RESUMEN
Since the start of the COVID-19 pandemic, there has been interest in using wastewater monitoring as an approach for disease surveillance. A significant uncertainty that would improve interpretation of wastewater monitoring data is the intensity and timing with which individuals shed RNA from severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) into wastewater. By combining wastewater and case surveillance data sets from a university campus during a period of heightened surveillance, we inferred that individual shedding of RNA into wastewater peaks on average six days (50% uncertainty interval (UI): 6 - 7; 95% UI: 4 - 8) following infection, and that wastewater measurements are highly overdispersed (negative binomial dispersion parameter, k = 0.39 (95% credible interval: 0.32 - 0.48)). This limits the utility of wastewater surveillance as a leading indicator of secular trends in SARS-CoV-2 transmission during an epidemic, and implies that it could be most useful as an early warning of rising transmission in areas where transmission is low or clinical testing is delayed or of limited capacity.
RESUMEN
Wastewater surveillance of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) RNA is being used to monitor Coronavirus Disease 2019 (COVID-19) trends in communities; however, within-day variation in primary influent concentrations of SARS-CoV-2 RNA remain largely uncharacterized. In the current study, grab sampling of primary influent was performed every 2 hours over two different 24-hour periods at two wastewater treatment plants (WWTPs) in northern Indiana, USA. In primary influent, uncorrected, recovery-corrected, and pepper mild mottle virus (PMMoV)-normalized SARS-CoV-2 RNA concentrations demonstrated ordinal agreement with increasing clinical COVID-19 positivity, but not COVID-19 cases. Primary influent SARS-CoV-2 RNA concentrations exhibited greater variation than PMMoV RNA concentrations as expected for lower shedding prevalence. The bovine respiratory syncytial virus (BRSV) process control recovery efficiency was low (mean: 0.91%) and highly variable (coefficient of variation: 51% - 206%) over the four sampling events with significant differences between the two WWTPs (p <0.0001). The process control recovery was similar to the independently assessed SARS-CoV-2 RNA recovery efficiency, which was also significantly different between the two WWTPs (p <0.0001). Recovery-corrected SARS-CoV-2 RNA concentrations better reflected within-day changes in primary influent flow rate and fecal content, as indicated by PMMoV concentrations. These observations highlight the importance of assessing the process recovery efficiency, which is highly variable, using an appropriate process control. Despite large variations, both recovery-corrected and PMMoV-normalized SARS-CoV-2 RNA concentrations in primary influent demonstrate potential for monitoring COVID-19 positivity trends in WWTPs serving peri-urban and rural areas.