Combining serological assays and official statistics to describe the trajectory of the COVID-19 pandemic: results from the EPICOVID19-RS study in (Southern

17 Background: The EPICOVID19-RS study conducted 10 population-based surveys in Rio Grande 18 do Sul (Southern Brazil), starting early in the epidemic. The sensitivity of the rapid point-of- 19 care test used in the first eight surveys has been shown to decrease over time after some 20 phases of the study were concluded. The 9 th survey used both the rapid test and an enzyme- 21 linked immunosorbent assay (ELISA) test, which has a higher and stable sensitivity. 22 Methods: We provide a theoretical justification for a correction procedure of the rapid test 23 estimates, assess its performance in a simulated dataset and apply it to empirical data from 24 the EPICOVID19-RS study. COVID-19 deaths from official statistics were used as an indicator of 25 the temporal distribution of the epidemic, under the assumption that fatality is constant over 26 time. Both the indicator and results from the 9 th survey were used to calibrate the temporal 27 decay function of the rapid test’s sensitivity from a previous validation study, which was used 28 to estimate the true sensitivity in each survey and adjust the rapid test estimates accordingly. 29 Results: Simulations corroborated the procedure is valid. Corrected seroprevalence estimates 30 were substantially larger than uncorrected estimates, which were substantially smaller than 31 respective estimates from confirmed cases and therefore clearly underestimate the true 32 infection prevalence. 33 Conclusion: Correcting biased estimates requires a combination of data and modelling 34 assumptions. This work illustrates the practical utility of analytical procedures, but also the 35 critical need for good quality, populationally-representative data for tracking the progress of 36 the epidemic and substantiate both projection models and policy making. 37 10 large, population-based surveys aiming at tracking the COVID-19 pandemic in Rio Grande do Sul. For this paper, only surveys 1 to 9 will be used. Data collection for the 1 st and 9 th surveys occurred in April 11-13 2020 and February 5-7 2021, respectively. The study involves nine sentinel cities – that is, the largest city from each one of the eight intermediate regions of the state as defined by the Brazilian Institute of Geography and Statistics (IBGE). The exception was the metropolitan region, for which two cities (the state capital and an additional city) were selected. In each city, 50 census tracts (delimited areas with approximately 300 households each in urban area, also defined by the IBGE) were selected with probability proportional to size and ensuring geographical representativity. In each tract, 10 households were systematically selected. Between surveys, different households were selected within the same tracts. One individual per households was randomly selected for interview and testing using the rapid point-of-care lateral-flow WONDFO SARS-CoV-2 Antibody Test (Wondfo Biotech Co., Guangzhou, China), which assesses both IgG and IgM