Dr. Natalie Dean (@nataliexdean) is an assistant professor in the Department of Biostatistics and Bioinformatics at Emory’s Rollins School of Public Health. She received her PhD in Biostatistics from Harvard University, and previously worked as a consultant for the WHO’s HIV Department and as faculty at the University of Florida. Her primary research area is infectious disease epidemiology and study design, with a focus on developing innovative trial and observational study designs for evaluating vaccines during public health emergencies. She has previously worked on Ebola, Zika, dengue, chikungunya, and now COVID-19. She received the 2020 Provost Excellent Award for Assistant Professors at University of Florida. In addition to research, she has been active in public engagement during the COVID-19 pandemic. She is verified on Twitter with over 100k followers and has authored pieces in outlets such as the Washington Post, New York Times, and Stat News.
Episode Description
Twitter can be cacophonous at times on a given day, serious analysis of the situation in Afghanistan, news stories about climate change, and Parry Gripp’s Music for Cat Piano Volume 1 can all compete for a user’s attention. This has only become more clear during the COVID 19 pandemic as it seems almost everyone is tweeting about the disease, with varying levels of expertise
+Full Transcript
Pennington
Twitter can be cacophonous at times. On any given day, serious analysis of the situation in Afghanistan, news stories about climate change, and parry grips music for cap piano Volume One can all compete for a user's attention. This has only become more clear during the covid 19 pandemic because it seems almost everyone is tweeting about the disease. with varying levels of expertise. However, there have been some experts who've been able to tweak through the noise and we'll talk with one of them on this royal statistical society edition of Stats and Stories recorded for the RSS annual meeting. I'm Rosemary Pennington. Stats and Stories is a production of Miami University's Department of Statistics and media journalism and film, as well as the American Statistical Association. Joining me as panelists are John Bailer, Chair of Miami statistics department and Brian Tarran, editor of significance magazine. Our guest today is Dr. Natalie Dean. Dean is an assistant professor in the Department of Biostatistics and bioinformatics at Emory University's Rollins School of Public Health. Her primary research area is infectious disease epidemiology and study design, with a focus on developing innovative trial and observational study designs for evaluating vaccines during public health emergencies. She's previously worked on such diseases as Ebola and Zika. And now COVID-19. In addition to research Dean's been active in public engagement during the covid 19 pandemic, often offering long Twitter threads to help educate people about the disease. She's also authored pieces and outlets such as the Washington Post, New York Times and stat news. Natalie, thank you so much for joining us today.
Dean
Thanks. I'm really happy to be here.
Bailer
Oh, Natalie, it's a delight to have you join us. I I want to take you back in time to find out what got you interested in studying infectious diseases and vaccines? I mean, I'm really tempted. I'm gonna yield to the temptation. When were you first bitten by the bug? Oh, no, I can't help it. Come on.
Dean
Yeah, actually High School. And so I was fortunate to take a class about infectious diseases in high school, and we read the hot zone about Ebola. And then we also learned about HIV. And I just remember thinking about just the magnitude of those types of public health problems and realizing that's what I wanted to do. You know, I wanted to help in whatever way I could. And when I went into undergrad, I was a biology major. And I actually thought I would be a microbiologist, and I realized I really hated the lab. But I found I'm not good at those things. But I was introduced to Biostatistics and started working on epidemiological study. And then you know, the rest is history, I really realized that's what I like. So when I went into grad school, I focused on HIV. And then when I came to University of Florida, that was when the Ebola epidemic in West Africa had started and I got involved in projects there. And yeah, we've been working on emerging pathogens ever since.
Bailer
So when you say you're been working on emerging pathogens, when can you just sort of sketch out? What does that mean to work on emerging pathogens? And what are some of the dimensions of the problems that you deal with?
Dean
Yeah, absolutely. So it really, for me, started with working on Ebola. And so when I arrived at University of Florida, I got involved working on a phase three Ebola vaccine trial in Guinea. And it used an innovative trial design. So use this ring vaccination approach, which is this sort of flexible trial design, where it's a cluster randomized design, so you have these individuals who are these new cases of Ebola, and then the cluster is the people, the contacts and the context of contacts around that individual. And then those units were then randomized, and that was the way that we were able to evaluate the efficacy of that vaccine, which is now licensed. And so after that, who recognized that there was a lot more we could do in advance of epidemics to be prepared to evaluate new vaccines and, and diagnostics and therapeutics. And so they formed this research and development blueprint initiative. And so and so it was a group of experts that I was part of and are still so I'm a part of and you know, thinking about all these prioritize pathogens. So lassa fever, Crimean Congo hemorrhagic fever, Marburg is a list of pathogens and kind of how would we evaluate new vaccines, what are the you know, best practices? How can we tailor clinical trial design strategies to work in these very unpredictable resource limited settings and so on. So we've been thinking very generally about these types of infectious diseases, because some of the problems that come up are similar, like, it can be very unpredictable, where, you know, epidemics are going to take off or outbreaks are going to take off. COVID is a little different, because it's really everywhere. But for some of the other things like Ebola Nipa, these these, these sort of these outbreaks that burn really hot in particular locations. There's a spatial and temporal dimension, we've been very focused on. So we worked on Zika. And thinking about loss of at least Respiratory Syndrome, all the stuff and then and then COVID hit and that's really occupied everyone's attention for the last year and a half.
Tarran
And lastly, when did COVID come onto your radar? And at what point in time was it that you started to feel that, you know, this could be a problem? This could be a pandemic?
Dean
Yeah, well, so I'm sort of part of two communities, right. I'm part of this biostatistics community. And then I'm also part of an infectious disease community research community. And so I follow a lot of folks on Twitter, and the infectious disease community community. So you know, there was a really early like, pro med mail warning that came out. And I remember seeing that, and I remember everyone starting to, you know, little talking about that, and so early January, and the moment that I knew it was serious was, well, there's kind of two moments. So one is when China shut things down, and I thought, Wow, well, if they're doing that where they're seeing something on the ground, that is really serious. And then the second moment, I mean, the really big moment, I think, was when the epidemic started to take off in Italy. And so you know, as with those who were there there were discussions, really early on, I remember, our blueprint Working Group kind of came together. I don't know if it was late January, early February, but sometime we started talking about, you know, vaccines, study design, things like that. And so, so yeah, but for me, Italy was really that moment where I thought, Oh, well, it's coming here, you know, and we should plan accordingly.
Tarran
Obviously, one of the reasons we wanted to have you on the show was to talk about your experience in communicating science and statistics during the pandemic. So, you know, obviously, you've got this, this pandemic happening, there's a lot of pressure on your time, I guess, to be doing your work on the on the pandemic, what when was it that you realized that you had something to contribute to the public debate and that you wanted to get involved actively to help people understand what we were going through?
Dean
Yeah, well, actually, I was on maternity leave when the pandemic started. So I had my second child in December. And so I, you know, I was coming back in March. And so in February, when things were really taking off, I was taking care of a small baby. And so I wasn't, you know, I was involved in some research, like I remember, you know, working on some reviewing paper drafts and reviewing things a little bit during that time, but I didn't have a ton of time most of the time, but I realized that Twitter was a way to stay involved. And because I had this set of expertise and experience, I mean, I had a sort of a frame of reference for certain things like, I remember the discussions about, you know, evaluating new treatments and thinking about, oh, we have this frame of reference for MERS, we have his frame of reference for SARS, like we have this experience with, you know, other emerging pathogens, and I can help interpret the results and help people understand and it really appealed to my, I liked I like teaching and I like infectious diseases and infectious disease dynamics. And it was a way to so it started off very much about teachings like little bits of explainers. But then I think it did, it did evolve and peer review, public peer review became a really big thing, because we had this rise in preprints. And everyone's trying to digest information very quickly. And I think people also didn't, you know, it's hard to understand just how poor quality early data are, that can't be that come out of China, you know, when case definitions are evolving, and there's a huge amount of missing data, because there were, you know, barely any tests. And so I also felt another role was to kind of explain the limitations of studies and try and help people think scientifically about how you know, the quality and the biases and how to piece this information together. And Twitter has been a great forum for that. And there were a lot of people there who were very active. Have
Pennington
You noticed particular threads you've done that picked up steam more quickly than others and have an idea about, you know, what might people be finding appealing in that?
Dean
Yeah, so it's, you know, a mix of things. Certainly, I really like the educational content. I like explaining, you know, the concepts and how to kind of think about how we define vaccine efficacy. How do you know where the limitations of study are? Uh, what are, you know, these sort of different fundamental concepts and to break those down? What goes viral? what people really, like, there's anything that was a little controversial, you know, and I think that's true. And so, you know, when you do this sort of public peer review and you're critical of something, I don't know, Twitter, Twitter loves a bit of an argument. So, so, which is, you know, a little unfortunate that that's what sells it, you know, that is what gets rewarded a bit. And that's something that has turned me off at times. You know, but, but yeah, anything that's a bit critical or controversial can also get more attention. Yeah.
Pennington
One thing I really appreciated about your tweets have been the way you use visualizations at times, I think there was one involving Legos at one point, and I just felt like, That's such a space where, you know, if you can visually be fluid and, and communicate visually, like that's a way of getting a message out. And I just really have appreciated seeing that in the tweets. I have to say the work you do there, but I was like, the tweets you publish?
Dean
I know, I know, what is it? What is it even? Exactly, yeah, I'm, I am a very visual learner, extremely visual, or I, like, I am also a visual artist, too. I mean, I really like to paint and so it sort of intersects a few different things. But I'm a very visual learner. And I remember there were some results that have come out from asters and AstraZeneca trial, one of their Oxford trials. And people were trying to understand this result that was a bit confusing. And I ended up making this highlighter sketch and I have the highlighters here, actually just found those clean going through my files, and I found the paper. And if people saw just how quickly I did some of these things it was truly like 10 minutes of just making a little sketch. And that went really viral, because it really helped people. I think it is, you know, I'm such a, yeah, I'm such a visual person. And it's hard for me sometimes to hear certain concepts, when I see them, it becomes very clear. And I think that you know, people learn different ways so that, sort of, hit a nerve with people in a good way. And then someone suggested turning it into duplos. As like, that is the best idea I've ever heard. I have a three year old. And so we have plenty of duplos around the house. And I ended up kind of stealing them away just long enough for him not to notice they took them outside, I took some pictures. And people really liked that. And so and so that's been fun, too, for, you know, to find different things that resonate with people.
Pennington
You're listening to Stats and Stories at the Royal Statistical Society's annual meeting. Our guest today is Emory University's Natalie Dean.
Bailer
You know, I love this sort of dual mission that I've heard in your response, this idea of exposition of concepts, trying to help people understand that, but also, you know, kind of shaping and framing and contextualizing study results. And I, you know, and one of the things that you posted, you talked about, you know, this idea of serving as a guard rail for interpreting studies. And I thought that was a really wonderful image. And I was wondering if you could talk a little bit more about what kind of what or what's part of being a guard wear guard rail here?
Dean
Yeah, well, I remember, last spring, I did a podcast interview on slate. And the host was the only way she brings this up, you know, regularly on Twitter that I said something that really resonated with her, which is that science is a process, you know, the public doesn't usually see that process playing out in time, and how that is an evolving process. And our understanding is evolving. And, you know, instead I think, sometimes people, reporters, the public tend to think that kind of the most recent study is the most reliable one, when really what we want to communicate is okay, the most recent one still has some set of limitations, it still needs to be balanced against all of the other studies that have come out, and how do we think scientifically about integrating those together? So if something conflicts with laboratory studies, or if it conflicts with prior evidence from another place? How do we then sort of put a little star on that until it can be confirmed, you know, until those results make more sense? So, you know, trying to help people they're trying to really help people understand the limitations. I mentioned, data quality, now that there are a lot of vaccine studies helping people understand just you know, the potential for biases and confounding and, and so you have, you know, you have a point estimate and a confidence interval, but I add a little bit more on top in my mind, because there's just, there's this just a wider, you know, that's what you've measured, but there's just this unmeasured part that and so helping people understand the limitations there. And yeah, people get very fixated on point estimates. And so making sure people understand kind of the range of uncertainty
Tarran
Natalie, what was it like to, I guess evolve your contribution to stats communication from you know, from Twitter to? I think it was on your website said April 11 2020. Was your first newspaper article? And it was in the Washington Post. I mean, that's a huge place to publish your first newspaper article. I think my first newspaper article was in the Barking and Dagenham post, a much smaller scale, but what was that? What was that experience like, you know, whether it was trying to find a way to communicate the ideas in a different media or working with editors and not having the full control like you would over your Twitter output?
Dean
Yeah, it was exciting. So I was contacted by an editor and they sort of pitched a story but it was something that was born out of something I had been tweeting about. So the first thing I wrote in The Washington Post was about the need for randomized clinical trials to evaluate new therapeutics that was back in the hydroxychloroquine debate days like Do we have time to wait, just give it all, we got nothing to lose. And it was this slow down we needed to see what really works and so explaining that sort of scientific process to people and that was a lot of fun. I mean, it's kind of an intensive process, it's a different process, the pace is very fast. The turnaround times are really fast. But you know, there's nothing like a great editor and it's really fun to hear from people who have a lot of experience, you know, writing these things for broad distribution and can show you exactly what you're, you know, what you've done wrong and, and really show you show you the way and I remember one of the big things I learned was make your point very quickly upfront, okay? Because you really have a limited time to kind of capture people's attention. And so don't you know, bury the lede, right? Like you need, your point needs to be made very quickly. And so yeah, that's been fine. I've worked with a number of different people. So I've also done some stuff in the New York Times and so learned about like fact checking and all the you know, it's very intensive, it's actually funny one time they, I got fact checked on something, it was I used the word field trial to refer to a type of vaccine trial. And that's in the field. And they came back to me and said, well, oh, the who says, bla, bla, bla, bla, bla. And I said, Oh, you actually just sent me something I wrote.
So that at least made me feel more confident that I was not in the totally wrong place, you know, but it can be challenging, it's a lot of exposure, it's a lot of pressure to kind of get the message right. And that can come with its own stress. And I remember getting a lot of, you know, mostly good, but sometimes you do get, you know, negative feedback. And that's just from everyone on the internet now feels that they should be emailing you directly about everything. And so, you know, I think I also need a little bit of fortitude, as well to deal with that.
Bailer
Okay, if I can follow up, I mean, you've really been very successful in your communication efforts. And I'm, I would be interested to hear what kind of lessons you've learned, and what kind of communication tips you might have for us. And I'm wondering if it is part of that if you could think of it? Are there distinctions that you make for communicating in a public way? Or maybe through the media? I mean, are there sort of any other distinctions? So just general tips?
Dean
Yeah, that's a great question. Twitter is different from talking to reporters is different from going on TV, or doing podcasts or radio, or writing op-eds. I mean, they all have a little bit of difference that, you know, it kind of depends on your level of the effort you want to put in and talking to reporters is something I like because it's not a huge investment of time. And it's something that I can sort of contribute to distilling things down. And I would say, one of the key challenges with status, Well, okay, statisticians, we like to think about the study methods, we like to think about, you know, all the things that kind of go into the science of figuring things out, whereas reporters really want to know, like, what does it mean in the end? And, and so you have to have to be ready to answer that question. And you also have to be keeping on top of what's going on, which is challenging because they also want to know how it fits into everything else. And that is tricky, because it's just, it's like a firehose of information. So it's like, you know, a whole other activity just to try and stay and stay on top of everything. You know, Twitter, I think, allows for a lot of different opportunities to communicate with people. I mean, one thing I like to do is kind of read through an article and pick out what I find most interesting, just recognizing that a lot of people aren't going to Read the article. And so they kind of want, you know, it's sort of like you're filtering information through you. And you picking out what you think is most interesting is notable just because you have this set of, you know, expertise and so kind of filtering things through and then I don't know, I just, I try not to say something unless I really have something to say. Otherwise, it's just noise. And, uh, yeah, yeah, it's a great topic.
Tarran
Is that how you avoid getting too sucked into the sort of Twitter arguments that there are out there? You know, just keep focusing?
Dean
Yeah, well, I, you know, that's another challenge, because it is a professional medium. And I you know, you have to recognize that if this is your professional profile, you are followed by your professional colleagues, and act accordingly. Because the lines with social media can get blurry, but it is probably prudent to, to treat it as a, you know, professional, it's a way for people to see your personality, I think. I mean, it's a way for people to understand a little about you, like, sometimes I'll mention, you know, I have kids, or sometimes I'll mention just other things I'm thinking about, but in the end, it is still a professional medium.
Pennington
I wonder, Natalie, now that you have these various experiences working with and in media, if you have advice for journalists who are trying to cover, I can't remember what the term was that used emerging infectious diseases or what it or whether it's COVID, which sort of seems kind of this anomalous thing, or something like Ebola, or Zika? If you have thoughts on how, you know, they could work to ensure they're covering it as well as they can?
Dean
Yeah, well, I, you know, I've met a lot of great reporters, I wouldn't deign to say I, you know, would have a lot of advice for them and know how some people have been, you know, working on this beat for a long time. And they're really such professionals. There. You know, I think just because the amount of COVID reporting has skyrocketed, the number of infectious disease reporters has skyrocketed. So it's been a lot of people have had to kind of learn on the job, you know, I guess, sort of expertise and experts. And I do feel like there is sometimes this, you know, and I say this as someone who's on Twitter, and I know that like people have contacted me, because I'm on Twitter, but making sure that people are kind of finding the right people to contact the people with really that is sometimes there I have a lot of colleagues who are not on Twitter, who are nega experts, obviously, and, and so one thing I have tried to do is redirect a lot of people to names they might not know, and then they can develop, develop those relationships.
Pennington
I'm just thinking because like, Brian, I started very, very small. And if I was a reporter, who had had to start covering this suddenly at my tiny news outlet, it would have been sort of a struggle to figure out how to like, move through all of the information and data.
Dean
Yeah, well, actually. Okay, so I was on the advisory board for COVID, the COVID tracking project. And, you know, there were a number of things that we discussed there. And, you know, what, if it is about empowering local newsrooms to be able to understand the data, they're, you know, to really digest the local data, because there's a lot of complexity, like, how do they define test positivity? How are things read, if you really dig into it, it's extremely complicated. So that was one thing we talked about, like having teaching set some sort of resource available for local reporters, data reporters, data journalists, to help, you know, at least be able to process the information and understand the limitations as well.
Bailer
Sort of as a kind of a complimentary question to that is, is preparing statisticians to work with journalists or preparing statisticians to engage publicly with some of these questions. What other advice might you give?
Dean
Yeah, I would say, well, Sir David Spiegelhalter has done an excellent job providing, you know, guidance sort of, so I'd recommend that people look through some really great tips. I mean, I think what I would, you know, one of the big things is understanding the subject matter, or, you know, really deeply and I, you know, that's just one of the persistent challenges is that you can't really be just, you know, sort of, just have statistical expertise, you also have to have enough of the subject matter expertise to be able to, to talk about it, you know, intelligently and yeah, I mean, I think figuring out your main message, you know, focusing on just a few main messages, using a lot of examples, figuring out, you know, like metaphor or analogy, like figuring out ways to explain things that that resonate with people, bringing it back to, to kind of what's going to impact their daily lives, and that's what people are really interested in. So recognizing the things you're interested in may be different from the things that reporters or interest Certain things that readers are interested in and trying to at least find the intersection there.
Tarran
And if in doubt, remember, Lego is a must on the use method of statistical
Pennington
Legos forever. That's all the time we have for this episode of Stats and Stories. Natalie, thank you so much for being here.
Bailer
Thank you, Natalie.
Dean
Yeah, thanks for having me.
Pennington
Stats and Stories is a partnership between Miami University’s Departments of Statistics, and Media, Journalism and Film, and the American Statistical Association. You can follow us on Twitter, Apple podcasts, or other places you can find podcasts. If you’d like to share your thoughts on the program send your email to statsandstories@miamioh.edu or check us out at statsandstories.net, and be sure to listen for future editions of Stats and Stories, where we discuss the statistics behind the stories and the stories behind the statistics.