COVID-19 containment on a college campus via wastewater-based epidemiology, targeted clinical testing and an intervention

Background: Wastewater-based epidemiology (WBE) has potential as an early-warning tool for determining the presence of COVID-19 in a community. The University of Arizona (UArizona) utilized WBE paired with clinical testing as a surveillance tool to monitor the UArizona community for SARS-CoV-2 in near real-time, as students re-entered campus in the fall. Wastewater results were utilized as a leading indicator to support public health decisions. Wastewater surveillance of student dormitories was used as the foundation of a strategy to contain potential outbreaks of disease following student re-entry onto campus for the Fall semester of 2020. Upon positive detection of viral RNA in wastewater samples, clinical testing was conducted on every individual living in the dorm. Infected individuals, asymptomatic and/or asymptomatic, were identified and relocated into an isolation dorm to prevent further disease transmission. Testing wastewater also allowed for evaluation of the effects of an intervention introduced onto the university community, mid-semester, namely a 'shelter in place' order imposed on the students. This presentation describes the community-wide protocols, testing plans, interventions, and results that took place to prevent over 90 outbreaks throughout the academic semester. The success of this study was featured in hundreds of international news outlets throughout Fall 2020. This study also provided the blueprint for numerous universities across the nation that are utilizing similar techniques pairing wastewater surveillance and targeted clinical testing to prevent the spread of COVID-19 amongst the campus community. Methods: Upon arriving to campus prior to the academic year, every individual was required to report to a designated COVID-19 testing site and undergo an anterior nasal swab for antigen testing. Individuals were kept on-site until tests results were noted. If a person had a positive COVID-19 test, they were required to isolate for a minimum of 10 days (at home or in a designated isolation dorm) from the onset of symptoms or from the date the sample was taken, per guidelines from the United States Center for Disease Control and Prevention (CDC). Therefore, all individuals on campus at the start of the semester were assumed to be free from SARS-CoV-2 infection. UArizona administration assembled a Task Force and Campus Re-Entry Working Groups to prepare for students' safe return to campus and classrooms. The WBE expert team hypothesized that surveillance of defined communities in dorms would provide an effective means for identifying new cases of COVID-19 since: 1) each dorm contained a known population; 2) dorm students provide a representation of the overall status of campus health; 3) wastewater samples could be collected from individual buildings; and 4) actionable public health responses could be initiated in the event of positive wastewater detection. Therefore, the university created a plan for wastewater surveillance of student dorms throughout the academic year. Samples were collected twice-per-week from manholes downstream from 13 dorms prior to convergence or mixing with other sewer lines, resulting in samples specific to individual buildings with defined communities. Results and Discussion: A baseline survey of SARS-CoV-2 RNA in wastewater was conducted prior to the fall semester via wastewater sample collection from multiple dorms between August 18-20, 2020, when some students had already returned for fall semester. No SARS-CoV-2 RNA was detected in these samples, which corroborated with the requirement that all students were determined negative prior to dorm entry. Only a few days after classes started in-person instruction, the first positive wastewater result was detected from Dorm A. This triggered immediate UA Task Force meeting, which supported additional wastewater testing to confirm results and mandatory clinical testing of all dorm residents. Two clinical tests for COVID-19 diagnosis were performed on all residents: antigen via anterior nasal swab and RT-PCR via nasopharyngeal swab samples. Data was de-duplicated when both tests resulted in positive. Figure 1 provides a timeline for the public health response actions and events described throughout the study. The next day, five wastewater samples collected five-minutes apart confirmed identical concentrations of SARS-CoV-2 RNA detected the day prior (Table 1). This validated the presence for SARS-CoV-2 in the wastewater samples that were representative of the resident population living in the dorm. Simultaneously, 270 antigen tests and 260 PCR tests were conducted on-site at Dorm A among the 311 total residents (Table 2). Antigen testing identified one positive individual (Person A) despite demonstrating no symptoms. The other 269 antigen tests were negative. An additional individual reported experiencing symptoms (Person B) and tested positive via an antigen test. Person A and Person B were immediately relocated into an isolation facility to prevent viral transmission. Over the next two days, antigen tests (34 individuals) and PCR tests (24 individuals) were conducted among individuals that were not originally available for testing. All clinical tests were negative (Table 2), as were corresponding wastewater samples, indicating that the source(s) for SARS-CoV-2 had likely been removed from the dorm. Only two days later, wastewater results returned to positive, but at very low concentrations. Clinical PCR tests confirmed another infected individual (Person C). Interestingly, this person previously had inconclusive results during the original testing in response to the first wastewater positive. It is possible that Person C was not shedding virus or had low viral shedding during that time, likely due to early stages of infection. Regardless, Person C was removed from the dorm and virus RNA was no longer detected in wastewater. Importantly, another resident (Person D) returned to the dorm the same day wastewater was positive, after following isolation protocols. The successful use of WBE coupled with targeted clinical testing in Dorm A provided confidence for an early-warning system that can avert COVID-19 outbreaks. Therefore, this program was expanded to monitor 13 dorms during the course of the semester. Specifically, 91 out of 111 (82.0% positive predictive value) positive wastewater samples successfully provided a warning that at least one individual living in the dorm was positive with a SARS-CoV-2 infection (Table 3); results were confirmed via a positive clinical test within 4 days of the wastewater sample. This corroborates with a recent report that suggests peak shedding occurs within the first week of illness (Cevik et al., 2020). Also, 185 out of 208 (88.9% negative predictive value) negative wastewater samples concurred with no new positive individuals on dormitory screenings (Table 3). Only 43 out of 319 total wastewater samples were discordant with results from clinical tests (Table 3). Positive wastewater samples not followed with positive clinical tests could be due to infected visitors and/or on few occasions less than 100% of the dorm residents tested. At the beginning of the semester there were no infections in the monitored dorms. Following campus re-entry, there were clusters of cases during the first 18 days of the semester. This was followed by a dramatic spike in cases beginning September 14, likely due to student behavior around the Labor Day holiday on September 7. Infections resulting from the holiday-related activities would be reflected in new cases being detected likely about five days thereafter (Lauer et al., 2020). Following the spike, the president of UArizona, endorsed a 'shelter in place' policy imposed by Pima County on September 15-29, 2020. The effectiveness of this intervention resulted in a dramatic decrease in the number of daily new-reported cases by September 21 (Figure 2). The decrease in new cases was associated with decreased viral concentrations in the wastewater over this period. Following this intervention, case counts remained remarkably low, and often zero for the rest of September, October, and into November. The vast majority of students left campus the Friday before Thanksgiving, November 20, and did not return to campus. Therefore, there are no WBE results to report for the month of December as the final weeks of courses and exams were conducted remotely online. Concluding Remarks: The combined strategy of utilizing WBE coupled with targeted clinical testing was critical in COVID-19 containment. WBE strategies averted potential transmission from numerous students, which allowed the university to remain open. The data from this study provide evidence that validate WBE as an important tool that can and should be used in tandem with diagnostics and infection prevention practices to intervene in COVID-19 transmission, and ultimately save lives.