key: cord-0789088-rlevoj41 authors: Vijh, Rohit; Ghafari, Cher; Hayden, Althea; Schwandt, Michael; Sekirov, Inna; Morshed, Muhammad; Levett, Paul; Krajden, Mel; Boraston, Suni; Daly, Patricia; Lysyshyn, Mark; Harding, John; McLennan, Meghan; Chahil, Navdeep; Mak, Annie; McKee, Geoff title: Serological Survey following SARS-COV-2 Outbreaks at Long Term Care Facilities in Metro Vancouver, British Columbia: Implications for Outbreak Management and Infection Control Policies date: 2020-10-18 journal: Am J Infect Control DOI: 10.1016/j.ajic.2020.10.009 sha: 3043777623ebcdd3eeddc9581815afbc31bc5538 doc_id: 789088 cord_uid: rlevoj41 A cross-sectional serological survey was carried out in two long term care facilities that experienced COVID-19 outbreaks in order to evaluate current clinical COVID-19 case definitions. Among individuals with a negative or no previous COVID-19 diagnostic test, myalgias, headache and loss of appetite were associated with serological reactivity. The US CDC probable case definition was also associated with seropositivity. Public health and infection control practitioners should consider these findings for case exclusion in outbreak settings. A cross-sectional serological survey was carried out in two long term care facilities that experienced COVID-19 outbreaks in order to evaluate current clinical COVID-19 case definitions. Among individuals with a negative or no previous COVID-19 diagnostic test, myalgias, headache and loss of appetite were associated with serological reactivity. The US CDC probable case definition was also associated with seropositivity. Public health and infection control practitioners should consider these findings for case exclusion in outbreak settings. Given the high mortality rate associated with COVID-19 among LTC residents 1 , timely and evidenceinformed interventions are critical for mitigating transmission risk. Serological testing may be useful to evaluate and inform public health infection control practices by uncovering cases missed during an outbreak using current laboratory-based and clinical case definitions. 2 Communicable disease case definitions can be utilized in public health for a variety of purposes (ex. surveillance). In the context where diagnostic tests are not rapidly available or have limited sensitivity, symptom-based case definitions are essential. In LTC outbreaks, uncontrolled introduction of infections not identified through testing may perpetuate transmission despite outbreak control measures. Currently, various national probable/epidemiologically-linked (clinical) case definitions largely focus on respiratory symptoms (ie. cough and shortness of breath), with varying inclusion of systemic/generalized symptoms (ie. fever, chills, loss of appetite)(Appendix A). Given LTC residents often present with non-specific generalized symptoms for other respiratory pathogens 3 , potential cases of COVID-19 are likely missed and potentially contribute to propagation within LTC facilities. Our analysis aims to provide a descriptive overview of a serological survey of LTC residents and staff members following outbreaks at two facilities and evaluate clinical case definitions of COVID-19 used in LTC outbreaks against serological results. A cross-sectional serological survey of LTC residents and staff members was administered from Venous specimens were tested using an orthogonal approach 4 with five different commerciallyavailable SARS-CoV-2 antibody assays with varying target immunoglobulin and epitopes (Appendix B), in accordance with manufacturers' recommendations. Each individual was assigned by a medical microbiologist into "reactive", "non-reactive" or "equivocal" category based on degree of agreement/disagreement of aggregate antibody results from all tests. Clinical information (symptomatic/asymptomatic history, symptoms recorded, medical comorbidities, medications) for each individual was gathered by abstracting data from a standardized case report form (Appendix C), medical charts of LTC residents, and phone interviews. Resident symptoms were documented through a combination of resident report/staff observation and utilization of a standardized symptom checklist(Appendix D). Symptom onset dates were captured using both clinical information and diagnostic test data (Appendix E). Participants were classified as immunocompromised or immunocompetent using provincial criteria (Appendix F). Data on clinical information and diagnostic test results were abstracted from May 22 nd to June 5 th 2020. Descriptive statistics of the study population were summarized in R (v.3.6.2) and STATA (v.15). Multi-variable logistic regression (adjusting for age, gender and facility) was used to generate adjusted odd ratio (aOR) estimates of associations between serological results and different individual symptoms, symptom clusters (Appendix A), immunocompromise status (yes vs. no) and history of negative NAATs (<3 vs.3). Covariate selection accounted for differences between staff and residents (age, gender) and facility characteristics. Individuals for whom we could not access a clinical history were excluded from regression analyses (n=6). Research ethics board review was not required, as this study was part of routine public health operations for quality improvement and program evaluation. Serological testing was offered to all residents and staff in both facilities, with 44% (303/691) consenting to participate (48% staff, 39% residents). A total of 303 LTC residents (n=127) and staff (n=176) were included in the study. After excluding 12 individuals with equivocal serological results, 39% (n=113) were reactive and 61% (n=178) were non-reactive. Table 1 provides a descriptive epidemiological summary of study participants. The median time between symptom onset and serological collection was 50 days (IQR= 15) for the entire cohort, 52 days (IQR= 9.5) for NAAT positive cases, and 48 days (IQR=23.5) for no or negative NAAT cases. .60) were most prominently associated with increased odds of reactive serology ( Figure 1A ). All symptom cluster case definitions were significantly associated with seropositivity ( Figure 1C ). Serological studies of COVID-19 have largely focused on cluster identification and characterization 9 , assessment of seroprevalence 10 , and patterns of seroconversion 11 . A recent study among hospitalized patients also utilized serology to identify cases with negative NAAT or asymptomatic infections 12 ; however, no studies to date have used serology to inform clinical case definitions and subsequently infection control measures in LTC facilities. Our findings support using a low threshold for symptoms in LTC settings (particularly nonrespiratory symptoms) when considering exclusion and isolation of symptomatic staff and residents. Given the non-specific nature of symptoms found to be highly predictive, such as headache, myalgia, and loss of appetite, implementation of universal contact/droplet precautions early in the outbreak may be effective in curbing transmission within facilities, rather than relying on isolating residents when they present with fever and/or respiratory symptoms. Moreover, staff and residents with several negative NAATs for COVID-19 should warrant further investigation with serology and/or be considered a clinical case if repeat NAAT testing is due to persisting symptoms. Lastly, ongoing evaluation of the Canadian, European, and WHO probable case definitions in outbreak settings is necessary, given gaps in COVID-19 diagnosis highlighted by this and other serological studies. 12 Amendment to align more closely with the US CDC definition, which was more sensitive to historical infection in this analysis, may be appropriate in LTC outbreak settings. Strengths of this study include serological testing on several platforms and utilization of multiple sources (i.e. phone interviews, medical charts and public health data) to gather reliable clinical histories immediately after the outbreak; however, the study was limited by the small sample size, preventing further regression analysis stratified by case type. Given that systematic collection of clinical histories was refined over the duration of the outbreaks, symptoms may have been underreported for some resident cases. Our findings should be generalized to other settings with caution, as the study was conducted in an outbreak setting with a high pre-test probability for COVID-19. The use of serological testing introduced some additional limitations. Baseline serological testing was not available at the start of the outbreaks and thus prior cases may not have been identified; however, both LTCF facilities represent the earliest COVID-19 outbreaks and cases in Canada, reducing the theoretical probability of prior infection to the start of the outbreak. Due to the rapid and evolving nature of the pandemic response, there is also potential risk for misclassification bias, as the clinical and diagnostic laboratory data structures used to compare and interpret serology results underwent continual quality improvement and reconciliation. While diagnostic misclassification may also occur due to the performance characteristics of COVID-19 serological assays, tests used in this evaluation were found by the performing laboratory to have specificity of 97%-99.5% and sensitivity of up to 98% at >14 days from symptoms onset. An orthogonal approach to the interpretation of test results further improved the overall specificity. Our serological survey demonstrates that generalized/non-specific symptoms and repetitive negative NAAT testing are highly associated with seropositivity. The findings of this survey can help inform case identification when managing COVID-19 outbreaks in LTCFs. Impact of COVID-19 on residents of Canada's long-term care homes -ongoing challenges and policy response. Int Long Term Care Policy Netw The important role of serology for COVID-19 control New Respiratory Viruses and the Elderly SARS-CoV-2 (COVID-19) serology: implications for clinical practice, laboratory medicine and public health Real-time tracking of self-reported symptoms to predict potential COVID-19 Clinical features of covid-19 Asymptomatic and presymptomatic SARS-COV-2 infections in residents of a long-term care skilled nursing facility Connecting clusters of COVID-19: an epidemiological and serological investigation Low SARS-CoV-2 sero-prevalence based on anonymized residual serosurvey before and after first wave measures in British Columbia Different longitudinal patterns of nucleic acid and serology testing results based on disease severity of COVID-19 patients. Emerg Microbes Infect Antibody responses to SARS-CoV-2 in patients with COVID-19 Funding Statement: No financial sources/grants to disclose.