AUTHOR=Sharmin Mahfuza , Manivannan Mani , Woo David , Sorel Océane , Auclair Jared R. , Gandhi Manoj , Mujawar Imran 

TITLE=Cross-sectional Ct distributions from qPCR tests can provide an early warning signal for the spread of COVID-19 in communities

JOURNAL=Frontiers in Public Health

VOLUME=Volume 11 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/public-health/articles/10.3389/fpubh.2023.1185720

DOI=10.3389/fpubh.2023.1185720

ISSN=2296-2565

ABSTRACT=SARS-CoV-2 PCR testing data has been widely used for COVID-19 surveillance, but current forecasting models primarily rely on binary PCR results, limiting their ability to accurately predict pandemic trajectories. To address this, a model utilizing cross-sectional population cycle threshold (Ct - the number of cycles required for the fluorescent signal to cross the background threshold) values obtained from PCR tests (Ct-based model) was developed. The goal of this study was to enhance COVID-19 forecasting models with features derived from the Ct-based model to detect epidemic waves earlier than case-based trajectories.
PCR data was collected weekly at Northeastern University (NU) from August 2020 to January 2022. Epidemic trajectories for the campus and county were generated based on case counts. The deep learning forecasting model was improved by incorporating Ct-based features and tested in Suffolk County and NU campus. Ct-based epidemic trajectories, including the effective reproductive rate (Rt) and incidence, were derived from cross-sectional Ct values.
The results showed that the Ct-based model estimated epidemic waves 12 to 14 days earlier than the traditional case-based approach, with a correlation of 0.57. The enhancement of the forecasting models with Ct-based information significantly reduced absolute error (by 49.4 and 221.5 for 7 and 14-day forecasts) and residual squared error (by 40.6 and 217.1 for 7 and 14-day forecasts) compared to the original model without Ct features. The study highlights that Ct-based epidemic trajectories can provide an early signal for impending epidemic waves in the community and improve the accuracy of transmission peak forecasts.
In conclusion, public health agencies should consider publishing Ct values along with binary PCR results to aid in early and accurate forecasting of epidemic waves. This information can inform timely public health policies and countermeasures to mitigate the spread of COVID-19. By leveraging Ct features, COVID-19 forecasting models can be enhanced, ultimately leading to more effective responses to the pandemic.