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

  1. Fernando P. Hartwig
  2. Luís Paulo Vidaletti
  3. Aluísio J. D. Barros
  4. Gabriel D. Victora
  5. Ana M. B. Menezes
  6. Marilia A. Mesenburg
  7. Bernardo L. Horta
  8. Mariângela F. Silveira
  9. Cesar G. Victora
  10. Pedro C. Hallal
  11. Claudio J. Struchiner

Reviewed by

Read the full article

Listed in

Log in to save this article

Abstract

Background

The EPICOVID19-RS study conducted 10 population-based surveys in Rio Grande do Sul (Southern Brazil), starting early in the epidemic. The sensitivity of the rapid point-of-care test used in the first eight surveys has been shown to decrease over time after some phases of the study were concluded. The 9 th survey used both the rapid test and an enzyme-linked immunosorbent assay (ELISA) test, which has a higher and stable sensitivity.

Methods

We provide a theoretical justification for a correction procedure of the rapid test estimates, assess its performance in a simulated dataset and apply it to empirical data from the EPICOVID19-RS study. COVID-19 deaths from official statistics were used as an indicator of the temporal distribution of the epidemic, under the assumption that fatality is constant over time. Both the indicator and results from the 9 th survey were used to calibrate the temporal decay function of the rapid test’s sensitivity from a previous validation study, which was used to estimate the true sensitivity in each survey and adjust the rapid test estimates accordingly.

Results

Simulations corroborated the procedure is valid. Corrected seroprevalence estimates were substantially larger than uncorrected estimates, which were substantially smaller than respective estimates from confirmed cases and therefore clearly underestimate the true infection prevalence.

Conclusion

Correcting biased estimates requires a combination of data and modelling assumptions. This work illustrates the practical utility of analytical procedures, but also the critical need for good quality, populationally-representative data for tracking the progress of the epidemic and substantiate both projection models and policy making.

Article activity feed

  1. ScreenIT

    SciScore for 10.1101/2021.05.21.21257634: (What is this?)

    Please note, not all rigor criteria are appropriate for all manuscripts.

    Table 1: Rigor

    NIH rigor criteria are not applicable to paper type.

    Table 2: Resources

    No key resources detected.

    Results from OddPub: We did not detect open data. We also did not detect open code. Researchers are encouraged to share open data when possible (see Nature blog).

    Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:
    An important limitation of the study is the need of many assumptions throughout the correction process, including: fatality ratio is constant over time; and the sensitivity function estimated in the validation study (which was enriched for symptomatic cases) is applicable to the field (which includes the general population) after a calibration procedure. It is not possible to empirically verify the assumption of constant fatality over time without making additional assumptions that may lead to circular reasoning. Regarding the sensitivity function, its validity can only be assessed by repeatedly applying the rapid test over time to individuals sampled from the general population who had been diagnosed with known date. Establishing such cohort would be logistically difficult and time consuming, requiring large initial sample sizes to identify enough infected individuals. It should also be mentioned that the ELISA test is itself not perfect. Indeed, there is evidence indicating that some individuals do not seroconvert, and this may be associated with disease severity(13). Therefore, estimates presented here must be interpreted as the cumulative prevalence of positive ELISA tests rather than of true infections. This study is not a definitive guide, but rather an example of the usefulness of integrating different sources to correct estimates from an imperfect test to obtain a more plausible temporal trend of the COVID-19 epidemic. Although the study demonstrates the practical imp...

    Results from TrialIdentifier: No clinical trial numbers were referenced.

    Results from Barzooka: We did not find any issues relating to the usage of bar graphs.

    Results from JetFighter: We did not find any issues relating to colormaps.

    Results from rtransparent:
    • Thank you for including a conflict of interest statement. Authors are encouraged to include this statement when submitting to a journal.
    • Thank you for including a funding statement. Authors are encouraged to include this statement when submitting to a journal.
    • No protocol registration statement was detected.

    Results from scite Reference Check: We found no unreliable references.

    About SciScore

    SciScore is an automated tool that is designed to assist expert reviewers by finding and presenting formulaic information scattered throughout a paper in a standard, easy to digest format. SciScore checks for the presence and correctness of RRIDs (research resource identifiers), and for rigor criteria such as sex and investigator blinding. For details on the theoretical underpinning of rigor criteria and the tools shown here, including references cited, please follow this link.

  2. Version published to 10.1101/2021.05.21.21257634v1 on medRxiv