Will all SARS-CoV-2 seroprevalence surveys provide the right picture?

Valent, Francesca; Gongolo, Francesco

Abstract Objectives We discuss two main potential sources of bias that might affect the results of the Italian nationwide SARS-CoV-2 seroprevalence survey conducted by the Italian Ministry of Health and the National Institute of Statistics in a sample of 150,000 persons in 2,000 cities for epidemiological purposes, at the end of May-beginning of June 2020. Methods The first issue regards performance of serological tests. Comparing serological test results with real‐time reverse‐transcriptase polymerase chain reaction available in an area of the Italian Region Friuli Venezia Giulia, we suspect that test performance is less than optimal, but still acceptable for estimating prevalence of antibodies. The second issue regards participation bias, which might produce distorted estimates and reduce usefulness of the survey. Results We show that participation bias is possible in the Friuli Venezia Giulia Region and identify a number of potential causes. Conclusions We acknowledge the importance of good communication and careful planning for effective interventions.