Early Adoption of Longitudinal Surveillance for SARS-CoV-2 among Staff in Long-Term Care Facilities: Prevalence, Virologic and Sequence Analysis

Abstract

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) emerged in 2019 and has become a major global pathogen in an astonishingly short period of time. The emergence of SARS-CoV-2 has been notable due to its impacts on residents in long-term care facilities (LTCFs). LTCF residents tend to possess several risk factors for severe outcomes of SARS-CoV-2 infection, including advanced age and the presence of comorbidities. Indeed, residents of LTCFs represent approximately 40% of SARS-CoV-2 deaths in the United States. Few studies have focused on the prevalence and transmission dynamics of SARS-CoV-2 among LTCF staff during the early months of the pandemic, prior to mandated surveillance testing. To assess the prevalence and incidence of SARS-CoV-2 among LTCF staff, characterize the extent of asymptomatic infections, and investigate the genomic epidemiology of the virus within these settings, we sampled staff for 8 to 11 weeks at six LTCFs with nasopharyngeal swabs from March through June of 2020. We determined the presence and levels of viral RNA and infectious virus and sequenced 54 nearly complete genomes. Our data revealed that over 50% of infections were asymptomatic/mildly symptomatic and that there was a strongly significant relationship between viral RNA (vRNA) and infectious virus, prolonged infections, and persistent vRNA (4+ weeks) in a subset of individuals, and declining incidence over time. Our data suggest that asymptomatic SARS-CoV-2-infected LTCF staff contributed to virus persistence and transmission within the workplace during the early pandemic period. Genetic epidemiology data generated from samples collected during this period support that SARS-CoV-2 was commonly spread between staff within an LTCF and that multiple-introduction events were less common. IMPORTANCE Our work comprises unique data on the characteristics of SARS-CoV-2 dynamics among staff working at LTCFs in the early months of the SARS-CoV-2 pandemic prior to mandated staff surveillance testing. During this time period, LTCF residents were largely sheltering-in-place. Given that staff were able to leave and return daily and could therefore be a continued source of imported or exported infection, we performed weekly SARS-CoV-2 PCR on nasal swab samples collected from this population. There are limited data from the early months of the pandemic comprising longitudinal surveillance of staff at LTCFs. Our data reveal the surprisingly high level of asymptomatic/presymptomatic infections within this cohort during the early months of the pandemic and show genetic epidemiological analyses that add novel insights into both the origin and transmission of SARS-CoV-2 within LTCFs.

Keywords: COVID-19; SARS-CoV-2; coronavirus; epidemiology; infectious disease; long-term care.

FIG 1

SARS-CoV-2 infections in six Colorado LTCFs. (A) SARS-CoV-2 N1 vRNA levels in nasopharyngeal swab (circles) or saliva (triangles) samples. Saliva was only sampled a single time (week 5) at two sites (A and B) due to nasopharyngeal swab shortages. y axis represents N1 copy number per swab or saliva sample. Dotted line indicates limit of detection. Numbers across the top indicate number of samples tested each week. (B) Prevalence of SARS-CoV-2 each week at each site (percentage of samples with detectable N1 vRNA out of total number tested). (C) Incident cases were defined as individuals who tested positive for N1 vRNA for the first time and had tested negative for infection 1 or 2 weeks prior. Not shown are prevalent infections among workers tested for the first time in week 2.

FIG 2

Relationship between SARS-CoV-2 viral RNA and infectious virus. Samples with detectable SARS-CoV-2 N1 vRNA were evaluated for N2 and E vRNA and infectious virus. (A) Relationship between levels of N1, N2, and E vRNA transcripts. (B) Genome/genome ratios between N1/N2, N1/E, and N2/E (median values with interquartile ranges). (C) Relationship between levels of infectious virus and levels of N1, N2, and E vRNA. (D) Specific infectivity (genome/PFU ratio) of infectious virus relative to N1, N2, and E transcripts (median values with interquartile ranges). Dashed lines represent limit of detection.

FIG 3

SARS-CoV-2 symptom status, severity, and relationship to viral RNA. (A) Numbers of symptoms reported by vRNA⁻ and vRNA⁺ participants (mean values ± SD). (B) Percentages of vRNA⁻ and vRNA⁺ individuals stratified by number of symptoms. (C) Percentages of vRNA⁺ survey participants reporting total numbers of symptoms. (D) Cumulative symptom score (not reported = 0, mild = 1, medium = 2, severe = 3) for all 11 symptoms stratified by vRNA⁻ and vRNA⁺ participants (mean values ± SD). (E) Percentages of vRNA⁻ and vRNA⁺ individuals stratified by symptom score. (F) Relationship between cumulative symptom score and N1 vRNA levels (semilog nonlinear regression line fit). ***, P < 0.0001 by Mann-Whitney unpaired nonparametric test.

FIG 4

Individual infection courses and virus levels. Viral N1 RNA (left axis) and infectious virus (right axis) in select individuals with detectable N1 for 1 (A), 2 (B), 3 (C), or 4 (D) consecutive weeks. (E) Examples of individuals with detection of N1 vRNA after a period of undetectable N1 following initial infection. (F) N1 vRNA and infectious virus by week of infection are plotted for individuals with incident infections during the course of the study, with negative (N) tests immediately before and after positive (P) tests stratified by the length of infection (1, 2, 3, or 4 consecutive positive weeks) and by those who experienced a postnegative positive test (a positive test after 1 to 3 negative weeks following initial infection). Dashed lines represent limit of detection; samples with result of “not detected” are plotted at half the limit of detection.

FIG 5

Phylogenetic analysis of SARS-CoV-2 genomes collected from Colorado LTCFs. (A) PhyML tree constructed using Tamura-Nei distance model, including both transitions and transversions, in Geneious Prime. Numbers at nodes indicate bootstrap confidence based on 1,000 replicates. Distance matrix was computed, and the tree was visualized in Geneious Prime. Letters preceding taxon designations represent job codes (AC, activities; AD, administrative; AM, admissions; DT, dietary; MT, maintenance; NS, nursing; SS, social services; UK, unknown), and letters A to E indicate site of origin. Numbers after underscores indicate the date of sample collection. Reference sequences and four Colorado-derived sequences were obtained from NCBI. (B) Map of the LTCFs’ relative geographic locations and distances from one another.