Paths from research to impact: a year of collaborative research on Covid-19
March of 2020 was an unforgettable time at MIT. As faculty transitioned their courses to be virtual and students left campus, it wasn’t easy to imagine how the pandemic would impact learning, research, and communities — all essential threads in the fabric of MIT life.
Almost immediately groups around campus began to explore how their expertise could be applied to coronavirus challenges. IDSS was no different. “Even as we were locking down, conversations were already beginning,” says Munther Dahleh, a professor of electrical engineering and computer science and director of IDSS. “What can we do to help? How can we contribute?”
The result, says Dahleh, was an opportunity for the community to come together despite being far apart, to attract talented people from around the world, and to connect on an urgent problem where the possibility of having an impact was imminent. The IDSS Covid-19 Collaboration, dubbed Isolat, explored challenges like testing and re-opening, ultimately leading to recommendations for MIT and other universities and a testing impact calculator for organizations like warehouses, offices, and schools.
One year later, IDSS (including the Technology and Policy Program) hosted a day-long workshop to bring together the Isolat group and other researchers at MIT and around the world whose data-driven covid-19 research has informed policy decisions and impacted the pandemic response on both local and global scales. The event, ‘Paths from Research to Impact: a year of collaborative research on Covid-19’ included two panel talks and two sessions of contributed talks that covered testing, modeling, forecasting, measuring the impact of policies, equity, and other key issues associated with the pandemic.
Testing = Control
The morning panel, ‘Controlling the Pandemic: Testing, Predicting, and Advising Policy,’ showcased strategies to control the spread of Covid-19 through testing. Anette ‘Peko’ Hosoi, a mechanical engineering professor and IDSS faculty member, discussed Isolat’s impact on MIT policy and the creation of the resources at whentotest.org.
“We started with basic questions,” says Hosoi. “How much testing should we be doing? Who should we be testing? How many people? How often?” For Hosoi, who advised the Biden-Harris transition team, the biggest challenges were not noisy data or complex modeling. “The big challenge was operationalizing this method, from educating decision makers to understanding community constraints.”
Hosoi was joined on the panel by three others who helped operationalize testing around the world. Kimon Drakopoulos (SM ’11, PhD ’16), an alum from the MIT Laboratory for Information and Decision Systems (LIDS), partnered with the Greek government to develop a machine learning tool to guide the distribution of limited Covid-testing resources at the borders of his home country, Greece. Fernando Paganini, an engineering professor at Universidad ORT Uruguay who was a postdoctoral fellow at LIDS, was appointed to a cross-disciplinary group that advised the Uruguayan president on the Covid-19 pandemic. Mathukumalli Vidyasagar, a Distinguished Professor with the Indian Institute of Technology Hyderabad, worked with a committee to model progression of Covid-19 in India. All agreed that the work presented a set of challenges outside of the science itself, including explaining complex concepts without getting too technical and communicating cautiously with the media.
When working with policymakers, Drakopoulos offered this advice: “Be transparent and give concrete recommendations. Do the theoretical stuff in the background. Being practical is more important than your model.”
The researchers had advice for policymakers, too. “Don’t just declare victory,” added Vidyasagar. “Keep listening to the scientists.”
Equity and disparate impacts
Global vaccination distribution remains a key, ongoing pandemic challenge, one that brings the issue of equity into stark relief. While the US, Canada, and Europe race to vaccinate their citizens, people in other areas of the world — Africa, the Middle East, and much of South America — still struggle with vaccine access and infrastructure.
The pandemic has shone a spotlight on many areas of inequity, some local, others global. The afternoon panel of ‘Paths from Research to Impact’ explored ‘Disparate Impacts of the Pandemic: equity and economic challenges.’ IDSS affiliate and political science professor Adam Berinsky shared some of his work on social media, Covid misinformation, and vaccine hesitancy.
Berinsky was joined by UC Berkeley urban studies professor Karen Chapple, who investigated the pandemic’s eviction crisis by using machine learning to model areas of housing insecurity; Brown economist Emily Oster, who worked on the National Covid School Response Dashboard and examined the disparate impacts of school reopenings on different communities; and MIT economist Parag Pathak, who applied market design to the challenge of allocating scarce medical resources such as vaccines, masks, and ventilators.
The panelists agreed that their pandemic research has had a significant impact on their perspectives. “It’s changed the way I think about misinformation,” Berinsky said.
Added Pathak: “One thing I’m hoping to see after all these interactions over the last year — more data analysis skills!”
In addition to the two panels, the event featured contributed talks from speakers including Arnab Sarker, a student in the IDSS Social and Engineering Systems doctoral program who discussed forecasting the Covid-19 pandemic with mixtures of models and ‘curves;’ Xianglin ‘Flora’ Meng, a grad student with the Laboratory for Information and Decision Systems (LIDS) whose research suggests interventions that can reduce the brunt of suboptimal policies on socioeconomically disadvantaged populations; and IDSS professor Nicholas Ashford, who offered a comprehensive look at societal interventions that could reduce the inequalities that pre-existed the pandemic and were worsened by it.
“The Covid-19 collaboration reinforced how IDSS is a unit of MIT dedicated to addressing societal problems,” noted Noelle Selin, director of TPP, at the event’s conclusion. Within IDSS, there is not only more interest in addressing equity issues, echoed Dahleh, but a deeper appreciation of the expertise needed to impact policy with research — and in Isolat, a model for cross-disciplinary collaboration on real time challenges.
“It’s been inspiring to be a part of this group, and a part of this event,” said Emma Tegling, Associate Professor of Automatic Control at Lund University. Tegling is also a research affiliate with the Sociotechnical Systems Research Center (SSRC) and Isolat member who led the effort to organize the workshop.
Talks were not recorded, but presentation slides can be found on the agenda page of the event website.
