The Effects of School Reopenings on COVID-19 Hospitalizations

REACH COVID hospitalizations tech report cover 2021 1 04
View Technical Report

Click here for a searchable spreadsheet.

Below, we provide an overview of our study. A key finding of this study is that it appears safe to reopen schools in counties where there are fewer than 36 to 44 new COVID-19 county hospitalizations per 100,000 people per week. We have therefore created this searchable spreadsheet, which provides comparable data for every county in the country, to make it easier to compare these thresholds to current health conditions.

Here are the steps to search for your county information:

(1) Click “edit” in the top bar, then “find and replace” and type in the county name in the top box.

(2) Some county names arise in multiple states, so make sure you have the right one by looking at the state column (click “find” again at the bottom of the box to look for the next county with that name). Some counties are not included because data are not available.

(3) Look for the most recent week available (currently 12/25/20). We have also included data week by week back to late July, so that counties can see their trajectories. For example, a county with “20” in the second column indicates only 20 new COVID-19 hospitalizations per 100,000 county population per week.

Please note that we are not making a recommendation that any schools open or close to in-person instruction. We are also not responsible for the data, which are being provided by the U.S. Department of Health and Human Services (HHS). There is some evidence of missing data and data errors, including when you see consecutive weeks with identical hospitalization rates or sudden hospitalizations, which are unrealistic. Therefore, we also recommend checking with your local public health officials.

We encourage readers to read the full report and the notes at the top of the spreadsheet before consulting the spreadsheet. We plan to update the searchable spreadsheet every two weeks with the latest data made available by HHS. (This is the same underlying source as the COVID Tracking Project, but we are reporting data at the county level.)


Report Overview

The study is the first to examine how reopening schools in-person has affected COVID-19 hospitalizations. We found no evidence that reopening schools in-person or in a hybrid form increased COVID hospitalizations in the 75 percent of counties that had low COVID hospitalization rates during the summer, prior to reopening schools. Our results suggest that it seems safe to reopen schools when there are no more than 36 to 44 total new COVID hospitalizations per 100,000 people per week.

For counties that had higher rates of hospitalizations in the summer, our results were inconclusive, as they were inconsistent across data sources and statistical methods.

We used data on essentially all school districts’ reopening plans from Education Week and two private data collection companies, Burbio and MCH Strategic Data. We combined these with information on COVID-related hospitalizations from the company Change Healthcare, which has healthcare claims for 170 million people, roughly half of the entire U.S. population, as well as data from the U.S. Department of Health and Human Services (HHS) including essentially all U.S. hospitals. We examined data for every county from Jan. 1 through Oct. 30, 2020.

Previous Entry
COVID policy brief cover
Next Entry
COVER COVID Reopenings Fall Spring