Graduate Student Selected As Semifinalist In MIT COVID-19 Datathon

Wednesday, May 27, 2020
HIA and MHA student uses data science to uncover health solutions for marginalized populations

Health Informatics and Analytics and Master’s in Health Administration student, Timothy Sokphat, recently participated in the MIT COVID-19 Datathon, in which his team was chosen as a semi-finalist in the disparities in health outcomes from COVID-19 research track. 

The week-long virtual event hosted 30 teams of data scientists, clinicians, public health professionals and other subject matter experts to leverage existing datasets to develop meaningful insights and influence policy and decision making in the public and private sectors. Research tracks included topics such as misinformation during the pandemic, measuring the impacts of COVID-19 policies, epidemiology of COVID-19, and megacity pandemic response in NYC.

“Our research focused on mining and analyzing COVID-19 data to identify existing disparities, especially for our patients of marginalized populations,” Sokphat said. “My role in the project consisted of combing through the datasets to uncover insights, analyzing outcomes and drawing our final conclusions from the data.” 

The team analyzed a host of incomplete data to determine which states accurately collected and reported racial and disparity related data, with the goal of identifying the ties between marginalized populations and the social determinants of health, such as access to health care, access to nutrition and socioeconomic status. 

Their research reflects the inequities of the U.S. health care system among marginalized patient populations, highlighted by the pandemic. It shows that while linkages exist between negative COVID-19 outcomes and citizens from a lower socioeconomic class, there are different demographic factors that are linked to outcomes. 

“For example, the negative outcomes from the hispanic population are more so linked to their minority status versus that of socioeconomic status,” Sokphat said. “Essentially, any proposed COVID-19 solutions can’t be based on a one-size-fits-all framework. Rather, they must be tailored to each individual population that’s affected.”

Sokphat jumped at the opportunity of participating in his first datathon after two hospital-based summer internship offers were rescinded due to the COVID-19 outbreak. 

“I still wanted to make my summer as meaningful as possible, so I thought ‘what better way than to help create solutions around COVID-19 data," Sokphat said. “Also, as an HIA student with the School of Data Science, I wanted to implement the newly formed skills I had learned from the previous semester.” 

While his team did not advance to the final round of the competition, they were encouraged to contact MIT approved mentors and partners to publish their research.

“We believe that it’s extremely important to encourage health care professionals and the federal government to continue to collect not only epidemiological data around COVID-19, but also racial and socioeconomic data,” Sokphat said. “They’re all drivers of patient outcomes and it’s crucial to have as much complete data as possible if we want to find concrete solutions to COVID-19.”

Learn more about the team’s findings by watching their final presentation here. The code used throughout the project can be found here via Github.