Big Data Analytics Calculator Determines COVID-19 Mortality Risk

December 14, 2020 - Researchers from Johns Hopkins Bloomberg School of Public Health used big data analytics to develop a COVID-19 mortality risk calculator, a tool that could inform public health policies around vaccination.

In a study published in Nature Medicine, researchers noted that COVID-19 can impact people in drastically different ways. While children and young adults may experience mild illness or no symptoms at all, the elderly have infection mortality rates of at least several percent.

The pandemic has also highlighted stark racial and ethnic differences in disease outcomes. Black and Hispanic individuals in the US have died at much higher rates than whites, the team stated. People with comorbidities like diabetes have experienced worse outcomes from the virus as well.

To date, the US and other countries have mostly relied on community-based interventions, such as mask-wearing and social distancing. There are a range of pandemic scenario models that can predict future trends in infection and deaths at the population level, but predisposing factors have yet to be included in most mitigation efforts.

“Although we have long known about factors associated with greater mortality, there has been limited effort to incorporate these factors into prevention strategies and forecasting models,” said study senior author Nilanjan Chatterjee, PhD, Bloomberg Distinguished Professor in the departments of Biostatistics and Epidemiology at the Bloomberg School.

READ MORE: Case Studies Apply Big Data Analytics to Public Health Research

Researchers set out to develop a model that can estimate COVID-19 mortality risk associated with predisposing factors. The team used several COVID-19-related datasets, including a large UK-based study and state-level death rates published by the CDC, to train the model. Researchers then validated the model for predicting community-level mortality rates using recent deaths across US cities and counties.

The resulting COVID-19 mortality calculator enables users to determine individual risk based on factors such as age, sex, race or ethnicity, and medical history. Users can also use the calculator to define risk for a particular group, like a specific community, corporation, or university, based on the mix of relevant factors that define the group.

In the study, researchers used the calculator to describe the risk distribution for the whole US population. The team showed that only about four percent of the population at high risk – defined as five times greater risk than the US average – is expected to contribute close to 50 percent of the total deaths.

The group also showed that population-level risk varies considerably from city to city and county to county.

“For example, the percentage of the adult population exceeding the fivefold risk threshold varies from 0.4 percent in Layton, Utah, to 10.7 percent in Detroit, Michigan,” Chatterjee said.

READ MORE: Data Analytics Tool Helps Design COVID-19 Testing Strategies

The calculator combines information on individual-level factors with community-level pandemic dynamics to calculate mortality risk of individuals. When a big wave of infections hits a population, the risk estimates for individuals will rise in that community. The tool is updated on a weekly basis to incorporate information on state-level pandemic dynamics.

“People may understand broadly that with a preexisting condition such as obesity or diabetes, for example, they are at higher risk, but with our calculator they should be able to understand their risk in a way that takes multiple factors into account,” Chatterjee said.

The team also collaborated with PolicyMap, Inc., to develop interactive maps for viewing numbers and the proportion of individuals at various levels of risks across US cities, counties, and states. These maps will allow local policymakers to plan for vaccination, protecting high-risk individuals, and other targeted efforts.

Researchers expect that the risk calculator will help public health officials set priorities and allocate preventive resources, like N-95 masks. The tool can also help leaders allocate early COVID-19 vaccines, acting as a companion to guidelines from other organizations and ensuring that the right people are vaccinated first.

“Our calculator represents a more quantitative approach and should complement other proposed qualitative guidelines, such as those by the National Academy of Sciences and Medicine, for determining individual and community risks and allocating vaccines,” said Chatterjee.

READ MORE: Demographic Data Analytics Informs Public Policies During Pandemic

With the FDA recently issuing an emergency use authorization for a COVID-19 vaccine to Pfizer-BioNTech, the study’s real-world implications are even more critical.

“The FDA’s authorization for emergency use of the first COVID-19 vaccine is a significant milestone in battling this devastating pandemic that has affected so many families in the United States and around the world,” said FDA Commissioner Stephen M. Hahn, MD.

“Today’s action follows an open and transparent review process that included input from independent scientific and public health experts and a thorough evaluation by the agency’s career scientists to ensure this vaccine met FDA’s rigorous, scientific standards for safety, effectiveness, and manufacturing quality needed to support emergency use authorization.”