City-wide built environment SARS-CoV-2 detection for COVID-19 surveillance

Built environment surveillance has shown promise for monitoring COVID-19 burden at granular geographic scales, but its utility for surveillance across larger areas and populations is unknown. Our study aims to evaluate the role of built environment detection of SARS-CoV-2 for the surveillance of COVID-19 across broad geographies and populations. We conducted a prospective city-wide sampling study to examine the relationship between SARS-CoV-2 on floors and COVID-19 burden. We used non-parametric correlation analyses and linear models to evaluate associations between SARS-CoV-2 signals and COVID-19 outcomes, across multiple locations/populations over time. Sampling sites included schools, libraries, and emergency departments in the capital city of Ottawa, Canada, from October 2022 to March 2023. Floor sampling was performed across spaces, and outcomes were evaluated at aggregate levels. Detection (presence/absence) and quantification (viral load) of SARS-CoV-2 was determined by reverse-transcriptase polymerase chain reaction conducted on floor swabs collected weekly at study locations. The main outcomes, and measures of COVID-19 burden, were (1) weekly regional wastewater signal and (2) weekly admitted COVID-19 patient census from hospitals. We collected 1,863 built environment floor samples over the 6-month study period, with an overall swab positivity for SARS-CoV-2 of 45% (95%CI 43%-48%). We found a strong correlation between overall built environmental swab viral load and hospital COVID-19 census (Spearman's r = 0.64, p = 0.0017), but no correlation between regional wastewater and hospital COVID-19 census (Spearman's r = -0.15, p = 0.5). We found a strong correlation (Spearman's r = 0.76, p = 9x10-5) between hospital-specific swab viral copy number and hospital-specific COVID-19 census, which is a likely driver of the overall association between swab load and census. Built environment surveillance of SARS-CoV-2 from hospitals was strongly correlated with hospital burden with improved delineation of hospital COVID-19 cases compared to regional wastewater. These findings support the use of built environment surveillance for quantification of infectious burden amongst institutionalized groups.