Data-Driven Model Predicts Number of COVID-19 Deaths in US

April 30, 2020 - A Rutgers engineer has developed a data-driven model that accurately estimates the death toll related to the COVID-19 pandemic in the US and could be used around the world.

For more coronavirus updates, visit our resource page, updated twice daily by Xtelligent Healthcare Media.

The model, described in a study published in the journal Mathematics, predicted that the death toll would eventually reach about 68,120 in the United States as a result of COVID-19. That number is based on data available on April 28, and the model had a 99 percent confidence that the expected death toll would be between 66,055 and 70,304.

The model’s estimates and predictions closely match reported death totals. As of April 29, more than 58,000 Americans had died from complications related to COVID-19, according to the Johns Hopkins University COVID-19 Tracking Map.

"Based on data available on April 28, the model showed that the COVID-19 pandemic might be over in the United States, meaning no more American deaths, by around late June 2020," said Hoang Pham, a distinguished professor in the Department of Industrial and Systems Engineering in the School of Engineering at Rutgers University-New Brunswick.

"But if testing and contact tracing strategies, social-distancing policies, reopening of community strategies or stay-at-home policies change significantly in the coming days and weeks, the predicted death toll will also change.”

Other institutions have built data-driven models to predict the impact of coronavirus, and these researchers have also cautioned against ending social distancing practices too soon.

A team from MIT recently developed a machine learning model that leverages COVID-19 data and a neural network to determine the effect of quarantine measures and predict the spread of the virus. The model showed that with the current quarantine measures in place, the plateau for Italy and the US would arrive somewhere between April 15-20. The model also suggests that the infection count will reach 600,000 in the US before the rate of infection starts to stagnate.

“Our model shows that quarantine restrictions are successful in getting the effective reproduction number from larger than one to smaller than one. That corresponds to the point where we can flatten the curve and start seeing fewer infections,” said George Barbastathis, professor of mechanical engineering at MIT.

“This is a really crucial moment of time. If we relax quarantine measures, it could lead to disaster.”

Researchers at Stanford University have also demonstrated the potential consequences of reducing social distancing measures too quickly. Using an interactive, data-driven model, the group found that in order for short-term interventions to be effective, much higher levels of social distancing will be required.

“If we lift controls too quickly, we could see a resurgence, where cases pick back up quickly. In fact, if we completely stop practicing social distancing at almost any point in this model, we risk an epidemic that overwhelms hospital capacity,” the group stated.

“We will probably have to stay careful for a long time after it seems like COVID-19 is gone in order to keep it under control, or until a vaccine is developed (which could take twelve to eighteen months). That doesn't mean we will necessarily need to stay in our houses the whole time, though.”

The Rutgers team anticipates using their model to estimate COVID-19-related deaths around the world.

“Further work can be done to apply the proposed model to Covid-19 global death data as well as any other countries such as Italy and Spain where they also have a large cumulative number of deaths due to Covid-19. In the future, we intend to use the model in applications of population mortality and the spread of disease,” the research team stated.