Estimating the true prevalence of SARS-CoV-2

Estimating the true prevalence of SARS-CoV-2


The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has led to a global pandemic, but the true extent of the pandemic remains unknown. Many cases are asymptomatic and may therefore go undetected, and the accuracy of various newly developed diagnostic tests is not well known.

The researchers analyzed SARS-CoV-2 testing data collected by the state of Indiana. More than 3,600 randomly selected Indiana residents were tested for SARS-CoV-2 between April 25–29, 2020. Even though participants were selected randomly, substantial nonresponse rates and potential diagnostic testing errors may bias estimates of disease prevalence based on the data.

To address these concerns, the authors adjusted estimates of disease prevalence using Bayesian methods and data on test performance, daily and confirmed case and death counts, and census data. Census data suggested that non-White and Hispanic/Latino individuals were underrepresented in the sample.

The adjusted statewide prevalence estimate was nearly 40% higher than the unadjusted estimate, largely as a result of false negative tests and of underrepresented demographic groups being disproportionately affected by the pandemic.

Consequently, the authors estimated that there were 12 times as many total cases as confirmed cases statewide.

https://www.pnas.org/content/118/5/e2013906118

http://sciencemission.com/site/index.php?page=news&type=view&id=publications%2Fbayesian-estimation-of&filter=22

Edited

Rating

Unrated
Rating: