Lori Walsh: Elizabeth Racz is an epidemiologist and public health expert at South Dakota School of Mines and Technology. She teaches in the Pre-health and Biology program on campus and she's been helping us navigate the modeling and predictions of the pandemic. She's joining us now with an update and to answer any questions you have about the progression of coronavirus in South Dakota. Welcome back Professor Racz. Thanks for being here.
Elizabeth Racz: Oh, thank you for having me.
Lori Walsh: We heard governor Kristie Noem and representatives from Avera and Sanford and Monument Health gathered and we keep hearing that they are all on the same page with the modeling that they're using. And they went through at a recent press conference the models they had consulted or looked at or used. Epidemiologist, Josh Clayton pointed out, the limitations of the ones that they thought for how that would apply to South Dakota. So there have been things that have changed in how we're looking at this situation, really daily. But certainly since the last time you were on. Is you're following those changes, what are some of the significant takeaways you have from the most recent data that the state has released?
Elizabeth Racz: Well, some of the most significant takeaways that I got from the data or the models that are being chosen and data information that's being released by the state is that we are now having large levels of cooperation at high levels in our healthcare system with the state government, and everyone is getting on the same page and learning what their shortfalls are going to be and how to fill those gaps. Which is a major step and it's going to be very helpful for protecting the health of South Dakota.
As they assess these different models, remember, no model is perfect. We know that a model is a snapshot of what's going on in the real world. It's what's happening out in the real world, but it's simplified a little bit. And when a model is made, the modeler makes certain assumptions about what's going on. And that's why we see models that are so different from place to place, depending on what the assumptions are that go into the models and depending on the type of data points they're going in.
So when we start talking about the assumptions that are going into these models, we're talking about, is everyone on the same page about what is the basic reproductive number of this infectious agent? What is the generation time of this infectious agent? What other parameters about contact are being included? Or environmental factors? So there's a whole number of different inputs into a model, and as you go through and look at the different models that are out there, they do all have different parameters or assumptions to some degree. And that's what makes them different in part. And also again, the data points that are going in.
Lori Walsh: So is there a South Dakota model? Or are they referring, from what you can tell, taking these other models and using it to make decisions? Or is there a modeler in South Dakota that is saying, "Here is our model."
Elizabeth Racz: Well what we have is a group, doctor Jason Ash and myself. He's been developing a model and bouncing it off me a little bit at the school of mines. And we presented that model to Monument Health that they can use if they like. That is one model that was considered. There's also models Covid Act Now out of Oxford and there are internal models that they are making themselves to really try to tailor that created model to South Dakota and to our needs.
You can use one model for a longterm view and another for a short term view. There might not be one model that works, especially right now, because the data that we have coming in, although we are getting more test results in, they're biased and this is because we are saving and this is for the health of the people who are getting sick.
We are saving those tests to confirm the Covid-19 disease in very ill patients. Which, as a doctor treating a patient, that's something you really need that information. As an epidemiologist we would like more mass testing and get the antibody So based on what our healthcare system representative are presenting, I do know that there was a lot of agreement among the different hospitals on many of the factors that were going into the models. For example, "the are nots." So the basic reproductive number on generation time and of course how it spreads. I do think the real beauty of what I saw coming out of the press conference the other day was how everyone really was listening to each other and working together to really tailor the models that we do have to South Dakota.
Models are not perfect. We just tried to make them as good as we can. And right now, with the data that we have, it's a little like looking at a fuzzy picture and until we get more good data, it's like you're looking at a picture and you can't quite focus the lens. So you can see that there's a peak coming but you don't know the exact day. But it's so important that we have good information, or so important that we protect people because everyone is naive to this virus. It's so important that we really stretch that out, if need be, just to protect people.
Lori Walsh: Are we going to be getting more data or are we stuck with the parameters that we have? And what kind of data points do we expect to clarify that picture and bring it into focus?
Elizabeth Racz: Okay. To bring the picture into focus: we need mass testing. We need those little antibody cassettes that give you information about whether or not someone is still infected or not, or has been infected and has antibodies. And what that would do is if a person has antibodies, that means that they have had the virus and successfully recovered and they have either high or low numbers of antibodies, but they have them. And they are protected. We don't know for how long, again, because this virus is new.
But this would also be a key element to helping us get people back to work. It's something where you could test a person and say, "Okay, I have the antibodies. I know I have sero-converted, or had the infection and recovered, and I can go back to work possibly for a certain amount of time." We need to research that a little more because with this virus, we don't know how long the immunity will last once people have recovered.
But it would also be a way to capture the people who have sero-converted asymptomatically. And allow them to go about their business.
Lori Walsh: And a lot of people have that question right now. You're hearing it all the time. "I might've had this. I got sick. I couldn't get tested." Some of those people are going into the ER or getting tested and finding out their test was negative. And they're still like, "Mm, I don't know. This might have been it." And that antibody testing would be a way to determine that. Is that coming?
Elizabeth Racz: Correct.
Lori Walsh: Is that something that is even in the realm of possibility or is that sort of in a perfect world scenario?
Elizabeth Racz: Well, we would all like to see it very soon. I know some other countries are using the antibody cassettes, the little tests to put people back to work. I don't know, again, how long that immunity is conferred for, but it's absolutely an angle we should look at. Antibody testing would help us a lot. It tends to be a relatively cheap, but again, right now resources, it's a real bottleneck for us in all areas. Whether it's PPE for our very brave healthcare workers or ventilators for our most ill or antibody testing to help clear people.
Because we really need to, I think, work with supporting our public health system by, we the people, asking for cooperation in between different States. Who has supplies? How can we move them? Really get a nice coordinated effort so we can shuffle supplies around as needed, because everybody's short.
Lori Walsh: And we're hearing lots of reports about how that's not working very well. States are competing for limited resources. That's been a big conflict throughout from the beginning of this. I have a couple of things I really want to dive down on. One of them is this idea of: we hear governor Noem say, "South Dakota is not New York City." And that's established. We know the difference and we keep hearing population density. So one of the things I'm wondering when we create models in South Dakota, is it just because the population density is so much lower that people are less likely... We're already somewhat socially isolated? And that's why you see problems in the city like Sioux Falls that are maybe more intense or is there even something in the air? Are we talking about the particle matters in the air and that population density makes a difference? Tell us why South Dakota is not New York?
Elizabeth Racz: Okay. So part of the reason why South Dakota is not New York is for exactly some of the reasons you've just listed. Our population isn't as dense. In New York, you have a lot of people living in close proximity to each other. Walking around a lot next to each other. Social isolation is much more difficult just because there's not as much space to spread out in.
Now something to keep in mind, although we are barely rural, or we are rural, some areas have more densely packed populations like Sioux Falls. And we are fairly isolated. We don't have an international airport, but we do have people coming into the state who can travel in and out still. And these are sources that can detract from the protection you might get from being distanced from someone. So for example, we see these clusters of Covid-19 disease in urban areas. And when you model that or map that you can see, "Okay, wow, there's a lot more cases there."
The issue is, is when we have travel from one location to another with this type of virus where everyone is naive, very quickly we start losing that protection. When people start coming in and potentially bringing it with them from other locations or densely populated locations where they're gatherings and the spread can happen. So we start losing and it starts taking away some of that potential protection that we might have just from being a more rural isolated place. So those are things we need to keep in mind. And that shows up usually in the disease modeling. Again, depending on the assumptions people choose when building their model.
Lori Walsh: The numbers are very broad. In the last, there was a low number and then a really high number, 30% to 70% of the population might get this. When you start breaking down that math, it's fairly astonishing. People are asking, "We can't narrow it down closer to that." Tell us how to think about that broad range of possibilities?
Elizabeth Racz: Right. So right now due to the lack of testing data, what we're left with, that we can be sure of, are percentages of population. So we know our population, we know our numbers, but we can also say, "We can run a few different percentages." And we can get these percentages off of other people's models or we can look at the data that's coming out of other countries. For example, 80% of the people who get sick are symptom free. Asymptomatic. Or mildly symptomatic and can deal with it without a hospital intervention. But that leaves 20% of the people who might need extra help. And when you start thinking about our population, 20% of that number that might need extra help, we're still talking about a lot of people.
Then we start breaking that 20% down of people who might need hospitalization and cases. Whether you're looking at 30% of the population or 80%, you're really going to still have a huge number of people. And what you need to do is, we really don't know. This is a new virus. We don't know how it's going to act in all scenarios. And it's really good when we don't have the ability to focus in to at least know the range. At least know the sides of that image. And so you can plan as data keeps coming in, we can kind of see which way we're falling in that broad range: closer to the 30% or closer to the 60%, 70% of people having major issues or be infected.
Lori Walsh: So are we always planning for the worst case scenario? Are we always just saying, "Okay, where we need to focus our efforts are on getting rid of the worst case scenario," and then address the fact that if we do that and we're successful, some people are going to look back and say, "Well that model was wrong." And that's not really the case. Right? So explain that to people of how we plan for the worst.
Elizabeth Racz: In a way all models are wrong. We tried to pick the least wrong model. Everyone needs to realize all these people are pooling their intellectual capital to do the very best job they can in a very tough and fluid modeling situation with very little unbiased data. That's why we need to look at such a broad window there of percentages. Because you don't want to under-prepare.
It goes back to that under-preparing and over-preparing, and trying to find that right response. If we're a little over-prepared, that's okay. You know we save some people. Okay. Now we are working on a balance here, trying to balance our resources versus people coming in and what they'll need. Unfortunately, regardless of the percentages you pick for people getting ill and needing care, we are going to need outside help. We are going to need to have different hospitals within the state help each other.
From municipality to municipality. I know the governor mentioned that she was reaching out, trying to get more supplies. These are the kind of things we are going to need to do to address many situations within the modeling percentages that we have. Or the estimates that we have.
Lori Walsh: Another question is really has to do with this sense, and governor Noem mentioned at her press conference. I don't think it was yesterday. My days are a little confused right now. But she pointed out Beetle County. Right now with the numbers today, Beetle County has had 21 positive cases, but 18 have recovered. And she held up the efforts in Beetle County as a success story.
Now you look at Minnehaha County, today a 165 cases and only 26 have recovered. It was just at a hundred yesterday, I believe. It just crested 100 a day or two ago. So this is a pretty big jump, Minnehaha's increasing. Compare those two communities, the city of Sioux Falls and the metropolitan area. Because you had Lincoln County in there with another 27 cases versus Beetle County, which was able to have an early, "Everybody pay attention." And now it seems to be we're not seeing a lot of new cases from Beetle.
Elizabeth Racz: Right. So if we look at these two areas and compare them, we're talking about a more rural county compared to Minnehaha. We also have to say, "Thank goodness we have people recovering. That's wonderful news. We're happy about every single one of those." It does come down to, are we doing the mitigation techniques? Are we able to treat those patients? And it also comes down to a person-by-person's health status. Do they have underlying conditions? What was the level of their exposure? All of these factors will play into whether a person recovers. What's the health care that they're getting? That that sort of thing.
Now comparing that to a more urban area, or that's where we really, really have to stress that, those mitigation efforts as well. And again, it might be a little more difficult to do it there. There just more people to come into contact with. We know that people, just by talking, can spread this infectious agent. And it can hang in the air for hours. So if you're in a place where you're using, for example, an elevator. If someone was in there coughing a half hour before you and you walk in, you could be exposed. That's going to be less likely to happen in a rural environment. So there's all these little sort of micro situations, micro niches for this infectious agent to hang out in, in both urban and rural areas. But there tend to be more people interacting with them in urban areas.
Kind of have to think about your landscape and the environment you're interacting in as this ecosystem and where would this virus hangout? Although viruses aren't considered living organisms.
They do use parasitism as a way to reproduce. They parasitize, they're intercellular. They're using intercellular parasitism to reproduce.
Lori Walsh: In the modeling work that you offered to Monument Health, talk a little bit about Rapid City and the Black Hills area. What are some of the things that... Do you create the model and offer suggestions or recommendations? How do you help people interpret that information? What do you see as a head for the black Hills area?
Elizabeth Racz: Okay, well when we're looking at these models, and the one that Dr Jason Ash was talking to me about, and the one that went to Monument Health, we're looking at the basic reproductive number over time. So what that model does is it takes the last seven days of cases and it allows a prediction about the next few days to weeks. And then of course it's as you get away from current time, the air starts getting larger.
But it has done a nice prediction for us recently. And we do see, still, this exponential growth where you have growth increasing at a certain rate based on the value from that past day, for example. And what we're seeing is, is still exponential growth, which means that we have not gotten to our turning point yet for the curve. As we look at the different curves that have been modeled and made, we see this spread and maybe the peak coming in June or in May.
And what's important to keep in mind is, especially for our area, in the Black Hills, what we're looking at is we're ramping up until we hit that time.
We're going to have more cases until we hit that time and then after that time we don't know if we will plateau or if we will just turn down right away. So we're looking for two different turning points. They're called inflection points. That first inflection point, generally in epidemiological circumstances, when you hit your first inflection point, you know you're about halfway. But again we don't have dates exactly to associate with that because we can't focus in without more of that refining data. What we're looking at is at least for the next few weeks, next few months potentially. But definitely the next several weeks, a ramping up of cases. And we are seeing that. We've gone from just under a hundred to 187, to 280, to now over 300 in just the past couple of weeks.
We're going to see that growth continue.
Lori Walsh: We will keep having you back on the program to help us parse this out in the future. It's been so useful for us, so we really appreciate your time. Thank you so much for being here with us today.
Elizabeth Racz: Good. I'm glad I could help.