Site-Specific Wastewater-Based Surveillance in Early Detection of COVID-19 New Cases and Prediction of Mass Testing Outcomes in Long-Term Care Facilities
Author(s)
Jiabi Wen, MSc1, Ken K. Peng, MSc2, Bonita E Lee, PhD1, Rhonda J. Rosychuk, PhD1, Tiejun Gao, PhD1, Judy Qiu, PhD1, Michael Y. Li, PhD1, Eleanor Risling, MD3, Lorie A. Little, MD3, Christopher Sikora, MD3, Xiaoli Pang, PhD1, Arto Ohinmaa, PhD1.
1University of Alberta, Edmonton, AB, Canada, 2Simon Fraser University, Vancouver, BC, Canada, 3Alberta Health Services, Edmonton, AB, Canada.
1University of Alberta, Edmonton, AB, Canada, 2Simon Fraser University, Vancouver, BC, Canada, 3Alberta Health Services, Edmonton, AB, Canada.
OBJECTIVES: Long-term care facilities (LTCFs) were disproportionately impacted during the COVID-19 pandemic. While mass testing supports outbreak control, it is invasive and resource-intensive. Site-specific wastewater-based surveillance (WBS) offers a non-invasive alternative by capturing collective viral signals in LTCF wastewater. This study evaluated WBS’s effectiveness in signalling new COVID-19 cases in LTCFs.
METHODS: Wastewater samples were collected from nine LTCFs in Edmonton, Canada, from January 2021 to May 2023. Constrained distributed lag models with quadratic functions were employed to explore associations between daily new cases and wastewater viral loads. We identified a critical window of lead time during which positive wastewater samples significantly indicated emerging COVID-19 cases. Using this critical window, we evaluated the predictive accuracy of wastewater samples for mass testing outcomes (defined as testing 10-100% of the facility) Fisher’s exact test and Mann-Whitney U test compared WBS accuracy for predicting resident vs. staff cases and examined whether factors like sample type, quantity, outbreak duration, or collection timing influenced accuracy.
RESULTS: Among 2,505 wastewater samples collected, 909 were positive, alongside 825 clinical cases identified from 18,226 clinical specimens. Before clinical testing scaled down in 2022, critical windows for signalling new cases were consistently within five days, with eight facilities showing critical windows within three days. Of 198 mass testing events, 140 had one or two wastewater samples collected 1-3 days in advance. The accuracy of wastewater in predicting mass testing outcomes was not significant across testing scales at 20-100%. For a testing scale ≥20%, three-day-in-advance wastewater predicted 85% of negative and 60% of positive mass testing outcomes. WBS more accurately predicted resident cases than staff cases (74% vs. 33%, p=0.02). Other factors did not significantly affect prediction accuracy.
CONCLUSIONS: WBS effectively signalled COVID-19 cases among LTCF residents, supporting timely outbreak responses to protect vulnerable older populations and potential cost savings by optimizing testing strategies.
METHODS: Wastewater samples were collected from nine LTCFs in Edmonton, Canada, from January 2021 to May 2023. Constrained distributed lag models with quadratic functions were employed to explore associations between daily new cases and wastewater viral loads. We identified a critical window of lead time during which positive wastewater samples significantly indicated emerging COVID-19 cases. Using this critical window, we evaluated the predictive accuracy of wastewater samples for mass testing outcomes (defined as testing 10-100% of the facility) Fisher’s exact test and Mann-Whitney U test compared WBS accuracy for predicting resident vs. staff cases and examined whether factors like sample type, quantity, outbreak duration, or collection timing influenced accuracy.
RESULTS: Among 2,505 wastewater samples collected, 909 were positive, alongside 825 clinical cases identified from 18,226 clinical specimens. Before clinical testing scaled down in 2022, critical windows for signalling new cases were consistently within five days, with eight facilities showing critical windows within three days. Of 198 mass testing events, 140 had one or two wastewater samples collected 1-3 days in advance. The accuracy of wastewater in predicting mass testing outcomes was not significant across testing scales at 20-100%. For a testing scale ≥20%, three-day-in-advance wastewater predicted 85% of negative and 60% of positive mass testing outcomes. WBS more accurately predicted resident cases than staff cases (74% vs. 33%, p=0.02). Other factors did not significantly affect prediction accuracy.
CONCLUSIONS: WBS effectively signalled COVID-19 cases among LTCF residents, supporting timely outbreak responses to protect vulnerable older populations and potential cost savings by optimizing testing strategies.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
EPH215
Topic
Epidemiology & Public Health, Methodological & Statistical Research
Topic Subcategory
Public Health
Disease
Geriatrics, Infectious Disease (non-vaccine)