key: cord-0908884-l18fle8r
authors: Morris, Claudia R.; Sullivan, Patrick; Mantus, Grace; Sanchez, Travis; Zlotorzynska, Maria; Hanberry, Bradley; Iyer, Srikant; Heilman, Stacy; Camacho-Gonzalez, Andres; Figueroa, Janet; Manoranjithan, Shaminy; Leake, Debra; Mendis, Reshika; Cleeton, Rebecca; Chen, Christie; Krieger, Rachel; Bush, Patricia; Hughes, Tiffany; Little, Wendalyn K.; Suthar, Mehul; Wrammert, Jens; Vos, Miriam B.
title: Prevalence of SARS-CoV-2 Antibodies in Pediatric Healthcare Workers
date: 2021-03-12
journal: Int J Infect Dis
DOI: 10.1016/j.ijid.2021.03.017
sha: 6b2902305c2175ba4572093c3277ab0fac0ec06f
doc_id: 908884
cord_uid: l18fle8r

OBJECTIVES: To determine SARS-CoV-2-antibody prevalence in pediatric healthcare workers (pHCWs). DESIGN: Baseline prevalence of anti-SARS-CoV-2-IgG was assessed in a prospective cohort study from a large pediatric healthcare facility. Prior SARS-CoV-2 testing history, potential risk factors and anxiety level about COVID-19 were determined. Prevalence difference between emergency department (ED)-based and non-ED-pHCWs was modeled controlling for those covariates. Chi-square test-for-trend was used to examine prevalence by month of enrollment. RESULTS: Most of 642 pHCWs enrolled were 31-40years, female and had no comorbidities. Half had children in their home, 49% had traveled, 42% reported an illness since January, 31% had a known COVID-19 exposure, and 8% had SARS-CoV-2 PCR testing. High COVID-19 pandemic anxiety was reported by 71%. Anti-SARS-CoV-2-IgG prevalence was 4.1%; 8.4% among ED versus 2.0% among non-ED pHCWs (p < 0.001). ED-work location and known COVID-19 exposure were independent risk factors. 31% of antibody-positive pHCWs reported no symptoms. Prevalence significantly (p < 0.001) increased from 3.0% in April-June to 12.7% in July-August. CONCLUSIONS: Anti-SARS-CoV-2-IgG prevalence was low in pHCWs but increased rapidly over time. Both working in the ED and exposure to a COVID-19-positive contact were associated with antibody-seropositivity. Ongoing universal PPE utilization is essential. These data may guide vaccination policies to protect front-line workers.

Serologic studies examining prevalence of antibodies to SARS-CoV-2 vary widely by the group Rep, 2020). Prior to these data being released, utilization of PPE in outpatient pediatric settings was inconsistent, which may have increased workplace exposure in the outpatient pediatric setting. The aim of this study was to determine the prevalence of IgG antibodies to SARS-CoV-2 in pHCWs at a large pediatric healthcare system that is comprised of three children's hospitals, and to identify what characteristics may be associated with increased prevalence. In particular, we hypothesized that EDbased pHCWs were at higher risk of infection compared to those working in other areas of the hospital.

pHCWs were enrolled into a prospective, longitudinal cohort to determine the prevalence of IgG antibodies to SARS-CoV-2 in HCWs over time at a large pediatric healthcare system in Atlanta, Georgia that cares for children and adolescents up until their 21 st birthday. There are no adult patient services within this healthcare system. Cross-sectional data presented here are from baseline visits for J o u r n a l P r e -p r o o f the cohort participants, which took place from April-August, 2020. The study received Institutional Review Board approval from Emory University and Children's Healthcare of Atlanta.

Healthcare employees and support staff of the Department of Pediatrics at Emory University School of Medicine, Children's Healthcare of Atlanta (Children's) and private pediatric practitioners or contractors who regularly work at Children's facilities were eligible for study participation. pHCWs were recruited by email, by postings in the facilities, and by personal communication. Staff and providers working in the ED were targeted with additional emails and communications to increase participation levels. A healthcare worker who self-identified as regularly working at one of the institutions above were eligible for study inclusion. For staff safety, all participants were required to be asymptomatic at the time of in person visits. If a participant was symptomatic near the time of participation, he/she was rescheduled to a later date to ensure the visit was scheduled at least 14 days post-symptoms.

At baseline, verbal or electronic informed consent was obtained and each participant completed a brief online survey that included basic demographics (age range by decade, sex, history of comorbidity), job role, primary location within the pediatric facility, number of hours worked per week, perception of SARS-CoV-2 exposure, illnesses since January 2020, travel history, documented infection with SARS-CoV-2, current symptoms, belief as to whether they had anti-SARSCoV-2 antibodies, and level of interest in participating in donating blood for convalescent serum if positive (yes/no). pHCW participants were also asked their level of J o u r n a l P r e -p r o o f anxiety experienced due to the pandemic on a Likert scale of 1-5. All participants provided up to 30 ml of blood through venipuncture; blood specimens were processed within 1 hour, and serum and plasma were stored by the Children's Healthcare of Atlanta and Emory University's Children's Clinical and Translational Discovery Core for future analysis.

Venous blood for plasma and serum was collected, processed and stored at -20°C until analysis.

Qualitative positivity and quantitative SARS-CoV-2 antibodies were determined by measuring the IgG antibody responses to the receptor binding domain (RBD) of the spike (S) protein using an enzymelinked immunosorbent assay (ELISA) as previously described (Suthar et al., 2020) .

The dependent measure was prevalence of past SARS-CoV-2 infection as reflected by IgG antibody seropositivity. The primary job location was used to define the main independent measure as pediatric ED-based HCWs (ED or urgent care) or non-ED based pHCWs (pediatric intensive care [PICU], general wards, specialty wards, clinical and administrative services, and operating room). Other independent measures for the analyses included month of enrollment, HCW role, average number of clinical hours worked per week in 2020, potential risks for SARS-CoV-2 infection (children in the home, domestic and international travel), prior illness or known COVID-19 exposures since January 2020, and receipt of SARS-CoV-2 PCR testing. HCWs from urgent care settings were considered to be ED-based pHCWs. Pediatric ED-based HCWs, including respiratory therapists, worked primarily in the ED, and did not work in multiple units. The level of anxiety related to COVID-19 was quantified by dichotomizing the responses to the question "The COVID-19 pandemic has increased my anxiety." J o u r n a l P r e -p r o o f Scores of 4 and 5 were defined as "high anxiety", while scores of 1 and 2 were considered "low anxiety."

In this large single-healthcare system, policies around response to the SARS-CoV-2 pandemic evolved over time. The first case of COVID-19 reported in Georgia was diagnosed on March 2, 2020.

COVID-19 cohorting units at Children's for patients under investigation (PUI) with respiratory symptoms and those with confirmed COVID-19 were established during the first few months of the pandemic between March 9-June 10, 2020. These were all private rooms. Staff followed Enhanced Contact Droplet precautions that included: procedure mask (N-95 mask if aerosol-generating procedure; once more N-95s available, staff could use as their default mask for all care of these patients), eye protection, gloves, and gown. Since turn-around time for COVID-19 testing in March/April was over 72 hours, the COVID-19 status was often unknown for days. The PICU did not cohort PUI, however patients with respiratory symptoms were placed in isolation, and staff utilized Enhanced Contract Droplets precautions. Within the pediatric EDs at Children's, a universal PPE policy was implemented that included use of surgical masks and goggles for all patient-facing activities on March 21, 2020, with N95 masks utilized for traumas, resuscitations and intubations. Prior to this time, masks were not routinely worn in the pediatric ED or at triage unless a patient was placed on respiratory isolation.

Patients seen in the pediatric EDs or admitted to the hospital were not routinely screened for COVID-19; testing was performed when the outcome could potentially influence clinical decision-making, and rapid testing was not available.

J o u r n a l P r e -p r o o f Prevalence of antibodies for SARS-CoV-2 was assessed using summary statistics, and reported overall and by HCW roles, and for pediatric ED-based vs. non-ED-based HCWs. Univariate tests for differences in proportions between groups were conducted using chi-square tests. Associations between participant characteristics and SARS-CoV-2 seropositivity were assessed by calculating prevalence ratios (PR) and 95% confidence intervals (CI) from modified Poisson regression models (Barros and Hirakata, 2003) . Variables that were significant in bivariate analyses were included in the final multivariable model, and adjusted PR (aPR) and 95% CI were calculated for each exposure.

Analyses were carried out in SAS v.9.4 (Cary, NC) and CRAN R v.3.3 (Vienna, Austria), and statistical significance was evaluated at the 0.05 threshold.

A total of 642 pHCWs (202 ED, 440 non-ED) were enrolled, the majority of whom were 31-40 years of age, female, and physicians or nurses. Overall and stratum-specific prevalence of antibodies are summarized in Table 1 . Thirty-one percent of participants were pediatric ED-based, 42% reported a viral illness between January 2020 and date of participation and 8% underwent SARS-CoV-2 PCR testing during an illness. The majority of pHCWs (84%) believed they were at high risk of COVID-19 due to their occupation in healthcare, 99% of the cohort believed it was helpful to know their SARS-CoV-2 IgG status, and 91% would be interested in donating convalescent plasma if seropositive. In response to the survey question "The COVID-19 pandemic has increased my anxiety", 71% reported high anxiety.

The prevalence of SARS-CoV-2 seropositivity was 26 out of 642 participants (4.1%, Figure 1 ). Most of the IgG positive pHCW were physicians and nurses (73%). Only 31% of the entire cohort reported an exposure to a known COVID-19 contact. The majority of IgG-positive HCWs were pediatric EDbased with a prevalence of 8.4% vs. 2.0% in ED vs. non-ED pHCWs (p<0.001). No PICU-based HCWs tested positive for SARS-CoV-2 antibodies. One out of 36 (2.6%) respiratory therapist in our cohort was seropositive.

Thirty-five percent of IgG-positive pHCWs were tested for SARS-CoV-2 by PCR during an acute illness; 8/9 were PCR-positive. Seropositive pHCWs were more likely to report a history of viral illness since January 2020 compared to those who were IgG-negative (69% vs. 41%, p=0.004), and had a higher exposure to a known COVID-19 positive case (58% vs. 30%, p=0.004). There was no difference in travel history or presence of children in the household. Of the 26 IgG-positive participants, 31% reported no viral illness or symptoms since January 2020, suggesting asymptomatic COVID-19 disease. No seropositive HCW reported hospitalization, but 19% sought medical care for the acute illness, 31% described their illness as moderate and impacting daily activities in addition to staying home from work, while 2 (8%) reported severe symptoms including being bedbound for at least 1 day and/or requiring an ED visit. There was no difference in self-reported anxiety level between pHCWs in the ED vs. those outside the pediatric ED, nor was there a difference in those who were seropositive vs. negative.

Seropositivity by month of pHCW participant enrollment between April-August 2020 is illustrated in Figure 2A . Overall prevalence by month of enrollment decreased from 4.8% among those enrolled in April to 1.3% among those enrolled in June. Seropositivity among the pediatric ED participants J o u r n a l P r e -p r o o f remained stable in those enrolled in the months April through July, while an increase in IgG positivity was detected in participants enrolled in August regardless of job location. Reported COVID-19 cases in Georgia between April-August 2020 are illustrated in Figure 2B in order to demonstrate activity in the community during this same timeframe (CDC Tracker, accessed January 31, 2021).

In this prospective study of SARS-CoV-2 IgG antibodies in pHCWs, the overall prevalence of positivity was 4.1%. Factors associated with seropositive status included working in the pediatric ED, pediatric ED-based HCWs so early in the pandemic also suggests that the virus was likely circulating for some time before the pandemic risk was appreciated, during which stage pediatric ED-based HCWs may have been exposed to infectious children with unrecognized COVID-19, or their infectious parent, family member or caretaker. Use of universal PPE was not standard practice in our pediatric EDs before the Georgia statewide shelter-in-place order took effect on April 3, 2020. High volumes of children with mild symptoms are commonly seen in pediatric facilities. Low acuity "fast track" patients make up 40-60% patient volume across the three Children's EDs included in this analysis and represent pediatric facilities, that combined, evaluate over 240,000 ill or injured children annually.

With over a 4-fold increased risk of seropositivity compared to their non-ED based colleagues, these data show that pediatric ED-based HCWs are at increased risk of viral infection with SARS-CoV-2 compared to other pHCWs. These data add to growing evidence that children likely spread the novel coronavirus, and HCWs in the ED may be at higher risk for exposure and infection. These data also have implications for frontline pHCWs during future epidemics regarding workplace safety. PPE may contribute to the variation in HCW positivity seen across various studies. Chen and colleagues from Nanjing, China reported a high IgG antibody positivity rate of 17% in 105 HCWs known to be exposed to patients with COVID-19. Not all HCWs utilized PPE in this study. The authors identified a lower risk of seroconversion among HCWs that was closely related to wearing a facemask (Chen et al., 2020). In our study, the upward trend in seropositivity noted in July among non-ED pHCWs and in August among all pHCWs in our study likely reflects escalation in community spread as local businesses began to re-open and social activities increased, correlating with an increase in acute cases reported in Georgia following the 4 th of July weekend (CDC Tracker, accessed December 10, 2020) We hypothesize that exposure risk in the workplace early in the pandemic J o u r n a l P r e -p r o o f mitigated by universal PPE utilization transitioned to greater risk of exposure in the community as the prevalence of COVID-19 rose briskly in Atlanta (CDC Tracker, accessed February 27, 2021).

By professional classification, physicians were found to be at highest risk, however this association disappeared in multivariate analysis, possibly driven by the higher number of physicians in the pediatric ED-based participant group. Respiratory therapy could be considered a high-risk profession given the infectious exposures resulting from aerosolizing procedures. A report of high infection rates among respiratory nurses in the United Kingdom supports this notion (Bird et al., 2020) . However, only 1/36 (2.6%) respiratory therapist in our cohort was seropositive, similar to the prevalence among non-ED based pHCWs. In contrast to pediatric ED providers at our facilities, most respiratory therapists at our facilities typically wore PPE as standard practice when working with children in the hospital even before the pandemic, which may have provided increased protection (Ferioli et al., 2020, Sommerstein et al., 2020). It is interesting to note that no staff from the PICU, another closed unit with potentially high COVID-19 exposure risk, were found to be seropositive even though they made up 10% of our pHCW cohort. There are several possible explanations for the difference including differences in typical pre-pandemic PPE utilization and differences in patient exposure levels between inpatient and outpatient settings. Although recent studies suggest that children are just as likely as adults to become infected with the virus, they have fewer symptoms and less severe disease (Zimmermann and Curtis, 2020), so they may present to acute care settings with COVID-19 symptoms, while only a small portion will require hospitalization (Sinha et al., 2020, Zimmermann and Curtis, 2020). COVID-19 testing of pediatric patients at our facilities was limited at the time of this study, so most HCWs caring for a child infected with SARS-CoV-2 were unaware of their diagnosis.

Working less than 35 hours per week was another univariate risk factor that fell out of the multivariate model, leaving only ED-based work location and known exposure to COVID-19 as significant risk factors. The association with working fewer hours was likely driven by the ED-based physician seropositivity as this group commonly works less than 30 clinical hours per week. Travel and presence of children in the household were not predictors of seropositivity, although these factors are often assumed to be risks.

Testing for SARS-CoV-2 by PCR during acute illness was infrequent among our participants, with only 8% of the cohort tested for SARS-CoV-2, despite 42% reporting viral symptoms and flu-like illness since January 2020. Even among those who were identified as seropositive through this surveillance study, only about a third of staff were PCR tested during an acute illness, with 8 out of 9 tested found to be PCR-positive. Interestingly, 31% of seropositive HCWs reported no history of illness in 2020, representing an asymptomatic group similar to what has been documented in the literature (Oran and Topol, 2020).

HCWs who are exposed to SARS-CoV-2 in the workplace may be concerned about inadvertently bring the virus home to their families, adversely impacting household members. This could potentially generate fear and anxiety among HCWs not typically experienced previously in the US, although the psychosocial impact of past epidemic/pandemic outbreaks on HCWs is well described in the literature (Preti et al., 2020) . Our study confirms this suspicion, with 71% of our pHCW cohort reporting high anxiety during the current pandemic. Anxiety was similarly high between ED-based and non-ED HCWs, although 94% of pediatric ED-based HCWs reported a belief that they were at high risk of COVID-19 compared to 80% of HCWs located outside the ED. In a large cohort of over 1200 HCWs from 34 hospitals J o u r n a l P r e -p r o o f in China, 72% of HCWs reported distress and 34-50% reported symptoms manifesting as stress, anxiety, insomnia and/or depression. (Lai et al., 2020) Evidence is emerging that the psychological toll of the COVID-19 pandemic is being felt by HCWs worldwide (Chew et al., 2020) . HCWs are known for their stamina and emotional resilience in the workplace, however COVID-19 has changed the playing field.

Pressures of the pandemic may add to the long-recognized problem of physician burnout (Yates, 2020) , and emergency medicine physicians in particular are already at high-risk (Lim et al 2020) . Additional resources may be needed to address this growing issue among HCWs in order to preserve mental health and resilience, particularly among front-line workers (Santarone et al., 2020) .

This study has several limitations. First, this was a convenience sample, and the antibody prevalence of pHCWs who chose to participate may be systematically different from those who chose not to participate. Second, this study was not designed to identify the source of infection for the pHCWs.

Adult family members, parents and/or caregivers of the children seeking care were also a potential source of infection. Additionally, it is possible that pHCWs were exposed to SARS-CoV-2 by household contacts, through domestic or international travel, or infected via community spread rather than work exposure. However, neither children in the home nor travel between January-March 2020

were associated with seropositivity. The significant difference in seropositivity between ED-based pHCWs vs. those outside the ED supports the hypothesis that spread by children evaluated in the ED occurred early in the pandemic. Alternatively, it is possible that the high prevalence of SARS-CoV-2

IgG antibodies in ED-based pHCWs is because of spread among staff working closely together in an environment with a limited ability to socially distance within the ED (Cave, 2020). However, the consistently high prevalence in ED-based pHCWs across three separate pediatric hospitals within the Children's network makes this less likely. Given this study takes place in a region that acquired high SARS-CoV-2 activity, the conclusions may not be generalizable to all healthcare settings, particularly J o u r n a l P r e -p r o o f in communities with lower contagion prevalence. Also, although pHCWs were recruited from a large pediatric healthcare system, a small percentage of pHCWs in the network was tested, which could lead to selection bias and an overestimate or underestimate of the true prevalence. Finally, false positive or false negative SARS-CoV-2 IgG antibody test results could impact the integrity of the data, however the method used in this study has demonstrated a > 96% sensitivity and specificity (Suthar et al., 2020) .

The study has several strengths. This is a large, prospective cohort and represents many departments and position types among the pHCWs in our institutions. To our knowledge, it is the largest report on pediatric-specific HCWs to date. The participant enrollment began near the beginning of the epidemic, thus allowing us to characterize both the early seropositivity prevalence and the change over time as the level of COVID-19 has fluctuated within our region. These data also highlight increased risk to pediatric frontline workers, and the psychological toll of this pandemic on HCWs that is likely not unique to our institution.

In conclusion, we found that the seropositivity prevalence for antibodies to SARS-CoV-2 in pHCWs was higher compared to reported levels in our community early in the pandemic, while the percent of positive participants increased in those pHCWs enrolled subsequent to a time of increased community spread (July 2020,CDC Tracker, accessed January 31, 2021). Both known exposure to COVID-19 and working in the pediatric ED were independently associated with seropositivity, while travel and having 

We declare no conflicts of interests. Claudia R. Morris, MD, is the inventor or co-inventor of several UCSF-Benioff Children's Hospital Oakland patents/patent-pending applications that include nutritional supplements, and biomarkers of cardiovascular disease related to arginine bioavailability, is an inventor of several Emory University School of Medicine patent applications, and is a consultant for 

The study received Institutional Review Board approval from Emory University and Children's Healthcare of Atlanta; verbal or electronic informed consent was obtained from all participants.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Table 2 . Prevalence and correlates of anti-SARS-CoV-2-IgG among pediatric healthcare workers enrolled in a prospective cohort study, Atlanta, Georgia, April-August 2020 

Alternatives for logistic regression in cross-sectional studies: an empirical comparison of models that directly estimate the prevalence ratio

MMWR Morb Mortal Wkly Rep 2020

High SARS-CoV-2 infection rates in respiratory staff nurses and correlation of COVID-19 symptom patterns with PCR positivity and relative viral loads

SARS-CoV-2 antibody prevalence in health care workers: Preliminary report of a single center study

COVID-19 Super-spreaders: Definitional Quandaries and Implications

CDC Morb Mortal Wkly Rep, Coronavirus Disease 2019 in Children -United States

Cases & Deaths among Healthcare Personnel

Trends in Number of COVID-19 Cases and Deaths in the US Reported to CDC, by State/Territory

A multinational, multicentre study on the psychological outcomes and associated physical symptoms amongst healthcare workers during COVID-19 outbreak

Protecting healthcare workers from SARS-CoV-2 infection: practical indications. European respiratory review : an official journal of the

Pediatric emergency department responses to COVID-19: Transitioning from surge preparation to regional support. Disaster Medicine and Public Health Preparedness

Care for the psychological status of frontline medical staff fighting against COVID-19

Seroprevalence of SARS-CoV-2 antibodies among physicians from a children's hospital

Prevalence of SARS-CoV-2 Infection Among Health Care Workers in a Tertiary Community Hospital

SARS-CoV-2-specific antibody detection in healthcare workers in Germany with direct contact to COVID-19 patients

Factors Associated With Mental Health Outcomes Among Health Care Workers Exposed to Coronavirus Disease

Epidemiological and Clinical Characteristics of COVID-19 in Children: A Systematic Review and Meta-Analysis

Emergency medicine physician burnout and wellness in Canada before COVID19: A national survey

Prevalence of Asymptomatic SARS-CoV-2 Infection : A Narrative Review

SARS-CoV-2 infection among hospital workers of one of the most severely affected institutions in Madrid, Spain: a surveillance cross-sectional study

The Psychological Impact of Epidemic and Pandemic Outbreaks on Healthcare Workers: Rapid Review of the Evidence

Preserving mental health and resilience in frontline healthcare workers during COVID-19

COVID-19 infection in children

Risk of SARS-CoV-2 transmission by aerosols, the rational use of masks, and protection of healthcare workers from COVID-19

SARS-CoV-2 specific serological pattern in healthcare workers of an Italian COVID-19 forefront hospital

Rapid Generation of Neutralizing Antibody Responses in COVID-19 Patients

Prevalence of Healthcare and Hospital Worker SARS-CoV-2 IgG Antibody in a Pediatric Hospital

Physician stress and burnout

COVID-19 in Children, Pregnancy and Neonates: A Review of Epidemiologic and Clinical Features

The authors acknowledge the clinical staff, nurses and physicians working at the Emory University and Children's Healthcare of Atlanta facilities and in the Emory Children's Center Research Unit, who contributed to the success of this study. The authors also recognize the significant contributions of the participants, without whom this study would not have been possible. This study was funded in part by