarks?

THOMAS TSAI: I just wanted to start with a high-level overview of where we are with the pandemic, as well as cover a few of the recent developments to our COVID-19 risk-level dashboard on According to our risk-level dashboard, there are currently twenty-eight hotspot states, those are states with over twenty-five per hundred thousand new cases. That’s been an alarming trend over the last several days as well as several weeks. Over this past week, we’ve seen over a million new cases with a seven-day moving average of the daily tests, positive cases of over eighty thousand. Yesterday, we hit the unfortunate milestone of having more than one hundred thousand new cases per day. Our test positive rate remains high and it’s climbing with eight point five percent test positive rates. And now we’re averaging over nine hundred deaths per day, and we’re seeing with each day over a thousand deaths per day. Our testing has increased, but it has remained relatively static over the last several days. We’re now averaging approximately one point two million tests per day, an improvement from the earlier part of the pandemic. But these alarming numbers really suggests that we need to double down and really focus on our public health response. In terms of the updates to our COVID-19 Dashboard on, which is a collaboration between the Harvard Chan School of Public Health, the Harvard Safra Center for Ethics and the Brown University School of Public Health. We have launched two additional views over the last week. Approximately two weeks ago, we created a congressional district level view of the risk-level dashboard that offers two advantages. One, it does standardize the risk levels to its population because a commercial district, as we know, is fixed on a population level, but does vary in terms of geographic size and takes away some of the variability we’re seeing with some of these smaller rural counties with smaller populations. It is quite alarming that we’re seeing approximately 4 in 10 or 40 percent congressional districts across the country are now current COVID hotspots. The second reason why this is the relevant view is because the congressional district is an actionable unit of geography, and in light of our recent election, this is a time for policymakers to focus on the severe pandemics in our own backyards. The second part of our dashboard that’s been launched last Friday, is an update to our testing dashboard by including a testing calculator. We’re in a phase of pandemic where we can move from a more reactive diagnostic approach to testing of basically confirming symptomatic individuals or contacts to a more proactive and offensive role of screening asymptomatic individuals. But in order to do so, we need to be able to prioritize the groups of people we want to screen, as well as the frequency we want to screen for. And the testing calculator provides an actionable, data driven tool for local and state leaders to be able to calculate based off their own demographics in their state, the number of teachers students in K-12, for example, a number of university students, a number of first responders, health care workers, nursing home residents and staff and highlight how often they want to test these individuals and produce a number of what the testing strategy, the number of tests that would be needed to pursue such a strategy. And I’ll close my comments with a comment that Dr. Redfield from the CDC made yesterday was that, now is the time to develop a national testing strategy. And I agree with that. But I want to push it one step further and say we have been developing a national testing strategy for the last several months. Now is the time to actually implement the strategy, especially with the election behind us, two days ago. And one potential strategy is a bill that’s currently in Congress right now, which has bipartisan support, which is Suppress COVID-19 Act, which provides federal funding to support state efforts at testing and contact tracing. So I’ll stop there. Happy to take any questions around our testing strategy as well as our risk-level dashboards. Thank you, Nicole.

MODERATOR: Thank you, Dr. Tsai. OK, first question.

Q: Thank you for taking my call. You mentioned that yesterday the US hit the milestone of one hundred thousand daily COVID cases. And I just want to know, how do you see this trend going from here? Will those numbers continue to rise? Will they stagnate? Or can we anticipate a drop? I know in an earlier talk, Dr. Mina said he thinks we might see two hundred thousand cases a day in a month or so. Does that track with you as well?

THOMAS TSAI: Thank you. I think that’s an important question. And what’s clear from the data from the last several days and weeks is that we are still on the steep part of this unfortunate rising curve. We have not rounded the corner yet. We have not hit a peak of this current wave. That said, what I do hesitate to say in terms of what numbers will be in a month, because that number depends on what happens over the next days to weeks. One thing we’ve learned from this pandemic is, for better or for worse, the case counts, the hospitalization rates, the deaths, even. None of this is written in stone and it depends on our public health response in terms of testing, contact tracing, supported isolation, masking and physical distancing. So the models can produce a number, and the 200,000 numbers have been projected in certain models. But I think the message for our general public is that this is a time to make sure that we don’t hit that number. And this means following through with the guidances that are out there, but also means for our public health leaders or policymakers, to really focus on implementing a testing strategy now and in places that are hotspots, realizing that social distancing policy is not an on-off switch and that this is a dial that needs to be calibrated to the temperature. And in a lot of these places, these are hot spots and we may have to reinstitute more localized or focused physical distancing, such as bans in the indoor dining or restaurants in order to put the lid back on the pot that’s boiling over again. So it’s a very alarming trend. But as with the pandemic over the last several months, this is within our control to be able to suppress it. But we have to take decisive action now.

Q: All right, thank you very much.

MODERATOR: Next question.

Q: Hey, Dr. Tsai, thanks so much for doing this. Kind of a follow up on what you were just talking about, the need for perhaps some localized interventions. Are you seeing those happen anywhere, really? I mean, if there’s a need for them or are they happening, basically? And why aren’t we seeing, whether it’s governors or mayors, sort of taking interventions that would be required to stem the increasing transmission right now?

THOMAS TSAI: I think it’s important for our local and state leaders to understand that, you know, the train hasn’t left the station. We can always dial up our physical distancing response, as well as our masking. And really the most important strategy right now is also to rapidly scale up our testing strategy to move from the defensive approach we’ve been relying on the last several months and more offensive, by using screening tests and point of care test, which is antigen test, to break the silent chain of transmission among asymptomatic individuals. You know, I think there’s been pandemic fatigue among the general public and individuals. And my hope is that there isn’t also a pandemic fatigue among our policymakers, but certain states are leading the way. In our own state in Massachusetts, there’s been a reinforcement of some of the physical distancing in terms of not having dining after 9:30 at night, and more limiting the high-risk encounters. And I think this is a good example that other states need to follow is, you know, being able to calibrate and adjust your social distancing policies based off of where you are currently in the pandemic. And that’s where the risk level dashboard is very helpful to show our local and state leaders the direction that the pandemic is heading in their own areas. But I think it’s incredibly important and this is the time to take decisive action now. And part of the strategy is also from the federal government side to pass the funding and legislation that’s been in debate for the last several months, and be able to support the efforts on the state and local levels, not of state budgets or out of money to pay for testing. This is where part of the strategy isn’t just a plan, it’s also having the resources allocated to the communities that need them to be able to provide for testing and supported isolation. So this is really the moment. And the actions of individual municipalities, of states, regions, and collaborations across states together can help break the trend of the pandemic on a national level.

Q: Just because you mentioned, I can’t remember if it’s called the curfew or not here in Massachusetts, but the idea of restaurants closing at a certain hour. Is that actually an effective measure just because it doesn’t change the underlying activity? It just sort of limits the hours that that activity can occur, because you can still eat indoors. And whether you’re there at 8 or 10 p.m., that shouldn’t really necessarily matter, right? Or maybe it does.

THOMAS TSAI: That’s a good question. And I think it’s important to think about the risk of community transmission as a gradient, a spectrum of risk. You can both decrease the likelihood of individual action transmitting COVID. And things like six feet of distance, wearing masks could also change the frequency of those interactions, which is what the limited hours get at is, you’re also limiting the frequency of contact among bar or restaurant patrons, for example, so that can be effective. In the research that we’ve worked on in the collaborative project with a team from Google, we have analyzed the changes in mobility from people’s cell phone and data that’s been anonymized and differentially privatized. What we’ve seen both in the United States and in Europe as well, is that about a 10 percent reduction in social mobility or movement translates to about a 10 percent decrease in the number of COVID cases. So decreasing the amount of interaction through bans on dining or large gatherings can be effective. It was effective in the first phase of the pandemic. And now is a time to think about a lot of these hotspot states and hotspot counties and hotspot congressional districts, if that’s what’s also needed to mitigate the pandemic again.

Q: Thank you very much, appreciate it.

THOMAS TSAI: Thank you.

MODERATOR: Next question.

Q: Hello, thanks for taking my call. Can you hear me OK?


Q: OK, great. We’ve seen some lockdowns, countrywide lockdowns and sweeping lockdowns in Europe and England, but not here. Is there a sense that we would go back to those kinds of things, even in states, the lockdown? Or what do you think of that? And you kind of talked about it a couple of minutes ago, but I thought to just get your thoughts on whether that’s something that might happen here and why or why not?

THOMAS TSAI: Right. You know, I think we have more information and more tools in terms of our ability to mitigate and eventually suppress the pandemic. But, you know, thinking back in March and April where when the U.S. was very focused on flattening the curve, we didn’t have the capacity for widespread masks at the time. So now we have masks and we have more tools in our armamentarium. So, in terms of a national lockdown, you know, I don’t think we’re in the same place where we were in terms of our ability to respond as it was back in March and April. But we do have the opportunity for more focused interventions. And I mentioned bar or restaurant bans as an example, enforcing universal masking as a way to prevent needing a national lockdown, ramping up our testing and our screening of asymptomatic individuals to pick up silent spreaders of COVID-19, are ways to prevent a need for a more national or widespread lockdown. So we have tools that just weren’t at our disposal back in March where we had to shut down nationally. Our hospitals, locally some hospitals are hitting capacity now, but it’s a different stage where we were in March and April. Hospitals have had a chance to prepare from both a supply chain standpoint in terms of masks and respirators, but also from a human capital resource standpoint. We have better clinical algorithms for how to prepare and treat COVID-19. So if we can use a lot of these more targeted interventions decisively, in some ways it’s thinking about using a sharp or finely tuned instrument as opposed to a larger, blunter tool. But you have to use those sharper instruments in the right window of opportunity. And I think that’s the important part, is we don’t want to miss a window where the hundred thousand cases that we’re seeing, become one hundred twenty-five thousand, one hundred fifty thousand, two hundred thousand cases, at that point, we may need a wide scale lockdown. So, again, the pandemic is not written in stone. We can act decisively to control it. But I think this is a moment where this large basket of smaller interventions can be effective and needs to be implemented in a lot of the hotspot areas.

Q: Thank you very much.

MODERATOR: Next question.

Q: Yes, hi, thanks for taking our questions today. You know, Maine, up until the last month or so has seemed fairly insulated from the worse effects of the pandemic, but our numbers are going in the wrong direction. And I’m just wondering what you think of it. At what point is there any kind of rule of thumb for when exponential spread gets to the point where the smaller measures don’t work? Are we talking, say, you know, 10 percent positivity rates? Or is there any metric that you can point to for that?

THOMAS TSAI: Thank you. And I think Maine is a great example of the idea that no community is an island when it comes to the COVID-19 pandemic. And while Maine had relatively lower rates of COVID-19, we all remember the example of the wedding in Maine and how quickly that was able to spread and impact the safety and lives of people in that community. So I think that’s important to know for all areas across the country, whether it’s a more rural area or a more urban area or an area that was previously relatively spared, is that no community is an island. And in our risk-level dashboard, on the county level view, there were isolated counties that were green or yellow. But when we aggregate that up to a congressional district, we see that those counties were actually surrounded by hot spots in that entire district, was actually at high risk for accelerated spread of COVID-19. So it’s important for more isolated areas to not be complacent because the impact, now, should be even more profound. If you’re a rural area, having one COVID or two COVID cases can quickly overwhelm a small or critical access hospital. But in terms of the metrics, I think we are looking at the same basket of metrics that we were looking at earlier in the pandemic. And that is not just the number of new cases, but the trend in the growth of the new cases and how fast that the trend is accelerating. And the analogy I like to use is, it’s important to think about not just how fast the car is going, so the speed, going 60 miles an hour versus 30 miles an hour. But are how fast are you accelerating? We call that the second order derivative in math terms, but we have an internal dashboard that looks at how fast are you accelerating? Are you staying at 60 or are you going 60, becoming 80, 90 miles per hour? And I think that’s important to think about. How fast the pandemic is spreading is not just the number of new cases, but also the rate and the acceleration in the rate of COVID cases. That test positive rate, again, has become an important number. We have made big strides over the summer with some states like New York and Massachusetts having test positive rates less than one percent. But now we’re seeing our national test positive rates heading in the wrong direction, being about eight point five percent. So the test positive rate, again, is an important metric for states to consider. And then we still have our lagged indicators of hospitalizations as well as deaths. But the idea is to not wait for those lag indicators to become severe before we act, we need to respond in some ways to an early warning system. That’s where the testing data remains a critical piece of our public health response. So I think all those things together gives a sense of the directionality of where the pandemic is heading. So I think these are the metrics that we still need to be considering.

Q: If I could just have one follow up here, and I could just throw out some numbers for you and see where you think Maine is. Our PCR tests per one hundred thousand is right around six hundred. Six hundred PCR tests per one hundred thousand. And then we just announced a rapid antigen testing expansion at Walgreen’s where pretty much everybody can get an antigen test and get a result within 15 minutes. That’s going to start in probably two weeks. Our positivity rate is about one point three percent and we can still contact trace everybody who tests positive and we’re beefing up our contact tracing workforce through Maine CDC. Does that give us a chance to maybe turn things around? I’m curious what you think about that. Thank you.

THOMAS TSAI: Yeah, no, that’s a really important question. I think Maine is heading in the right direction. And Maine has had a very robust public health response and has had very robust public health leadership in its department public health as well as its state CDC. The increasing testing is Maine is ahead of some other states nationally. But I think what’s also important is the second part of what you mentioned is a move from the more reactive testing strategy in Maine to a more proactive one around screening asymptomatic individuals. I think this is a moment to really implement that policy. The technology that we’ve been waiting for months has arrived. We do have these antigen tests now. They are cheaper, they’re faster and in some ways more effective at identifying individuals when they’re most infectious and likely to spread COVID-19. And the second part of that is also what we call the key performance indicators of how effective the contact tracing efforts is. You know, we’ve moved from just discussing structure, the number of tests, number of hospital beds. But now is really the time we need to measure how well our processes are working. So it’s one thing to have contact tracers, but what proportion of individuals are able to actually reach? What proportion of the contacts from an identified case are you able to reach? What proportion are actually getting tested? Of those, how many of those are getting results back within forty-eight, within twenty-four hours, to be able to meaningfully change their behavior and undergo supported isolation to break those chains of transmission? So we need these key metrics of how well our interventions are working. And obviously we also have to rely on the outcomes. So, you know, in pre-pandemic times, most of my research was focused on health care quality and health policy, we think about structure, a process and outcome measures. And I think this goes back to the earlier question is the basket of metrics we need to look at. We need that full basket of metrics that tell us structurally, do we have enough resources to have an effective strategy, the process measures to guide how effective is our strategy actually working? And ultimately, we have to rely on the outcomes, but the outcomes can be lags. We have to be able to react to some of these key performance indicators, as you mentioned.

Q: Thank you very much. I appreciate it.

THOMAS TSAI: Thank you.

MODERATOR: Next question.

Q: Hi, thanks for taking my question. So when we talk about a proactive testing strategy and sort of moving from the defense to the offense, I’m wondering what specifically sort of needs to be done, because we have more testing now, but it certainly sort of not ubiquitous in that anyone can sort of get a test at any time. So do we just need to be smarter about where we’re using the tests that we have? Do we need to get diagnostic companies to ramp up their production to make more a point of care tests? What do you see sort of needing to be done on the testing front?

THOMAS TSAI: Yeah, I think you’ve hit the nail on the head. And this is the moment where we’ve been thinking mostly in terms of our public health response for testing on the supply side, the number of tests, the number of swabs, but now is when we need to focus on the demand side of the testing. So it’s not just about getting the noses to the swabs, we have to get the swabs to the noses and as opposed to the noses, the swabs. And that means, you know, being able to load balance. So there are areas where there is a high capacity for testing, but a roughly smaller outbreak like the Northeast, whereas you have areas in the upper Midwest where there’s very high levels of COVID-19 cases, but fewer resources for testing. So this is where a load balancing is really important. And in the absence of a coordinated federal response to do that, this is where regional collaborations across states can be a very effective. The Rockefeller Foundation, for example, has led an effort. And now there’s 10 states that have engaged in an interstate compact or regional collaboration to help signal to the test makers that there is a ongoing demand for testing. I think that’s been part of the challenge, as well, between the supply and demand side is that there’s been very mixed signals for most of this pandemic to a test maker about what the role for testing is going to be. But I think it’s very clear that the testing was necessary, is necessary now, and is going to be necessary in the future because the vaccine has not arrived at warp speed. And even when it does arrive, it won’t be available to all individuals immediately and reach all areas immediately and it won’t be 100 percent effective by itself. So the testing remains a core part of a public health response. And we need to be able to signal that to our testing manufacturers. And that can be done through the Defense Production Act by pre-purchasing tests on a federal or state level. The federal government has purchased the antigen tests, 150 million different tests through the Abbott BinaxNOW test. That’s one example. But on a regional scale, there’s also important ways to advance that through what we call an advanced market commitment, to signal to our test makers that this demand is going to be there by the time that you make the test.

The third piece of the puzzle is what the tests are going to cost to the individuals. And most of the testing is still focused on a medical indication for testing, so, relying on insurance reimbursement, a doctor ordering a test. And as we moved from this more proactive stage of testing of screening asymptomatic individuals, we also have a change in how we think about how we pay for testing and for a testing strategy to be effective. It has to be not just available, but also affordable. And we still hear stories about people being charged hundreds of dollars out of pocket payments from the COVID-19 strategy and, you know, 10, 11 months into the pandemic, I hope we can do better. And one version of what doing better looks like is, as I mentioned at the top of the call, is this bipartisan effort sponsored by Senator Cassidy from Louisiana, Senator Smith from Minnesota, the Suppress COVID-19 Act, which provides 50 billion dollars for testing and supported isolation to shore up the demand for testing. And 5 billion of that is specifically allocated for states to cooperate together to be able to bulk purchase tests and load balance. Because if you’re a small rural state, we don’t want to be in a situation where we are in March where you have larger states bidding against smaller states, bidding against the federal government. So part of the bill also incentivizes regional cooperation across states. So there’s a way forward to this. But I think we need to shift the way we think of our testing from being limited by a scarcity mindset to, you know, thinking from not just supply-side solutions of the number of tests, but thinking about how we can increase the demand of testing. And third, is making that also affordable to the general public, but also affordable to the states, because lots of states have very tight state budgets currently. And that’s where the federal help can come in. And there are bipartisan solutions that are out there, and they’ve been discussed and debated. But I think, you know, as we’ve moved past Election Day, this is a moment for the nation to collect around is, that this is not a political decision. This is a public health crisis. And we should let the data and let science guide the way in terms of how to respond to this growing pandemic.

MODERATOR: Do you have a follow up?

Q: Yes, one more question is about at home testing and sort of how when we see those coming to market, we’re anticipating those maybe by the end of the year, in the next couple of months. How do you picture that fitting into the public health response? And what’s the best way to use those tests, sort of given the opportunity for, I’ll call it sort of chaos, if we just have them sort of freely in the marketplace? What is the best way to sort of handle that?

THOMAS TSAI: Yeah, I think that’s why we need to have a strategy. And it’s not just about a free for all. And even with the antigen test as an example, now, you know, part of the challenges states and municipalities are facing is how do we take the data, the test positive from the antigen test, and feed them back into the public health system. We’re relying on molecular tests which were performed in labs with existing data pipelines from the diagnostic labs to public health agencies was a lot easier to collect that information. But as testing becomes more widespread and in the communities or even in people’s homes, is how do we still maintain that data collection ability to one, track where the pandemic is headed, but also, two, is how do we still provide public health and medical guidance and information to people who do test positive when the test results are happening at home? So I think, again, this goes back to the earlier answer is yes, the number of tests and where they’re delivered is important. But it’s also just as important to have the strategy and the context for that testing. And this is the time to really develop the data infrastructure, the public health information and guidance infrastructure, to prepare us for the moment when we do have enough tests, when they’re cheap enough to be performed at home, that we are set up to take advantage of that. And I think in some ways we’ve always been playing catch up with our testing approach. And this is a time where we have to be more forward-looking and be able to have the resources in place when these tests become available, to actually be able to implement that in a meaningful way.

Q: Great, thank you.

MODERATOR: Next question.

Q: Hi, thank you, Doctor. I was wondering if you’ve heard of any localities using that test with the abbreviation LAMP, which is that color-based test that’s similar to PCR. It’s the Loop-mediated Isothermal Amplification. I’ve heard that it’s cheap and it’s not perfect, but some college campuses are using it. And I’m wondering if you’ve heard of any municipalities around the country using that and if you think it would be a good idea. Thank you.

THOMAS TSAI: Yeah. So I don’t know specifically of any places that are using that technology, so I don’t have any specific insight to that. The bigger comment, though, I think is that there isn’t going to be a single test or testing modality by itself. Really what we need is a whole suite and a layered approach to testing and take advantage of all the different testing resources that are coming online. For example, you know, there is not just antigen test, there is a next generation genome sequencing that’s happening. So we need to take a layered approach to all the different tests. And that goes back to sort of the theme of this conversation today, really that means having a strategy around testing is as some of these point of care technologies like antigen tests, for example, if somebody tests positive, you confirm that with the PCR test or through a next generation genome sequencing test, that happens in the lab. So it’s really about how do we take all these different pieces of the puzzle together? And I was using, you know, a sports analogy before, we’ve been talking about, you know, do we pass the ball, or do we do we run? But it’s not about individual plays. We need a game plan in all four quarters. And I think that’s the part we have to focus on.

Q: Thank you, I do have a follow up, is that OK?


Q: Can you also give us just a very basic sense of what settings are currently driving the spread of the disease? And if it’s, for example, if it’s in households, is there any recommendation being thought about to have people mask even when they’re in their own homes? You know, it’s a little unclear to me like what settings people are getting sick. Because in Texas, we had a dramatic drop in cases after Governor Abbott implemented the statewide masking ordered, but it has leveled out and is now climbing again. Yet we’re all masking. So what are we supposed to do besides that? We do have bars and restaurants open, and I understand that that’s an option. But is there anything else that we could be doing if that is not going to be done?

THOMAS TSAI: Yeah, I think that that’s a really important question, and some of the contact tracing reports from the last several weeks have suggested that about a third of our COVID-19 transmissions are happening via household transmission or household clusters. And in conversations we’ve had with public health officials leading some of the contact tracing efforts. The concern is that people are more vigilant in the public about wearing masks and physical distancing. There are better protocols in place at schools or places of work or even retail environments or restaurants. But people are not being as vigilant in private when they come home and in terms of their social interactions with their neighbors or friends or families in their own neighborhoods and communities. And I think on the public side, there is a need to recognize ongoing vigilance, whether you’re in a public setting versus a private setting. So I think that’s one key paradigm shift that needs to happen. The second part, this goes back to our testing conversation, is the masking and distancing, that people are better at doing now, can prevent the transmission of COVID-19, but it doesn’t provide you the information to guide you in terms of once you’re home. This is why a screening strategy around testing asymptomatic individuals is extremely powerful, because, you know, you may be masking in public, but if you don’t realize that you may have tested positive or asymptomatic carrier, for example, you may run the risk of going home and exposing COVID to your own family members or friends. So, you know, the testing is a key part of that early warning system of being able to detect silent transmission of COVID-19. and being able to use that to inform behavior change, then breaks that cycle of transmission. Because if you know that you’ve tested positive on an antigen test, for example, you can safely self-isolate at home, versus sort of carrying on with sort of blissful ignorance. So I think the vigilance and the behavior is important, but two, is we need the information to guide people’s behavior that comes full circle again to testing.

Q: Thank you.

MODERATOR: Next question.

Q: Hi, Dr. Tsai, thank you very much for taking this question. I think that I heard you say that we need more targeted surveillance testing. And I’m wondering how much information there is about whether targeted surveillance testing is working. For instance, in Massachusetts, we have this campaign called Stop the Spread, where testing sites were set up in communities where there was a high case rate. But I haven’t seen anything that shows if the people who are getting tested in those communities are actually living there or in populations where it’s known that there’s a high risk, there is very little information out there. And in a state where we’re very well resourced here, we have tremendous amount of testing capacity. And most of these tests, they’re free. So how can we get to a place where we can really have targeted surveillance testing without having a better structure for knowing how well we’re doing that?

THOMAS TSAI: So I think there’s two ways to think about it. We can do targeted surveillance geographically, but in some ways that’s more of a reactive approach because you’re reacting to areas that have accelerated growth or are hotspots already. So in some ways, you’re trying to snuff out the flame, but there’s already a fire. The other approach is to take a wider scale view of screening asymptomatic individuals, and not focusing on geographies, but focusing on high risk or high priority populations. And I think schools is a really good example of that. While children in elementary school, for example, are maybe lower risk, but it may be a very important social priority to keep schools open. And if testing teachers and staff or even students, creates those conditions for a safe opening of schools, that’s also an important policy priority as well. So I think the other way of thinking is to move from a focus on testing high risk areas to testing high risk populations or high priority populations. And the college campuses are a good example of that, of a high priority and also high-risk population for COVID-19. And our testing calculator on global epidemics might help state leaders identify the proportion of individuals in their states that fit those categories and also calibrate the frequency of testing to produce a testing number. But ultimately, the heart of your question, I think is absolutely right, is that information needs to be transparent to the general public as well.

And I keep going back to our earlier conversations about when states were first developing their reopening plans back in April and May, about a phased approach to reopening and what the metrics were to guide that reopening. I think we need to refocus our efforts, as well, about in this current phase of the pandemic, what are the metrics that will guide our decision making about if the strategy is working. This goes back to some of the comments I made that, as you know, we need these key performance indicators, which is effectively what you’re saying is in these targeted testing approaches, how well is it actually working to identify cases? In some ways, it may be self-defeating and that’s OK, because if you’re testing very broadly and people are responding to that information and staying at home, if they test positive and the test positive rate stays low despite wide scale testing in a high risk group of individuals or high risk community, that’s a good thing. And that shows that the strategy is actually working. So I think that’s where we need that information to be fed back to overall policy. And within Massachusetts, you know, the states had focused testing. She’s mentioned stop the spread in key communities. And we saw over the course of the summer, a number of cases decrease or test positive rates go down over the summer. So that at least gives both on a city level, but state level of how well that strategy was working. I think that test positive rate in Massachusetts earlier on declining, shows the effectiveness of that approach.

Q: Can I have a follow up?


Q: Thank you. But although that may be true now, the number of communities that are now in the high-risk category, despite this focus testing, is increasing. It has increased every week and keeps going up. So still, the question is, who is getting tested? We don’t really know who’s getting tested.

THOMAS TSAI: Right, so there’s two parts to that, right. So one is we absolutely need better information on who is being tested, who is testing positive and the information from the contact tracing that follows the testing. So the demographics, the race, the occupation, the information about their exposures and clusters. All that absolutely is really important information. Our information gathering on that has improved, but it definitely could be better. The second part of that is also that testing is not therapeutic. Testing is necessary, but not sufficient as part of a suppression strategy around COVID-19. But testing is effective in so much as it informs a set of behaviors, both on an individual level as well as a public level, as actually suppresses the pandemic. On an individual level, that means undergoing supported isolation to break the chain of transmission. On a public and community level, that means if there is the testing data in a high-risk community or high risk group of individuals showing an outbreak or a hotspot. And that means putting in the public health measures to suppress that outbreak or hotspot and know in some way at least, we’re seeing cities react to that. You know, Boston in terms of closing the schools down, we’ve seen that in terms of the bar and restaurant curfew on hours. We’ve seen calls for that and more national level in terms of widespread universal masking. So the testing is absolutely necessary, but it is not therapeutic. It is not sufficient in of itself. And I think that’s why we have to put all the pieces of the puzzle together. And my worry is that on the public health policy side, there’s a tendency to narrowly focus on one intervention or one solution. But really, it’s about a change or a sequence of interventions that together work to suppress COVID-19.

Q: Thank you very much.

MODERATOR: Next question. She would like to know; your data shows Texas is way behind in testing. El Paso has surged in cases. The state filed for an injunction to stop the county judge from imposing restrictions. What are you seeing in this area? What should they do on a safety standard? And what can the rest of the folks in Texas learn from what’s happening?

THOMAS TSAI: Thank you. I don’t have specific details on El Paso or that part of Texas, but I think we’ve seen the failure of the state interfering with more robust local efforts. And, you know, one example is what happened in Arizona earlier, where the governor prevented city leaders from imposing mask bans and having to backtrack on that. And we’ve seen the failure of a patchwork response across communities and we’ve seen that unfortunate story repeated over and over again. And this is why states need to coordinate a public health response across their municipalities, across their counties, because no community is an island, but also means that local leaders need to have the autonomy to be able to enact more robust measures in their own jurisdictions, if needed, to contain a very local outbreak. So one is the local public health leaders need to be supported if they’re making these hard decisions to impose local shutdowns or masking orders because they’re trying to respond to a local outbreak. But then, two, is because no community is an island. We also need to coordinate these efforts across local jurisdictions and even across states because we’ve seen the wildfire, the pandemic march from the northeast down to the south, to the upper Midwest, to the Mountain West, and then now back down to the south again. And, you know, the patchwork response didn’t work. It’s not working now, and it won’t work. So state leaders really need to be able to coordinate their response within their own state.

MODERATOR: OK, great. Thank you. Quick follow up. You said that about one out of three cases currently are happening at home and she was wondering which study that might be?

THOMAS TSAI: Yes, I can send you an email. That comes from some of the information from contact tracing in terms of identifying the clusters of infections, so it’s not necessarily one third of all cases, but at least one third of clusters. But I’ll send you information on that.

MODERATOR: OK, great. Thank you. Looks like we don’t have any other questions. Do you have any final comments you’d like to share with us?

THOMAS TSAI: Yeah, I think I’ll close with just a few themes from this past hour is that in some ways, this conversation, as it often does over the course of the pandemic, comes full circle. And I think really the last set of comments is most important. A testing strategy nationally is incredibly important. We have that strategy now. And that’s the federal funding to support state and local efforts to support the demand for testing, support the affordability of testing and move testing into the communities where they need it. So the challenge is to implement that testing strategy. The second part of that is using that information from testing to inform the set of public health interventions we have at our disposal so that testing is necessary, but not sufficient. And again, we’re back in the phase where we need to use testing information to guide action, both on an individual level but as well as a community and policy level as well. So again, I always sort of end with a positive note, which is that despite these rising number of cases, hospitalizations and deaths, you know, we can turn the corner. We haven’t yet, but we can. And as we move past the focus on the election, this is the time to double down and focus on our ability to not just mitigate the pandemic and really suppress it. And the analogy I use is we keep treading water in sight of shore, and we keep waiting for a vaccine or a therapeutic to come. But what we really need to do is just swim to shore and we know what that looks like. And that’s the testing, support isolation, masking, and physical distancing. And we just have to do it well. So I think that’s the main message I have to leave with.

This concludes the November 5th press conference.

View more press conferences