SARS-COV-2 Seroprevalence among Health Care Workers in Private and Public Hospitals in Tehran, Iran

Background Estimating the prevalence of SARS-COV-2 antibody seropositivity among health care workers (HCWs) is crucial. In this study the seroprevalence of anti-SARS-COV-2 antibodies among HCWs of ve hospitals of Tehran-Iran with high COVID-9 patient’s referrals was assessed. Methods HCWs from public and private hospitals were included and were asked questions on their demographic characteristics, medical history, hospital role and usage of personal protective equipment (PPE). Seroprevalence was estimated on the basis of ELISA test results (IgG and IgM antibodies in blood samples) and adjusted for test performance.


Introduction
As the severe acute respiratory syndrome coronavirus 2 (SARS-COV-2) reached a pandemic level, with more than 132 million cases globally by April 2021, the risk of virus transmission among health-care workers (HCWs) with close contact to patients with coronavirus disease (COVID-19) has increased (1, 2).
Hence, to reduce the risk of virus transmission and to assess safety precautions in hospitals, estimating the prevalence of antibody seropositivity among HCWs is crucial (3). Nevertheless, factors such as surge in COVID-19-reated hospital admissions and limited access to diagnosis test could partly restrict the efforts in conducting seroprevalence surveys among HCWs (4,5).
Overall, a higher prevalence of SARS-COV-2 antibodies among HCWs compared to non-HCWs were reported in studies conducted in different countries, including the USA and Sweden (2,4,6). Iran was among the rst countries that reported widespread outbreaks of SARS-COV-2 in several provinces (7). In a recently conducted study in 18 cities across 17 provinces in Iran, the difference in prevalence of antibody seropositivity among front-line and non-front-line HCWs was low (21.6% vs.18.0%) (7). Similarly, in a study conducted among staff of Mo d children's hospital in Tehran, Iran, no difference was observed in risk of seropositivity among HCWs versus individuals working in administrative departments (8). The similar seroprevalence estimates among HCWs versus non-HCWs in these studies could partly be due to lack of compliance to safety protocols and/or limited access to personal protective equipment (PPE) in some hospitals (8,9).
Although these studies provide some insight into the prevalence of SARS-COV-2 antibodies among HCWs in Iran, their data did not include information on the potential variation in antibody seropositivity by HCW's hospital role, type of hospital (e.g., public vs. private), and hospital departments. Furthermore, although use of PPE for reducing the risk of transmission in departments treating COVID-19 patients has been recommended, it has been shown that by April 2020, 1,710 COVID-19 infections and 116 deaths among HCWs in Iran were related to insu cient access to PPE (9,10). Hence, additional data on PPE protection against prevalence of SARS-COV-2 antibodies among HCWs in hospitals is required for worksafety policy decision-making.
To address the current knowledge gaps, in this study we assessed the seroprevalence of anti-SARS-COV-2 antibodies among HCWs in the ve hospitals of Tehran-Iran that had the most COVID-9 patients in the rst few months of the pandemic.

Study design and population
In this cross-sectional study, we used serological testing to assess the prevalence of SARS-CoV-2 antibodies among HCWs, in ve hospitals in Tehran, Iran. We included both government-based (i.e., public) teaching hospitals as well as private hospitals. The three public hospitals were selected as they had major COVID-19 patient's referral in Tehran. Two private hospitals were randomly selected from the listed private hospitals in Tehran.
Compared to the public hospitals, private hospitals are smaller and have lower number of HCWs. Hence, all HCWs who were working within two assigned days in the selected private hospitals were invited to participate in our study and samples from those who agreed to participate were collected in public hospitals, due to the higher number of HCWs, more time was required for sample collection (i.e., total duration of data collection from public hospitals: three weeks and 3 days).

Test characteristics
Detailed information on test characteristics were reported previously (7). In summary, Pishtaz Teb SARS- usage, categorized as mask only or mask and other equipment including gowns, shield and/or goggles, HCW's hospital role, and working department. Furthermore, data on COVID-19-related symptoms were collected and participants were categorized based on the experienced number of COVID-19-related symptoms into asymptomatic, paucisymptomatic (1-3 symptoms), or symptomatic (≥ 4 symptoms). History of the COVID-19-related symptoms such as anosmia, sore throat, headache, shortness of breath, diarrhoea, conjunctivitis, weakness, myalgia, arthralgia, altered level of consciousness, and chest pain in the preceding 12 weeks were also requested. (7).
A laboratory technician then collected 5 mL of venous blood into an EDTA-coated microtainer which later were couriered to DDRI laboratory and stored at stored at -80°C (7). Detailed information on sample collection and ELISA kits has already been published elsewhere (7).

Statistical analysis
Baseline characteristics of participants among seronegative vs. IgG or IgM positive individuals were reported. Two-sided x 2 test was used to compare PPE usage among hospital types, HCW's hospital roles, and hospital departments, and to compare the antibody seropositivity among different roles. To assess the seroprevalence of SARS-CoV-2-speci c antibodies among HCWs, the overall crude frequencies of positive tests and test performance adjusted estimates, strati ed by age categories, sex, BMI, presence of comorbidity, HCWs job type, department of work, hospital type, diagnosed COVID-19, contact with infected patients, and symptom categories were estimated. The 95% con dence intervals (CIs) for crude seroprevalence were estimated using exact binomial models and a bootstrap method was used to construct the 95% CIs for the adjusted estimates (7, 11). All statistical analyses were conducted by STATA software, version 12. The statistical approach used for the test performance adjustment and the bootstrap method is provided in detail elsewhere (7).

Results
In total among the 2065 participants, 1825 (88.4%) and 240 (11.6%) HCWs were recruited from the public and private hospitals, respectively (Table 1)

Discussion
In this cross-sectional study among HCWs, the frequency of PPE usage as well as seroprevalence of SARS-COV-2 varied considerably by hospital type, hospital department, and HCW's hospital role. Overall, the highest prevalence of seropositivity was observed in the private hospitals, COVID-19 patient's ward department, and individuals aged 60 years and older. Furthermore, concurrent use of mask and any other type of PPE was signi cantly higher among HCWs of public hospitals, ICU or surgery ward, and nurses.
Finally, 17.6% (60/340) of participants who had positive test results for SARS-COV-2 antibodies did not report experiencing any symptoms.
In general, our overall test-adjusted SARS-COV-2 seroprevalence estimate of 22.5% among HCWs in Private and Public Hospitals was similar to the reported seroprevalence estimates in conducted crosssectional studies of HCWs in the UK (24.4%) (Shields, 2020) and New York City (27.0%) (2,12). Consistent with other studies, we also observed variation in SARS-COV-2 seropositivity by HCWs hospital role and department (4,12). The highest seroprevalence was observed among nurses and assistant nurses, which could be due to closer contact with COVID-19 infected patients (4). In contrast, antibody positivity in our study was lower among Janitor/building superintendents, who compared to nurses, may follow different hospital policies with respect to SARS-COV-2 safety precautions (4,13). For instance, according to the World Health Organization (WHO) rational use of PPE, the type of PPE that should be used among hospital cleaners who enter the room of patients with COVID-19 partly differs from what HCWs should use (e.g., using Heavy-duty gloves) (14).
Similarly, several studies reported the same ndings, indicating the potential virus transmission among HCWs within hospital departments (12,15,16). In a study conducted in multistate hospital network in the USA, 29% of participants with detected antibodies, reported no symptoms of COVID-19 (16). These ndings highlight the potential "subclinical nature" of COVID-19 disease spectrum and the importance of testing HCWs regularly to prevent the virus spread within the hospital environment (15)(16)(17). In addition to screening HCWs for SARS-COV-2 seropositivity, assessing viral load among asymptomatic and symptomatic participants could provide some information about the viral transmission and pathogenesis of SARS-COV-2 in hospital setting (2).
In this study we found that overall, the combined usage of mask and any other type of PPE among HCWs of public hospitals was signi cantly higher than private hospitals. As a result, the seroprevalence of  (18)(19)(20). Hence, su cient access to PPE as well as assigning HCWs with comorbid conditions to hospital wards with lower risk of infection, may need to be considered as the potential strategies to reduce the risk of infection and mortality among HCWs (3).
To the best of our knowledge, this is the rst seroprevalence study in Iran that reports the SARS-COV-2 seropositivity among HCWs by hospital types, department, and participant's role. Additionally, the prevalence of seropositive tests was strati ed according to HCW's baseline characteristics and PPE usage. Despite the strengths, this study had some limitations that should be considered. First, since patients were recruited during short period of time in each hospital, the study participants may not be representative of all HCWs working in each centre. Second, data on baseline characteristics, presence of COVID-19 symptoms, and contact with infected patients were collected using a self-reported questionnaire, which may introduce recall and/or misclassi cation bias in the study. Finally, among seropositive HCWs, it was not possible to differentiate between the community acquired or hospital transmitted infections. Hence, the potential routs of SARS-COV-2 transmission among HCWs remains unknown.
In conclusion, the ndings of this study imply that seroprevalence of SARS-COV-2 among HCWs depends on hospital type, hospital department, and hospital role. The PPE usage, as a main strategy in infection prevention, was suboptimal, especially among HCWs in private hospitals. HCWs with close contact to COVID-19 patients and with comorbidity conditions are at higher risk of infection and continued effort in access to adequate PPE, regular screening of hospital staff for detecting asymptomatic personnel, to