Meet the Mind: Dr. Lauren Meyers
ALL OF US kinda, sorta know we’re living in unprecedented times—but Dr. Lauren Meyers can actually confirm the sentiment. The University of Texas at Austin Professor of Integrative Biology is one of the world’s foremost experts in the field of epidemiology, and specifically epidemic modeling, an area that leverages public health data to compute potential outcomes and offer guidance to policymakers. Suffice to say, Meyers’ work has been very much in demand over the last year with the rise of COVID-19.
“This is unprecedented; it’s hitting home like nothing we’ve experienced in our lives and the demands are so many,” Meyers tells Cognitive Times. “I’ve gone from a team of me and my five students/post- docs to 60 people on a call every morning, all collaborating, all working around the clock. The pace of work has increased tenfold, if not more. It’s intense.”
When Meyers says “unprecedented,” she knows. Her roughly two-decade career in epidemiology has coincided with perhaps some of the field’s all-time biggest changes and challenges. As a postdoc student studying evolutionary biology at Emory University in the early 2000s, she fell into epidemiology when the CDC came down the road looking for modelers to “model the spread and containment of respiratory infections in close settings like health care facilities and nursing homes, something that’s very relevant today,” she recalls. That work landed her a call from the British Columbia Centre for Disease Control, where staff had read a paper she co-authored with the CDC and wanted help with a new disease called SARS. From then on, Meyers would go on to help guide the response of policymakers through her research and modeling on diseases like H1N1, Ebola, Zika, HIV, and now COVID-19.
“[When I started doing] pandemic or epidemic modeling, it didn’t exist,” she says. “There are so many things that have changed in the last decade—the technology, all the models, the machine learning, everything we use—but there’s also been incredible growth in the field. There were maybe a dozen of us when I started, but now I’m on multiple calls with the CDC every week and there are dozens of academic modelers on those calls from universities from around the world.”
As that string of headline-grabbing diseases unfolded in the 2000s, epidemiology itself was transforming. Around the turn of the century, new epidemiologists like Meyers were borrowing theories from other academic disciplines and applying them in their field (like statistical physics’ percolation theory, which models how a fluid flows through a bunch of channels that are either open or closed in order to predict the extent of spread). Simultaneously, technology rapidly evolved—allowing for more near-real-time data collection, reporting, storage, and computation—so epidemiologists also adopted these tools as quickly as possible to ensure their new modeling approaches could be based more and more on hard data.
“[In the last 20 years,] we’ve been through a data revolution,” Meyers says. “When I first started, there was almost no data, there really was so little. We were a data-poor field, because it was so hard to get info on how many people were infected on a given day until very recently. Instead, we had to do a lot of counterfactual work— if it were like this, this is what it would look like. Let’s put it into the model and see what unfolds.”
“The real innovations came from taking the models we were developing and figuring out how to fit them to data, to more directly engage these complex models with what was being seen in the world,” she continues. “So there are all sorts of methods— Bayesian statistics, AI and machine learn- ing—that have come online and allowed us to take these models that could capture a lot about the real world if they have the right inputs, and directly engage them with data to infer how things are spreading. That lets us make more accurate predictions about how epidemics might unfold or to evaluate how different intervention actions could impact the spread of disease, things we’re very interested in with COVID-19.”
When it comes to epidemiologists and the novel coronavirus, Meyers identifies one other important change within the last few decades—the reception awaiting the work of the doctor and her colleagues. While some of today’s highest-profile data naysayers routinely make the nightly news with ill-informed policy decisions based on incomplete or nonexistent data, Meyers still sees our current culture as being more open to data-driven epidemiology than ever. She has distinct memories of walking into public health agencies as recently as 2009 and being met with near-instant dismissal: “Someone invited me, but someone else led the room. They said, ‘I don’t believe in models. I think they’re crap and we have better things to do,’” Meyers says. It was a common conversation 10 years ago, she recalls, but thankfully modern epidemiology and its mathematical, machine learning methods meet more amenable minds these days.
“Most decision-makers today want evidence, data, and science. I get weekly requests from, the White House and CDC to local leaders in cities across Texas and the country. They’ve seen our models online, and they want to know how actions they may take can impact the health and safety of their communities,” she says. “These models aren’t perfect; they’re only as good as the inputs so they’re always changing. But it’s always so much better to bring whatever little evidence you can to the table rather than shooting from the hip. That’s what we can do right now.”
To watch Dr. Meyers speak on the pandemic, watch her latest Time Machine AI Session talk at: go.timemachine.ai/ covid-19.