Through two testing strategies, a total of 16 SNFs offered testing to either residents, employees, or both. The first testing strategy was directed by Public Health of Seattle & King County (PHSKC) and focused on SNF resident testing with employee testing offered at select sites. The second testing strategy was facilitated by the Seattle Flu Study (SFS) and directed at testing only employees. PHSKC testing was conducted by providers from University of Washington, between March 29, 2020, and May 8, 2020, at 13 SNFs and one assisted living facility, of which six offered both resident and employee testing and eight had only resident testing. Employee testing by SFS was designed to coincide with resident testing done by PHSKC when possible. SFS testing was conducted between April 14, 2020, and May 13, 2020, at 13 SNFs. At three SNFs, both the PHSKC and Seattle Flu Study teams tested SNF employees.
Population
PHSKC identified SNFs in need of SARS-CoV-2 testing, including sites with known COVID-19 cases, facilities with no known cases, or where COVID-19 testing of residents had not occurred. For testing through PHSKC, teams of healthcare workers collected nasopharyngeal (NP) swabs from all residents in a SNF during a single visit. For testing through Seattle Flu Study, facilities identified by PHSKC were contacted by the study team for employee testing. Facilities agreeing to participate messaged all employees before the visit to inform them of the upcoming testing event and distributed a copy of the informed consent form for previewing. Employees were eligible to participate if they worked at the facility and were over 18 years old. All testing was voluntary and not required by the employer, and employees were advised that results would not be reported directly to employers. Employees who reported prior testing for SARS-CoV-2 through other mechanisms were eligible for enrollment. Study staff consented individuals in English or in the participant’s language of preference using an interpreter. After informed consent was obtained, individuals completed an electronic tablet–based questionnaire (Project Redcap in REDCap, Nashville, TN) and self-collected a mid-nasal swab under observation by trained study staff.
Laboratory Methods
For testing through PHSKC, NP swabs from SNF residents were placed in universal viral transport media (Becton Dickinson, Franklin, NJ) and transported to the University of Washington Virology Laboratory for testing via a one-step real-time reverse transcription polymerase chain reaction (RT-PCR) assay following the SARS-CoV-2 CDC assay protocol, as previously described.17 No samples tested through PHSKC were resulted as indeterminant.
For testing through Seattle Flu Study, self-collected mid-nasal nylon-flocked swabs were placed in universal viral transport media (Becton Dickinson, Franklin, NJ) and transported to the Brotman Baty Institute for Precision Medicine and the Northwest Genomics Center for testing using a laboratory-developed test for SARS-CoV-2, as previously described.18 Briefly, SARS-CoV-2 detection was performed using real-time RT-PCR with a probe set targeting Orf1b and S with FAM fluor (Life Technologies 4332079 assays # APGZJKF and APXGVC4APX) multiplexed with an RNase P probe set with VIC or HEX fluor (Life Technologies A30064 or Integrated Data Technologies custom made) each in duplicate on a QuantStudio 6 instrument (Applied Biosystems). Three or four replicates for RNase P and SARS-CoV-2 were required to have a detection cycle threshold less than 40 for a sample to be considered positive for this laboratory-developed test, or both replicates must be positive in the research assay. Samples resulting with two replicates of positive SARS-CoV-2 detection were defined as inconclusive. Because tests determined to be inconclusive had SARS-CoV-2 detected in multiple replicates, these results were grouped with positive results for reporting purposes.
Data Collection
For individuals tested through PHSKC, data available included name, date of birth, date of testing, and whether the individual was a resident or an employee. For employees tested through the Seattle Flu Study, data included participant date of birth, date of testing, race and ethnicity, location and nature of work, new symptoms experienced during the last 7 days, and history of SARS-CoV-2 testing (Appendix 1 in the Supplementary Material). Information on SNF policies regarding absenteeism, infection control, and employee health were collected from SNF management by email 2 weeks following employee testing using a standardized data collection form (Appendix 2 in the Supplementary Material).
Data Analysis
Time between resident and employee testing was calculated as the days elapsed between first testing dates for each group at a SNF. For sites with multiple testing dates for employees, residents, or both, tests from all dates for a given group were combined to calculate the prevalence at each site. All data analysis was conducted in the R statistical language (R Foundation for Statistical Computing, Vienna, Austria). Frequencies were tabulated for social and demographic data. To test the association between residents and employees who tested positive for SARS-CoV-2, a two-tailed McNemar’s test was used. p values < 0.05 were considered statistically significant.
Reporting
For employees, positive or inconclusive SARS-CoV-2 test results were reported directly to participants by phone within 48 h and to the Washington State Department of Health. Resident results were reported to the ordering physician at the SNF.
Ethics
The Seattle Flu Study was approved by the University of Washington Institutional Review Board. Other testing of residents and employees was conducted as a public health surveillance activity under the direction of PHSKC.