about agency
About Us

Johns Hopkins & University of Maryland Research Team

The tool is developed with supervision from Nilanjan Chatterjee, PhD, Bloomberg Distinguished Professor of Biostatistics and Oncology, School of Medicine. Dr. Chatterjee's research over many years in the past have focused on developing and evaluating models for the assessment for individualized risks of non-communicable diseases integrating information on genetic, demographic, anthropometric, life-style and environmental factors. The trainees who led various data analyses included:

  • Neha Agarwala, Department of Mathematics and Statistics, University of Maryland, Baltimore County
  • Jin Jin, Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health, Post-doctoral Fellow
  • Prosenjit Kundu, Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health, Post-doctoral Fellow
  • Yuqi Zhang, Department of Biomedical Engineering, Johns Hopkins University

The webtool was developed by Benjamin Harvey, DSc, a Senior Research Associate in the Department of Biostatistics, Johns Hopkins University, and the lead data scientist for the laboratory of Nilanjan Chatterjee, PhD

Our Method

Understanding Our Methodology

Please find more information on our risk-score calculation below.

The current risk-score calculation was developed using information on the risk for COVID-19 mortality associated with age, gender, race, social deprivation and 12 different health conditions published in a recent large UK study. The risk score was adopted to US setting by information on mortality rate by age and various race/ethnicity groups published by the Center for Disease Control. Further relative risks for individuals are provided by standardization with respect to average risk for the US population, calculated using available information on prevalence and co-occurrence of various risk-factors available form national databases. The model has been validated based on tens of thousands of recently observed deaths and projected risks across US cities and counties. Finally, the tool combines information of relative-risk with state-level forecasted death rates from a pandemic scenario model to estimate an absolute risk of mortality over a future specified time frame. A manuscript describing the details of the methods and showing projections of risk based on this model across a large number of US communities can be found here.