Back

The IHME vs Me: Modeling USA CoVID-19 Spread, Early Data to the Fifth Wave

Eng, G.

2021-08-24 epidemiology
10.1101/2021.08.16.21262150 medRxiv
Show abstract

Epidemiologists have never had such high-quality real-time pandemic data. Modeling CoVID-19 pandemic data became a predictive tool in-stead of an afterwards analysis. How early CoVID-19 model predictions impacted US Government policies and practices is first reviewed here as an important part of the pandemic history. It spurred independent modeling efforts, such as this, to help develop a better understanding of CoVID-19 spread, and to provide a substitute for the IHME (Institute for Health Metrics & Evaluation, U. Washington) 4-month predictions for the expected pandemic evolution, which they had to revise every couple of weeks. Our alternative model, which was developed over the course of several earlier medrxiv.org preprints, is shown here to provide a good description for the entire USA CoVID-19 pandemic to date, covering: (1) the original CoVID-19 wave [3/21/20-6/07/20], (2) the Summer 2020 Resurgence [6/07/20-9/25/20], (3) the large Winter 2020 Resurgence [9/25/20-3/19/21], (4) a small Spring 2021 "Fourth Wave", [3/19/21-6/07/21], and (5) the present-day Summer 2021 "Fifth Wave" [6/07/21-present], which the USA is now in the midst of. Our analysis of the initial "Fifth Wave" data shows that this wave presently has the capacity to infect virtually all susceptible non-vaccinated persons who practice NO Mask-Wearing and minimal Social Distancing.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.