Analysis of COVID-19 case numbers: adjustment for diagnostic misclassification on the example of German case reporting data
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (ScreenIT)
Abstract
Background
Reported COVID-19 case numbers are key to monitoring pandemic spread and decision-making on policy measures but require careful interpretation as they depend substantially on testing strategy. A high and targeted testing activity is essential for a successful Test-Trace-Isolate strategy. However, it also leads to increased numbers of false-positives and can foster a debate on the actual pandemic state, which can slow down action and acceptance of containment measures.
Aim
We evaluate the impact of misclassification in COVID-19 diagnostics on reported case numbers and estimated numbers of disease onsets (epidemic curve).
Methods
We developed a statistical adjustment of reported case numbers for erroneous diagnostic results that facilitates a misclassification-adjusted real-time estimation of the epidemic curve based on nowcasting. Under realistic misclassification scenarios, we provide adjusted case numbers for Germany and illustrate misclassification-adjusted nowcasting for Bavarian data.
Results
We quantify the impact of diagnostic misclassification on time-series of reported case numbers, highlighting the relevance of a specificity smaller than one when test activity changes over time. Adjusting for misclassification, we find that the increase of cases starting in July might have been smaller than indicated by raw case counts, but cannot be fully explained by increasing numbers of false-positives due to increased testing. The effect of misclassification becomes negligible when true incidence is high.
Conclusions
Adjusting case numbers for misclassification can improve this important measure on short-term dynamics of the pandemic and should be considered in data-based surveillance. Further limitations of case reporting data exist and have to be considered.
Article activity feed
-
-
SciScore for 10.1101/2020.09.29.20203877: (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: Thank you for sharing your code and data.
Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:Our analysis has some assumptions and limitations. The adjustment for misclassification in COVID-19 diagnostics depends on accurate information with respect to the number of examined individuals. Such information is not directly available for the Bavarian data and we rely on a model based approach for relating the reported case numbers to the reported number of positive tests from Bavarian laboratories. The results appear plausible but …
SciScore for 10.1101/2020.09.29.20203877: (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: Thank you for sharing your code and data.
Results from LimitationRecognizer: We detected the following sentences addressing limitations in the study:Our analysis has some assumptions and limitations. The adjustment for misclassification in COVID-19 diagnostics depends on accurate information with respect to the number of examined individuals. Such information is not directly available for the Bavarian data and we rely on a model based approach for relating the reported case numbers to the reported number of positive tests from Bavarian laboratories. The results appear plausible but can not be directly validated. Another assumption are constant misclassification probabilities with respect to the person-specific COVID-19 diagnostics over time. This assumption might be violated in case of changes in the diagnostic procedures, changes in workload for the laboratories, additional laboratories that perform parts of the testing, and changes or improvements in operating processes. Indeed there is evidence that at least one of the laboratories newly entrusted with no-fee testing at the Bavarian southern borders was reporting single target positivity as positive test results. There is, however, still very few direct information on the quality of the PCR-testing available under field conditions and especially no comprehensive information on temporal changes. In consequence, misclassification bias due to imperfect specificity may vary regionally with respect to the area served by different laboratories and may well have been at the upper range of the assumptions for false positive results locally. Since nowcasting is performed on the...
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.
-