Probabilistic forecasts of the impact of vaccines and variants on the trajectory of the US COVID

In a summary of the report released today, Thomas McAndrew, a computer scientist and assistant professor at the College of Health at Lehigh University, includes probabilistic predictions of the impact of vaccines and variants on the trajectory of U.S. COVID in the United States. over the next few weeks. According to McAndrew, the purpose of the report is to “support public health officials, infectious disease modeling groups and the general public.”

Highlights of the report:

• A consensus of 91 forecasters predict that B.1.1.7. The variant will be found in 42% of all genetic sequences with an S gene mutation in the first two weeks of March and in 72% in all sequences between March 29 and April 4, 2021.
• The consensus among infectious disease modeling experts and trained Metaculus forecasters is that by February 28, 55,420,000 people will have received at least one dose of a vaccine. General forecasters at Good Judgment Open (GJO) – an online forecasting platform open to any interested member of the public – responded similarly with an implied median of 52,200,000 people receiving one or more doses of the vaccine. Preliminary data from the CDC shows that 49,772,180 people received an initial dose on February 28.
• Consensus forecasts from Metaculus and Good Judgment Open predicted a decrease in the rate of cases, deaths and hospitalizations for the last week of February (21-27).

The team will share with members of the Centers for Disease Control and Prevention (CDC), State Council and Territorial Epidemiologists, and members of MIDAS.

McAndrew’s approach to forecasting is different from the traditional approach. Rather than building a computational model to predict COVID cases, deaths and hospitalizations, it asks trained experts and forecasters to predict these targets and combine their predictions into a single consensus forecast.

Additionally, he and his team produce a meta-forecast, which is a combination of a set of computational models and their consensus forecasts.

The idea behind this approach is to combine computer models with human judgment to make more accurate predictions of the outbreak in the United States. “

Thomas McAndrew, Computer Scientist and Assistant Professor, College of Health at Lehigh University