Professor works with World Bank to track COVID-19 outbreak

When the coronavirus hit the largest urban areas of South Asia (specifically India) in late March, the usual rural-to-urban flow of migration was reversed. People were trying to return home from their city jobs, running into many others following the same route.

Under harsh conditions such as mounting heat and crowded foot traffic, many South Asians currently struggle to find proper transportation and shelter in their attempt to get where they need to go.

“In many countries, the spread of the virus is still in early days. It is the lockdown which is having the most immediate and sharpest consequences,” wrote FIU economics professor Abu Parves Shonchoy and his research team in a blog post on NYU's Financial Access Initiative covering migration mapping in the region. “As people stop going out, many households – particularly those that rely on income from microenterprise, informal labor and agriculture – have no cash coming in.”

For those who are going out, however, they are likely to bring the coronavirus with them.

Shonchoy collaborated with experts at the World Bank, as well as Oxford, UCLA and NYU to monitor migration patterns.  Shonchoy and his team are currently working on trying to use the data on pre-COVID-19 migration patterns to predict outbreak patterns in Spring 2020. The team will then compare their results with the actual outbreak to gauge the effectiveness of their model.

“It took us some time (about 3 weeks),” said Shonchoy. “Our team pulled together all the data from available sources and then we could create the map. We also took help from a World Bank visual graphics artist to make it pretty.”

Before the creation of the map, Shonchoy had been working with epidemiologists from the International Centre for Diarrhoeal Disease Research in Bangladesh and found that COVID-19 outbreaks outside of Dhaka (the capital of Bangladesh) were strongly correlated with migration.

“I began analyzing the data on Bangladesh and saw the pattern,” said Shonchoy. “The local migration stock information is a good predictor for the COVID-19 outbreak in Bangladesh.”

Shonchoy then started circulating his initial findings.

“I was previously at NYU,” said Shonchoy. “My colleagues got excited about the initial draft and decided to join the force to do a bigger pitch using the same methodology, focusing on South Asia, which share a similar origin, history, culture and also migration propensity.”

In the future, the team will analyze international migration as well, correlating migration patterns with COVID-19 cases, and integrating a broader range of economic and health data.

“Our plan is to create a contagion time-series map and to conduct rigorous analyses on the disease spread going forward,” said Shonchoy. “This will be a new data tool not only important for the policy makers but also for the epidemiologists.”