Skip to main content
Loading Explore guides…
Domestic & Global Research · Acedra
Home
Explore
Medicine
Domestic & Global Research
Sign in to save
Chat
If YouTube asks you to sign in, open the video on YouTube.
Watch on YouTube
Reload player
Transcript
153 segments
Copy
Hide video
0:00
Most of my work has been on domestic— most of my work recently, I should say, has been since, the pandemic, on domestic work, domestic questions, healthcare-associated transmission and transmission of COVID and other respiratory viruses here in Utah. But a theme throughout my entire research career has been working in global spheres. This is talking about that piece. But one of the interesting things that happened to me during COVID when I
0:31
started working with the Utah Department of Health to understand how it was spreading here in Utah, was that I realized that a lot of the work I had been doing before was interesting scientifically, but it was sort of languishing in scientific journals rather than being used to, to stop the spread of diseases. And so one of the things that I've really, really focused on since the pandemic is trying to make sure that the work that I'm doing is motivated by public health policy
1:01
changes and working directly with those policymakers to help implement those changes. And so here I'm going to talk about sort of two different topics that fit into that, that's working very closely with global stakeholders.
1:16
Okay, so that's not— how do we— oh no.
1:19
Sorry guys, I lost my presenter mode. There we go.
1:25
All right, so just since you guys are not all infectious disease modelers,
1:32
just a quick overview on the global burden of infectious diseases.
1:38
Infectious diseases have been increasing in the 21st century. There's a number of reasons for why this is happening, but a few of them are increased globalization, which leads to higher connectivity of global populations and more opportunities for spread. We saw this a lot, you know, we saw this during, during COVID It didn't take very long for COVID to get from China to the US. It's a 15-hour flight. We're now so much more connected as opposed to,
2:01
you know, 19— early 1900s when it was like a multiple-week journey by boat. The other thing is we have increased population growth. So increased population growth leads to increased population density. So, that means you're contacting more people when you're in dense populations, which is more local transmission opportunities. And finally, climate change.
2:24
Climate change has increased the range of many infectious diseases, especially vector-borne diseases where previously, you know, places where mosquitoes couldn't live because it was too cold, it's now getting warm enough for them to live or live year-round, which is leading to increasing the range and severity of many pathogens. And this is despite improved access to healthcare and improved sanitation. So I stole this paper, or this figure, from Baker et al.
2:51
It's kind of hard to see with the glare, but it's just sort of an overview of some of the major— thank you.
3:01
You're gonna turn off the lights. Yeah, it's sort of an overview of the major outbreaks that have happened since 2000 and it stops at 2020. But we're really sitting on the brink of another major outbreak. And in some cases we've already sort of crossed that with H5N1 where it's spreading really rapidly in birds, in cattle. And as you know, and it's been this major wildlife pandemic, it's not far from, it doesn't take a lot to imagine that that'll soon start spreading in people or could.
3:36
At the same time, global vaccination coverage is declining. COVID-19 disrupted health systems worldwide, and that includes global vaccination programs. And global vaccination has not yet returned to pre-pandemic levels. So here you can see a plot of the diphtheria, tetanus, and pertussis 3-dose vaccination rate over the last 20 years. And you can see it was increasing really nicely until COVID and then it dropped. And we have not yet in any region, we have not yet really
4:09
returned to the pre-pandemic levels. At the same time, we've seen increased vaccine hesitancy. So that's not only did we have interruptions to people who wanted to get vaccinated, we're now having more people who don't wanna get vaccinated. And so this is leaving us really vulnerable to disease reemergence.
4:30
And the problem is all of this combined, it leads us to sort of require us to revisit how we respond to infectious diseases. The playbook that was written, you know, in the 1900s is not going to work anymore. We really need to think about how we're going to, how we're going to change and how we're going to do better.
4:52
So transmission models are one tool that can be used for this response.
4:58
And they can be really helpful in reimagining how we respond to infectious diseases. The pandemic thrust models into the spotlight. I'm sure many of you, if not all of you, have seen this figure. It's adapted from the New York Times about flattening the curve models. This is really infectious disease modeling is talking about how we flatten the curve and how big the curve is. But models have been at the forefront of outbreak response for the last 40 years since the HIV pandemic.
5:26
Models are particularly useful in providing guidance in understanding and responding to pathogens when data is limited. And this is because they can rely on mechanistic assumptions. They can also allow us to, there's the, they can also allow us to do counterfactual scenarios where we couldn't run these in real populations. We can compare if we do this intervention or we do that intervention, what is the impact and which one should we do.
5:58
And we can use those to guide clinical trials, in ways that you can't just experiment on, on real people.
6:09
And so I'm going to give two examples of how my research program has been using models to improve treatment guidelines. The first one is a completed project. So we'll be talking about cholera treatment guidelines, working with Med— working with Médecins Sans Frontières and the Global Task Force for Collar Control. And the second one is in-progress work, and I'm gonna leave you hanging a little bit. It's on diphtheria treatment guidelines.
6:32
So we started out working with the WHO, and this has turned into an R01 submission that's going in next week. So sort of leaves you hanging on that because it's not done yet.
6:44
But before we do that, I wanted to introduce you to our group. So I'm part of the Infectious Disease Dynamics Group. Here at the U in epidemiology.
6:54
It's a multi-PI group. It's myself, Katherine Walter, Kareem Cotter, Matt Seymour, and Damon Toth.
7:02
And we're sort of unified by our use of mathematical and statistical models to understand disease spread. The, the central focus of our group is understanding how infectious disease transmission impacts our ability to control and eliminate these pathogens. The majority of the people in our group are focused on domestic issues, um, but myself and Katherine Walter work in the global sphere. All right, drink break to save my throat.
7:37
All right, so as I said, I'm going to talk about two different projects. The first is this cholera treatment guidelines.
7:45
So for those of you who don't know, I know we have some cholera experts in here, so this is gonna be real, real overview for Daniel. But cholera is a bacterial disease spread through contaminated food and water. It causes severe diarrhea, dehydration, and death. And while many people or most people have mild or no symptoms, about 10% of people develop very severe symptoms that can result in death in a matter of hours. It's estimated that there's anywhere from 1.3 to 4 million cases each year
8:16
globally, resulting in sort of a broad range, but somewhere around 21,000 to 140,000 deaths.
8:28
So current cholera treatment guidelines reserve antibiotics for the most severe illness, and everyone else is treated with rehydration. And so we're, we're reserving these for the people at highest risk of death, and that's partially because patients with moderate illness experience self-limiting symptoms. They resolve with the appropriate self-supportive care,
8:51
and they only receive a modest benefit personally from taking antibiotics, primarily through the reduction in symptoms.
8:59
But antibiotics can reduce the shedding duration, so the ability to transmit by up to 90%. So you go from transmitting for about 10 days to transmitting to about 1 day.
9:10
And antibiotic resistance is a significant threat for cholera. It's particularly good at evading the antibiotics that we have. And so this is— so the worry about maintaining the integrity of the antibiotics that we have for cholera is governing the current treatment guidelines.
9:33
But our colleagues from Médecins Sans Frontières reached out to us and said, could you use modeling to determine if expanding antibiotic treatment guidelines to include moderately symptomatic individuals could actually help control cholera outbreaks? So this is stemming from the idea that antibiotics reduce the duration of shedding, so they reduce the transmission in the population by up to 90%. And so if you take this big group of people, the moderate infections
10:01
representing most cholera infections, you treat them, you make them less transmissible, they're not able to infect as many people. Could that actually stop the outbreak or limit the outbreak? Could this be an opportunity for control?
10:19
Absolutely.
10:21
Is cholera in the US at all?
10:23
No, cholera has not existed in the US in a long time, except maybe in travelers.
10:30
There are non-epidemic strains that circulated in lakes and waterfowl, but in a couple of cases, travelers that come in from the DRC or Haiti. [Inaudible]. Yeah.
10:53
Yeah, so as I said, antibiotic resistance is a big problem for cholera. And so we have to think about the trade-offs with using antibiotics. So we, when we think about, when we think about antibiotics, most of the time we think about there being, you know, sort of patient, positives and harms, benefits and harms,
11:16
but at the population level. So patient benefits include, you know, reducing your symptom, reducing your duration or severity, and, you know, not dying.
11:28
The negative impacts include impacting your gut microbiome, potentially subsequent, increasing your risk for subsequent resistant infections. But when we think of the population level, a lot of times we only think about the harms. So we think about this increased risk of selective pressure, right? So this increased risk of antibiotic resistance developing when we use antibiotics. But what this project is proposing for cholera is that there actually might be population-level benefits as well.
11:58
You know, if the antibiotics are able to reduce the
12:03
duration of shedding, reduce your transmissibility, it's going to result in fewer cases. And so could we have a scenario where we've reduced transmission so much that you actually use fewer doses of antibiotics over the course of the outbreak? And so that's what we're sort of looking at here.
12:23
So we built a model and I did not include any equations because, you know, I know my audience. I don't wanna turn you guys all off to this, but here's a nice pretty model diagram of what is happening here. So we developed the simulation model. I can walk through it very quickly. So we have susceptible, this S box, and they become exposed. Exposed is people who have successfully become infected but are not yet able to transmit. Once you become infectious, so able to transmit the infection,
12:55
you can either become asymptomatic, which is the top box, you can become moderately symptomatic, which are these green boxes, or you can become severely symptomatic, which are the blue boxes. And then the green and the blue are further separated into are you seeking care or are you not seeking care? So the U is not seeking care in both cases and the T is seeking treatment here.
13:21
So importantly, every severe infection who seeks treatment, seeking antibiotics, receives antibiotics. And then they recover.
13:32
And the question that we're looking at is changing this, these, looking at between these two arrows in the moderately symptomatic and seeking treatment. So right now, the current treatment guideline is that anybody who seeks treatment does not, who's moderately infected, does not get antibiotics and goes up to that top box. So right now this, oh, I'm on my tippy toes, this box,
13:55
for those of you on, I could probably do this, this box
13:59
does not exist in under current treatment guidelines. Nobody who's moderately infected gets antibiotics. But what we're looking at is what if we change this proportion? How does it impact the transmission dynamics? And just so that I don't overlook it, we split our recovered or removed box because we're mathematicians, we think of them as removed.
14:23
Into untreated, treated with antibiotics, and dead.
14:28
So we can keep track of how, of these different epidemiologic outcomes. And so we developed this model, we extracted parameters from the literature, and we use Latin hypercube sampling to explore the range of parameter uncertainties. So you don't need to worry about what that is, but it allows us, it's this complicated way that allows us to search over the range of all parameters when you have dozens or even hundreds of parameters
14:54
in a scientific way, scientifically rigorous way. And so we calculate the outbreak final size, so the total number of people infected at the end of this outbreak. We calculate the number of antibiotic doses used, in particular compared to not treating anybody, so additional doses. And then we look at cases averted and deaths averted.
15:21
And so what we find is that when you expand
15:27
treatment guidelines to include moderate infections, the outbreak final size, which is what's plotted here, goes down in all cases. So here we have 100% of moderate infections seek care. So everybody who is moderately infected is presenting to a health facility. It's not particularly realistic, but bear with me here.
15:47
And right here we have this sort of where it intersects zero is where none of them get treated. And so that's sort of the current status quo. And then as you move across the x-axis, we're treating a higher proportion of those. So everybody's come and we're turning at the beginning here, we're turning everybody away. No antibiotics for moderately infected, only severe. Currently what's happening, and this dotted line sort of tracks that outcome.
16:16
And then when you get all the way over to 1, that means every single moderate infection is treated with antibiotics. And so what you can see is the final size of the outbreak, so the total number of people or the total proportion of the population goes down in all cases. And we look at these 3 different outbreak scenarios. So you have a low reproductive number, so this is like some outbreaks in Africa. You have an intermediate reproductive number, which sort of happens everywhere.
16:43
And then you have, sort of happens not everywhere, but it's sort of more sporadic of where it's happening. And then you have places with high reproductive numbers, so really high transmission.
16:53
And that's sort of like Southeast Asia when you have outbreaks there. But either way, in all three of these different outbreak scenarios, you see that treating moderate infections reduces the overall outbreak size.
17:05
And that makes sense, right? You, when you start treating people, they're not able to spread and the outbreak doesn't continue. And so what happens when we think about more realistic expectations in terms of who's seeking care? So right now, moderate infections don't get antibiotics. And so we expect probably a lot of them are like, oh, I have a moderate infection. I'm not gonna go to the doc. I'm not gonna go seek care because then they're not gonna give me anything anyways.
17:30
A lot of people in the US do this with flu, right? You have influenza, you have respiratory virus, you're like, well, I'm not going to go to the doctor because what's the point? What are they going to do? They're going to send me home and tell me to take DayQuil, right? Sort of same idea. And so in that case, we can look at what happens if only 5% of moderates seek care. We can increase it to 50%, and then I already showed you 100%.
17:49
What we see is that even when only a small proportion of moderates seek care, we still see a benefit from treating them, although the benefit's reduced, right? You're not treating as many people.
18:01
So what we've sort of said is that if we're gonna do this, you need to increase awareness, you need to increase, you need to put out an informational campaign that alongside this, that if you're going to expand treatment, you need to let people know that you've expanded treatment.
18:17
But what about antibiotic resistance? So we can't, we're not modeling antibiotic resistance explicitly here, But what we can say is the number of antibiotic doses is correlated to the risk of developing resistance. And so we have, so we can look at that. And so what we see is that this is the, the Y-axis is the proportion of infections averted, so cases averted, and the Y-axis is the number of additional doses of antibiotics used. And so as you go, as you go this way higher, you,
18:56
you're using more doses and so more risk of antibiotic resistance. But what we see, we've split it up colored nice and pretty,
19:06
but what you really need to focus on is these purple dots which are negative.
19:10
So in these two scenarios, so you have 100% of moderates getting
19:16
antibiotics for a, for a low reproductive number and 100% for, um, intermediate reproductive number, these are actually using fewer doses over the course of the outbreak. So not only have you now brought down the total outbreak size, but you've brought it down by so much that by expanding treatment, you're using fewer doses because you have so many fewer cases in the outbreak that you're actually able that even when you treat everybody,
19:44
you're actually able to use fewer antibiotics because the transmission reduction has been so great. I'm getting a lot of head nods, which is great. I love that.
19:54
So yeah, in certain circumstances, it's not everyone, right? Like, these are a bunch— each dot represents a different simulation with a different parameter value. So there are a lot that are happening in these sort of intermediate ranges, and we don't need to go into what what the different colors mean, but there are a number of simulations, quite a few, where you see they use fewer doses over the course of the outbreak. Well, what we noticed was that it looked like there were these
20:22
thresholds that were happening where certain parameter values were leading to
20:28
these benefits and certain parameters were not. And so I got a master's student in math to come and take a stab at this
20:37
and looking at, looking at like analytically solving the system of equations for these thresholds. Again, I promise there's no equations. You just have to bear with me that that's what we did. And what we found is we found that there's basically two thresholds. We named them the dose utilization threshold and the outbreak prevention threshold. And the dose utilization threshold is where expanded antibiotic treatment reduces the number of antibiotic doses over the course of the outbreak below
21:04
that with current treatment guidelines. So those were those purple dots. Basically, they were so good that, you know, it's so good that you're actually using fewer antibiotic doses over the course of the outbreak. So in this case, that is a time when you definitely wanna be expanding your treatment guidelines because you're going to use fewer doses. So you're protecting the antibiotic stores that you already have compared to your current guidelines.
21:28
The outbreak prevention threshold, the other one, is where expanded antibiotic treatment reduces the effective reproductive number below 1 and actually halts the outbreak. So this means that we're so good, I didn't show any simulations of this one because it's a little harder to show. But this one, you're so good, you're actually preventing the outbreak from ever taking off. So this only happens in low transmission areas, but you can actually stop an outbreak before it can start.
21:57
And so again, I promise no equations, but we calculate these. So the outbreak prevention threshold is when treating moderate infections prevents the outbreak from taking off. We solve the system of equations. We found that using sort of the average of all of the
22:13
cholera parameters that we found, that this occurs when the reproductive number is 1.42.
22:19
And so on this plot, this shows
22:25
the percent of the population that's infected and this is the proportion of moderates treated. And so anywhere that it crosses the x-axis is where
22:36
treating moderate infections drops you below this outbreak prevention threshold, stopping the outbreak. And so you can see it sort of crosses right about here where it's 1.42, R naught is 1.42. And so in order to achieve that, when R naught is 1.42, you have to treat every single moderate infection. But, so that's probably not realistic, but as you get like lower and lower reproductive numbers, you need to treat fewer and fewer proportions of the population.
23:06
So for example, if your reproductive number is about 1.3, you just need to treat about 55% of the moderates. So maybe a more realistic value,
23:19
but it gives you an analytic solution so you can plug in your parameter values for your outbreak and you can say, do I want to treat or do I not want to treat moderates? And then how many do I need to get?
23:39
The other threshold we talked about was the dose utilization threshold. So this was the purple dots. And so we can see when treating moderate and severe cases reduces the number of doses used compared to treating severe cases alone. Again, solve the system of equation, and this happens a little bit higher. This happens when R naught is equal to 1.53. So here's our reproductive numbers again. And so 1.53 is here. Oh, this is an old plot.
24:07
It should be in this plot.
24:10
This is definitely an old figure. So in this one, it's crossing at 1.43. Oh, no, sorry. It's crossing at the— again, I'm sick. My brain's not 100%. It's crossing at this dot where we've added. So you can see that when you start, we'll follow this trajectory, this light blue trajectory. So when you start treating moderate, so proportion of moderates treated, and this is the percent of the population treated. You first increase, so you're treating more compared to this
24:39
when you treat this where it crosses the y-axis, where x is 0.
24:47
First you treat more, you go up, and then you start to come back down, so you're treating fewer and fewer. And then this point indicates where you cross the— what you would have done. And so anything below this proportion treated So about 1.55 for
25:06
R naught of 1.4. Anything below that, you end up, you're using fewer doses. So this was the purple. I think it's a little bit more intuitive to look at it on the purple. But yeah, so we've, we were able to solve this analytically.
25:23
So in conclusion for this project, for very low values of R naught, treating moderate cholera infections can actually stop the outbreak. For intermediate values of R naught, treating moderate infections can reduce the number of doses used overall. And so these thresholds can help, can be achieved. We've shown that these thresholds can be achieved under realistic parameter values for cholera.
25:45
But in doing all this, we realized that many of these key parameters are not well characterized and they have huge ranges of uncertainty. And so before, we can actually say, yes, you should do this. We really need better estimates before implementing this. And again, we've highlighted that the proportion of moderate infections who seek care is this key quantity and we don't really understand much about it. We don't know how many people are currently showing up and getting turned away.
26:15
And we also think that probably current treatment guidelines are likely driving this, right? Like if you're, if you know that you're not gonna get treatment, you might not present to care. And so it's necessary to increase care-seeking behavior to achieve these benefits.
26:31
But you know, previously it's sort of said that there's this unavoidable trade-off between antibiotic use and the risk of resistance. And here we're showing some scenarios where antibiotic use actually creates this population-level benefit. Yeah.
26:46
And so we've sort of, again, we've sort of tailored all this to cholera. We've been working with the Global Task Force for Cholera Control and MSF. But I think this is a theoretical framework that may be useful for other similar pathogens. And it really challenges the sort of traditional approaches to antibiotic prescribing and stewardship and proposes this sort of strategic context-specific framework for thinking about how we might use antibiotics to stop transmission.
27:13
All right.
27:17
This is a big part of your story where we are seeing the differences in the findings of using the antibiotics. With this modeling, did you consider the age of the— like, is this will be with the kids under 5 or more with the cholera and high risk? Is it like some parameter which was— will it be a different result if you consider the age as 0 to 5 versus the adults? So we didn't consider the age. We didn't consider age in this. The model parameters were already sort of exploding.
27:46
There's a lot of parameters and very little consensus about what the parameter, the true parameter values actually are. And so breaking it down even further or adding more complexity,
27:59
we didn't consider. I think one of the issues that we're struggling with is getting good estimates for what these are. And I think we don't have, we don't have good estimates for what the difference between children and adults are when you add in that complexity. So for right now, we just say some proportion of the population is moderate, some proportion is severe. Children are probably more likely to end up in the severe box, as you're saying, but we don't actually look at it by age.
28:27
Children also might be more likely to seek care, right? Parents might be more likely to take their children in to seek care than than if they were an adult. So that might be a, yeah, that might be a trade-off. But yeah, I have a question on Zoom.
28:42
Does your model assume 100% adherence to medication regimens? It does, yeah. Which is an, which again is a not so realistic, but we had to start somewhere. Yeah, that makes sense. I mean, I think that that obviously would have implications both for I think resistance emergence as well as the effectiveness of curbing an outbreak is what are the adherence patterns in patients like this as well. Yeah absolutely.
29:13
Yeah, I will say this is sort of a first pass. This is not meant to be, you know, we've been talking with the, with MSF and GFTCC about using this, but they're certainly not implementing it as this first as our first model. This is more of a guideline for thinking about what parameters do we need to better estimate? What do we better, you know, what can we, what can we find that's not in the literature? What, what sort of clinical trials might be beneficial to collect this information
29:41
to be able to move forward with it?
29:47
All right, so moving on to diphtheria. So as I said, this is sort of newer work that is not completed. I'm gonna leave you a little bit on a cliff. At the end. So in case you have not thought about this since 1888, diphtheria is a disease caused primarily by toxin-producing Corynebacteria diphtheriae. Transmission occurs person to person. It spreads through droplets, and humans are the only reservoir. It can either be respiratory or cutaneous.
30:20
Respiratory is a significantly higher risk of mortality. It forms basically a plaque in the back of your throat that releases and strangles you from the inside. Diphtheria used to be known as the strangling angel. And the other thing is cutaneous is not reportable. So cutaneous diphtheria happens probably a lot more frequently, particularly in intravenous drug users, but it's not a reportable disease. So this was once a major cause of childhood morbidity.
30:51
And mortality, and it's all but disappeared, primarily through the introduction of highly effective therapeutics. So those, and those therapeutics include the diphtheria antitoxin, which was introduced in 1888, the diphtheria toxoid vaccine, which now is combined with tetanus and pertussis, DTP, and it follows a 3-dose schedule for children and adult boosters every 10 years.
31:25
As well as, so the, the vaccine simulates the, the protection of antitoxin and does not prevent colonization, reduces your risk of getting infected by a significant amount, but you can still get diphtheria even if you've been fully vaccinated. It just protects against symptoms, so you'll be asymptomatic.
31:43
But the other therapeutics include penicillin.
31:47
So for treatment, antibiotics basically. And you can see there was a major— this is diphtheria cases in the US, and there was a major drop-off after the scale-up of the DTP3 vaccine in the 1970s. So it's incredibly effective at preventing transmission or preventing symptomatic disease.
32:11
So yeah, so diphtheria is current treatment is currently relying on a two-pronged approach. So you need antibiotics to clear the infection. So this is sort of the clearance rates. So red is treated, you clear it much faster than blue, which is untreated. And you also need antitoxin to prevent death. So it reduces the case fatality rate from about 30% to under 10%.
32:38
Um, but it's highly important to get antitoxin rapidly. The longer the delay, the— basically, the less effective that antitoxin becomes.
32:49
So you might be thinking, why am I talking about a disease that's basically not around anymore? Well, diphtheria has been resurging, over the last 8 years. There have been a number of outbreaks,
33:03
The largest, initially the largest, was among the Rohingya refugee population. These were forcibly displaced Rohingya from Myanmar into Bangladesh. They had a major outbreak of over 8,000 cases in 2017, 2018. But recently, and this is not updated, this is an old slide that I did not check. Recently, there's been an outbreak in Nigeria. It's still ongoing. There are over 43,000 cases right now. But there have been a number of cases.
33:35
Many of these have been— these outbreaks have been emerging following infrastructure failures or civil unrest.
33:42
But we're starting to see more and more diphtheria cases. And that's partially because on the first slide that I showed you, the DTP3 coverage since 2020 has fallen.
33:55
But we have a new problem.
33:57
We've run out of diphtheria antitoxin. So diphtheria antitoxin is an equine antitoxin, so it requires the use of horses. You infect them with diphtheria, you somehow extract the serum. That's not the kind of research that I do, so I don't totally understand how any of that works.
34:12
But it takes a very long time to produce antitoxin. And so we're scrambling to respond to these new large outbreaks. There's low immunization coverage. But the real problem is that demand for antitoxin currently exceeds global supply. We do not have enough antitoxin. And not only that, but we don't have enough antitoxin in the parts of the world that we're having these diphtheria outbreaks. So here are, um, 4 different outbreaks that are happening. The first one is the Rohingya refugee.
34:41
So 89% of them, um, received antibiotics. Um, in Kano— in these, these 3 are in Nigeria— uh, 4% of them receive— sorry, not antibiotics, they received antitoxin. Um, but only 4% or 0% in these Nigerian outbreaks. We're really struggling to get antitoxin to places that need them.
35:04
And so last year, I worked with the WHO to reimagine, to revive the Diphtheria treatment guidelines, the clinical management for diphtheria, and they're still recommending that every case gets diphtheria antitoxin. But the real problem is, and these questions emerged that couldn't be answered during this process of making these treatment guidelines, is what do we do when diphtheria antitoxin is not available?
35:31
How do we treat people then? And then the other piece of it is, since the antitoxin effectiveness is so time sensitive, can we anticipate where these outbreaks are gonna be happening so that we can ensure that stockpiles in those regions are robust enough to handle these outbreaks?
35:55
And so this comes into our R01 proposal that's going in resubmission next week.
36:01
So we are proposing to address these questions using modeling. So the first one, we are proposing to develop a simulation model to explore alternative outbreak response strategies. So this includes mass drug administration of antibiotics. So when you see an outbreak starting to happen, if we just treat the entire population with antibiotics, are we able to stop diphtheria before it spreads? There's also the option for treating household contacts following an index case.
36:29
So if someone in your household becomes infected, infected with diphtheria, we just give antibiotics to everybody else in the household to hopefully stop them from getting infected. And then we're also looking into reactive vaccination campaigns. What happens if we do these? Are we able to slow this outbreak and prevent some morbidity and mortality?
36:49
So the other question is, where can we anticipate where these outbreaks are gonna happen? So the way that we're going to sort of look at that is proposing to use serology to quantify diphtheria immunity in the population. So we know what the vaccination rate is. We've, we have vaccination coverage data and we're seeing outbreaks in places with supposedly high vaccination coverage, not high enough to, to prevent all transmission, but higher than you would expect.
37:15
And so we're going to look at this to identify gaps in coverage
37:20
that may have occurred because of sort of heterogeneity in vaccination coverage. So you might have populations that have low vaccination coverage within a region that has relatively high vaccination coverage. Because there's maybe less access or there's more vaccine hesitancy in those populations. And so we're sort of, we're gonna use this information to hopefully identify these gaps and be able to give some information about where in the world you might have diphtheria outbreaks.
37:50
So as I said, I was gonna leave you on a cliff there.
37:54
And so with that, I'd like to thank all of the collaborators and people who helped with this work. The funders, and I will take questions.
38:07
So you had mentioned that you were working on the Diptheria guidelines for WHO, at least briefly? Yeah. What's kind of your opinion then right now with the withdrawal of the US from the WHO, the impact on global outbreaks, especially in Nigeria? I mean, I think that the WHO is doing fantastic work, and public health in the US and abroad is sort of under attack right now. And so I think it's, it's bad. I mean, I don't know what to say. Like, it's terrible? Like, I don't know.
38:45
It's, you know, the US is a major funder for the WHO, and so it's going to put a lot of strain on public health globally.
38:54
But so is the NIH grant sort of pause and CDC cutting. And, you know, there's a lot of pieces to it that are all going to,
39:06
yeah, it's not, it's not great. So yeah. Cholera stuff. Yeah. So, like the public health campaigns or policy changes that happen, like in response to the model's outcomes, how do,
39:24
like, how do they plan on balancing like pushing antibiotics and like that message to moderate infections while also balancing like the lack of like labs to have diagnostics to figure out that it is a bacterial infection versus like a viral, like kind of pushing people towards care that's maybe going to end up getting antibiotics for not necessary things and capturing those moderate infections? Yeah, I think that's a really great question and I honestly have no idea.
39:53
I don't know the answer to it. My sort of, Daniel might actually know, but my sort of piece in this puzzle is to identify what scenarios where this may be useful. The next step in the cholera work, which I didn't share, is that we're gonna work to sort of refine parameters and see if we can estimate specific parameters for specific outbreaks, specific locations.
40:16
And say, you know, in these locations it may be beneficial, in these specific locations it may not be beneficial.
40:23
But from there we sort of pass it off. And having worked on the diphtheria treatment guidelines for WHO, there are a ton of experts from all different pieces, and they're very strict about only letting you speak to your specific expertise. So if you try to make a comment that's sort of outside of your scope, they actually get very mad at me. Yell at you to stop talking. My— one of the people who I collaborated on the diphtheria stuff with, Sean Truelove, made some— like, got into,
40:51
like, an argument, and they, like, muted him because they were basically like, you're not— this is outside of your scope and you're not allowed to talk about it. You're only on the treatment guidelines as an expert in this particular thing. Even if you have expertise in other things, they bring you on for very specific things. So I will say that once you get to that stage, they're very, like, strict about that. But I don't know, maybe Daniel knows. I mean, this is for outbreaks, right?
41:16
Yes, this is not for endemic settings. This is not for, I think in Bangladesh where there's, you know, there's some seasonality to cholera, but really aren't outbreaks of cholera. So this is for your outbreak in Haiti or your outbreak in the DRC or places where there's not always some cholera circulating. And once there's a cholera outbreak, it's recognized as an outbreak. And almost anybody who comes in with what they call AWD, so acute watery diarrhea, large volume of watery diarrhea,
41:46
especially with moderate and severe, it can be quite, quite noticeable that it's cholera.
41:53
That, and it's a one dose, I mean, you need a one dose antibiotic to treat it.
42:00
And so there's probably not going to be too many non-cholera cases in a cholera outbreak setting. If it happens during rotavirus season, I can see maybe some
42:16
nonspecific treatment of rotavirus cases. But in epidemic settings, you're likely to catch mostly cholera cases. But yeah, in an endemic zone, I don't think the model ever— No, the assumptions that we've made have all been around
42:37
everybody in the population is susceptible, there's no prior immunity. So this is, this is very much an outbreak setting. Yeah. And I think a lot of that, a lot of the results would change if you had people who were somewhat immune, partially immune because they've gotten cholera recently. So yeah, I think we're, this is very specifically not an endemic setting. This is, yeah. A one-dose treatment regimen obviates the question about adherence too, if it's just a one dose, which, that's helpful.
43:08
Yeah, I think the azithromycin is one dose.
43:10
Daniel says it is one dose. It's azithromycin. Yeah yeah. Confirm the other ones that Doc's saying.
43:19
Yeah. Lindsay, thank you very much. So for the modeling on cholera. Yeah.
43:28
How much work is needed to get to your model
43:34
then to go to some other infectious disease, is it easy to model something else or are you starting from scratch again and recreating the wheel?
43:47
So that's a great question. So I think the hardest part in all of this is finding parameters.
43:54
Going through the literature and trying to figure out, okay, what's the, What's the incubation period? What's the latent period? How long are people infected for? What's the reproductive number? And it requires, you know, a good amount of work to just read literature and try to figure out what sort of the consensus is, or if there is no consensus, to figure out what the possible range of parameter values are.
44:18
I think that writing a model is very easy. Writing the differential equations is pretty straightforward.
44:28
But the sort of conceptualization of thinking about it as treating more people leads to fewer doses. I think that's sort of the novelty that hopefully will inspire people. And then, you know, it's a normal— I would say like a normal amount of work for a paper to redo this for another disease. So the conceptualization, the sort of conceptual framework is like the big lift. We've done that. We've said, here's some other— here's some places where this may be useful.
45:00
And then, yeah, it's probably about as much work as doing another paper. So yeah, a PhD student could do it. It'd be a nice chapter for a dissertation if anybody's looking for one, wants to add some modeling into their repertoire. It's not that hard. You could do it. Does that answer your question? Yeah. Okay. Do you want to start modeling?
45:25
We got some beautiful— Yeah. I can't even read the alphabet. It's all Greek to me. Yeah.
45:36
Any other questions from Zoom or? Awesome. Well, thank you guys so much for having me.
45:51
It's great that you got through all this without a math equation.
Here to help you learn
Sign in to keep a private conversation with your course tutor.
Ask me anything about the material…
Sign in to chat