Probability-Based Estimates of Severe Acute Respiratory Syndrome Coronavirus 2 Seroprevalence and Detection Fraction, Utah, USA

Abstract

We aimed to generate an unbiased estimate of the incidence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection in 4 urban counties in Utah, USA. We used a multistage sampling design to randomly select community-representative participants >12 years of age. During May 4-June 30, 2020, we collected serum samples and survey responses from 8,108 persons belonging to 5,125 households. We used a qualitative chemiluminescent microparticle immunoassay to detect SARS-CoV-2 IgG in serum samples. We estimated the overall seroprevalence to be 0.8%. The estimated seroprevalence-to-case count ratio was 2.5, corresponding to a detection fraction of 40%. Only 0.2% of participants from whom we collected nasopharyngeal swab samples had SARS-CoV-2-positive reverse transcription PCR results. SARS-CoV-2 antibody prevalence during the study was low, and prevalence of PCR-positive cases was even lower. The comparatively high SARS-CoV-2 detection rate (40%) demonstrates the effectiveness of Utah's testing strategy and public health response.

Keywords: COVID-19; IgG; PCR; SARS; SARS-CoV-2; United States; Utah; antibodies; case detection; coronavirus; coronavirus disease; immunoglobulin G; incidence; infections; nasopharyngeal swabs; population surveillance; probability sampling design; rRT-PCR; respiratory infections; reverse transcription PCR; sensitivity; seroepidemiologic studies; serology; seroprevalence; severe acute respiratory syndrome coronavirus 2; specificity; viruses; zoonoses.

Figure

We corrected estimates of seroprevalence for assay error by applying the following formula, where P1 is the estimated prevalence within a given category of a predictor variable provided by the generalized linear models.