JMIR Publications

ABSTRACT

Background:

Wastewater surveillance provided early indication of COVID-19 disease in municipalities. Residents of long-term care facilities (LTCFs) experienced disproportionate morbidity and mortality early in the COVID-19 pandemic. We implemented LTCF building-level wastewater surveillance for SARS-CoV-2 at six facilities in Kentucky to provide early warning of SARS-CoV-2 in these vulnerable populations.

Objective:

To evaluate the performance of wastewater surveillance for SARS-CoV-2 at LTCFs in Kentucky.

Methods:

We conducted a mixed methods evaluation of wastewater surveillance following Centers for Disease Control and Prevention (CDC) guidelines for evaluating public health surveillance systems. Evaluation steps in the CDC guidelines were 1) engaging stakeholders, 2) describing the surveillance system, 3) focusing the evaluation design, 4) gathering credible evidence, and 5) generating conclusions and recommendations. We purposively recruited stakeholders for semi-structured interviews and undertook thematic content analysis of interview data. We integrated wastewater, clinical testing, and process data to characterize or calculate seven surveillance system performance attributes (simplicity, flexibility, data quality, sensitivity and positive predictive value (PPV), timeliness, representativeness, and stability).

Results:

We conducted eight stakeholder interviews. The surveillance system collected wastewater samples (N=780) 2 to 4 times weekly at six LTFCs in Kentucky from March 2021 to February 2022. Synthesis of credible evidence indicated variable surveillance performance. 1) Simplicity: Surveillance implementation required moderate human resource and technical capacity. 2) Flexibility: The system efficiently adjusted surveillance frequency and demonstrated ability to detect additional pathogens of interest. 3) Data quality: Software identified errors in wastewater sample metadata entry (3.5% of fields), technicians identified polymerase chain reaction (PCR) data issues (1.7% of reactions), and staff entered all data corrections into a log. 4) Sensitivity and PPV: Using routine LTCF SARS-CoV-2 clinical testing results as the “gold standard”, a wastewater SARS-CoV-2 signal of >0 RNA copies/ml was 30.6% (95% CI 24.4%-36.8%) sensitive and 79.7% (95% CI 76.4%-82.9%) specific for a positive clinical test at the LTCF. The PPV of the wastewater signal was 34.8% (95% CI 27.9%-41.7%) at >0 RNA copies/ml and increased to 75.0% (95% CI 60.0%-90.0%) at >250 copies/ml. 5) Timeliness: Stakeholders received surveillance data 24-72 hours after sample collection with delayed reporting due to lack of weekend laboratory staff. 6) Representativeness: Stakeholders identified challenges delineating the population contributing to LTCF wastewater because of visitors, unknown staff toileting habits, and the use of adult briefs by some residents preventing their waste from entering the sewer system. 7) Stability: The reoccurring cost to conduct one day of wastewater surveillance at one facility was approximately $144.50, which included transportation, labor, and materials expenses.

Conclusions:

The LTCF wastewater surveillance system demonstrated mixed performance per CDC criteria. Stakeholders found surveillance feasible and expressed optimism regarding its potential while also recognizing challenges in interpreting and acting on surveillance data.