UT researchers created a new online dashboard that tracks COVID-19 hospitalization rates in Texas and provides a forecast for how the epidemic is spreading for different regions in the state.
UT’s COVID-19 Modeling Consortium uses hospitalization data as well as anonymous cell phone mobility data to predict how the pandemic will progress in each of the 22 Trauma Service Areas that encompass Texas, said Spencer Fox, associate director of the consortium. The consortium created the new dashboard to help decision-makers at the local level as well as help community members understand infection rates and likely virus trends in their area, Fox said.
“What our model does is it looks at the noisy data we’re getting about the state of the (COVID-19) pandemic, and it provides a quick summary of the current state as well as what projections might look like over the next few weeks,” Fox said.
Transmission rate in a region can be determined in part by its reproduction number, or how many people a positive case is likely to infect, Fox said. If the reproduction number is above one, such as when Texas reopened in the summer and in March, the spread of COVID-19 is increasing.
Fox said most regions in Texas are experiencing a decline in the spread of the coronavirus, but if trends start to rise for the fall, community members can use this dashboard to help make informed decisions, such as whether or not to send their children to in-person schooling.
Researchers are closely watching how in-person schooling impacts the trends shown on the dashboard, Fox said. Anonymous cell phone data provides researchers with metrics such as the average amount of time spent at home and the number of visits to restaurants, bars and schools, Fox said.
Fox said projections estimate Austin will continue to have stable or declining trends over the next few weeks as long as the community maintains social distancing measures.
“Over the past week or two, Austin has noticed a slight uptick in cases, and Mark Escott (interim medical director and health authority for Austin Public Health) mentioned that a lot of these cases were from the high school-age population,” Fox said. “What we saw over the summer was that infections that start in younger individuals don't stay there … If there is transmission happening, it makes its way to vulnerable populations.”
Biology graduate student Remy Pasco, who is a research assistant for the COVID-19 Modeling Consortium, said reopening measures, such as businesses operating at 75% capacity as of Thursday, will also be watched carefully to determine impact on transmission rates.
“Those are the kind of things we are watching because we knew it had a big impact in June,” Pasco said. “Now, things look OK, but we need to be very careful.”
UT spokesperson J. B. Bird said in an email that the dashboard is better suited to the Greater Austin area since it focuses on hospitalizations, and UT’s younger demographic has lower hospitalization and ICU levels.
“Our planners look carefully at data for Austin and Travis County as part of their ongoing, regular assessments of pandemic conditions, and these new tools from the Modeling Consortium will be a great asset in that work,” Bird said.