By Joe Garvey
A COVID-19 forecasting tool developed by researchers at Old Dominion University's Virginia Modeling, Analysis and Simulation Center (VMASC) is being used by Virginia health officials to inform their strategy for distributing COVID-19 vaccines.
ODU is assisting the state's departments of health, emergency management and emergency support team with forecasting daily new cases and vaccinations, COVID Community Vaccination Center (CVC) right-sizing, the emergency management support dashboard and answering forecast questions, such as how many Virginians will be vaccinated every two weeks until July 15?
"We are now providing VDEM and VDH with 21-day vaccination forecasts for all health districts and counties in Virginia for all three vaccines," said Barry Ezell, VMASC deputy executive director. "There are facility sizes of CVCs ranging from 250 to 6,000 doses per day. We are supporting with analysis what the correct size should be once the state makes the decision to stand them up. These are expensive decisions and hard to scale down if the state chooses the wrong size and location. A bad outcome would be to put a 3,000-doses-per-day CVC in a location and only get 400 people vaccinated per day."
Several VMASC researchers played leading roles in the vaccination effort:
Senior Project Scientist John Shull - data integration across multiple government systems.
Research Assistant Professor Erika Frydenlund - analysis in vaccination equity for vulnerable populations.
Lead Project Scientist Alex Neilson - COVID vaccination dashboard development.
Lead Project Scientist Craig Jordan - CVS right-sizing.
The state asked the ODU team about a month ago to help with modeling related to its vaccination efforts, Maria Reppas, a spokesperson for the Virginia Department of Health, told The Virginian-Pilot.
"It began with assistance in forecasting and analysis of equitable distribution for vulnerable populations," Ezell said.
The basis of the vaccine forecasting model is the Virginia County COVID-19 Daily Case Total Forecaster, which was developed by VMASC researchers Christopher J. Lynch and Ross Gore in March 2020. Their model was designed to predict the spread of the disease down to the city and county level. Their project was taken on independently by VMASC as part of its mission to develop transparent computational models that help find solutions in complex areas.
