Probabilistic forecasts of the impact of vaccines and variants on the U.S. COVID trajectory

In a summary of the report published today, Thomas McAndrew, a computer scientist and assistant professor at Lehigh University’s College of Health, provides probabilistic projections about the impact of vaccines and variants on the COVID trajectory in the US over the next few weeks. The goal of the report, McAndrew said, is “to support public health officials, infectious disease modeling groups and the general public.”

Highlights of the report:

• A consensus of 91 forecasters predicts that the B.1.1.7. The variant is 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.
• Infectious disease modeling experts and trained forecasters from Metaculus agree that by February 28, 55,420,000 people had received at least one dose of vaccine. Generalistic forecasters from Good Judgment Open (GJO) – an online forecasting platform open to all interested members of the public – responded similarly, with an implicit median of 52,200,000 people receiving one or more vaccine doses. Preliminary data from the CDC shows 49,772,180 people were given a starting dose on February 28.
• Consensus predictions from the Metaculus and Good Judgment Open projected a decline in cases, deaths and hospitalizations for the final week of February (February 21-27).

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

McAndrew’s approach to forecasting is different from the traditional approach. Instead of building a computational model to predict cases, deaths and hospital stays due to COVID, he asks experts and trained forecasters to predict these goals and combines their predictions into a single consensus forecast.

In addition, he and his team create a metaforecast that represents a combination of an ensemble of calculation models and their consensus forecast.

The idea behind this approach is to combine computational models with human judgment to make more accurate predictions about the US outbreak. “

Thomas McAndrew, computer scientist and assistant professor at Lehigh University’s College of Health