Seroprevalence of SARS-CoV-2 in Emergency Department Healthcare Workers at Sírio-Libanês Hospital,Brazil

Abstract

COVID-19 has spread rapidly worldwide. Information on its prevalence and factors associated with infection are important for protecting both professionals and patients in healthcare centers. This study evaluated the seroprevalence of antibodies against SARS-CoV-2 and its association with the degree of exposure and use of personal protective equipment by healthcare professionals dedicated to the treatment of patients with flu-like illnesses in the emergency room. The research team included an analysis of healthcare professionals who underwent enzyme-linked immunosorbent assay serological testing for SARS-CoV-2 between May 28 and June 26, 2020, in the emergency room of Sírio-Libanês Hospital in São Paulo, Brazil. Participants answered individual questionnaires on occupational information, medical health history, and factors associated with exposure to the novel coronavirus. The questionnaire variables were compared based on the serological results. Of the 164 study participants, 96 (58.54%) reported at least 1 flu-like symptom and 42 (25.61%) presented serology results that were compatible with SARS-CoV-2 infection. The asymptomatic declared group accounted for 62 participants; of these, 8 (12.90%) had positive serology results (neutralizing antibody and IgG) for SARS-CoV-2. Data analysis showed a positive correlation with duration of work, safety in wearing and reusing personal protective equipment, and presence of anosmia, and showed a negative relationship with duration of mask use. Our findings suggest that the perception of symptoms by healthcare professionals is not a good screening parameter for the diagnosis of an infectious disease with respiratory symptoms, such as COVID-19. The main influencing factor for the control of infection is the elaboration of workflows and safety protocols based on simple and clear rules as well as investments in team training.

Introduction

SARS-CoV-2 infection has spread rapidly worldwide, and the resulting COVID-19 pandemic has saturated health systems and caused shortages in hospital equipment and supplies.¹ In Brazil, after the first case of COVID-19 was diagnosed on February 25, 2020, health services were reorganized to contain the spread of the virus and to optimize management of the pandemic.

Of the strategies used to manage COVID-19, prevention of infections in healthcare professionals is essential for maintaining the integrity of healthcare systems and the quality of care. Between January 2020 and May 2021, approximately 115,000 healthcare professionals died due to COVID-19 complications worldwide.² However, the actual count is unknown because of underreporting and poor surveillance systems.³ In China, by February 25, 2020, approximately 3,387 healthcare professionals had contracted COVID-19, with at least 18 confirmed deaths due to the disease.⁴ These numbers surpass those of other epidemics, such as SARS in 2003, wherein 966 professionals working on the frontlines of health services were infected in China.⁵ Similarly, countries that faced the novel influenza A (H1N1) pandemic in 2009 reported a considerable number of cases among their health workers. During the H1N1 crisis, seroprevalence studies in Asia and Australia showed that 7.0% to 30.8% of professionals dealing with this disease were infected with the virus.^6-8 Of those infected, approximately 50% were asymptomatic but had the possibility of transmitting the disease.⁹ It is therefore important to emphasize that the infection rates in this group of professionals varied according to their training level in using personal protective equipment (PPE), the supply and availability of PPE, and organization of health services.^10,11

Similar to other epidemics, numerous studies have shown that asymptomatic SARS-CoV-2 carriers or infected individuals during the incubation period have an important influence on the spread of the virus.^12-14 The health team represents a considerable part of the group of asymptomatic transmitters, but this issue has scarcely been studied.¹⁵ Healthcare professionals are at higher risk of SARS-CoV-2 infection because of their substantial exposure to symptomatic and asymptomatic patients; however, data on this topic are still lacking.¹⁶ For example, in March 2020 during the first peak of the epidemic in Italy, 20% of notified COVID-19 cases were estimated to be healthcare professionals.¹⁷ In Spain, this number was even higher at 24.1% of all cases in May 2020.¹⁸ A multicenter study on seroprevalence was conducted from April to June 2020 to understand the involvement of this specific group, and showed that 6% of healthcare professionals in the United States had positive serology (ie, IgM, IgG, or IgA) for SARS-CoV-2, with one-third of this population being asymptomatic.¹⁹ In Brazil from March to May 2020, a study of healthcare professionals with flu-like symptoms at a tertiary center in São Paulo showed that 42.37% were COVID-19 positive.²⁰

In addition, the availability of diagnostic equipment for COVID-19 is unstable in Brazil. A study conducted in April 2020 on hospital services in Brazil and Latin America showed a lack of diagnostic resources of up to 25% in private services in the first months of the pandemic.²¹ During the sample collection period for our study, serological testing for detecting neutralizing antibodies was not available for routine use and was restricted to scientific research. Furthermore, serology for IgM and IgG antibody detection has recently been developed and is still outside the capabilities of most healthcare providers. This situation has led to a lack of knowledge regarding the serological status of healthcare professionals.

After the first few months of the pandemic, there was a perception in our health service that uninfected professionals began to believe that they had been previously infected and that they were already immune to the disease. Some of the scientific data at that time suggested that once one had been infected, antibody and T cell responses would offer a temporary protection against new infection for variants that were present at that moment.²² This is understandable because of reports of asymptomatic cases, the high exposure of healthcare professionals, and the low availability of serology tests at the time. This kind of thinking also generated a dangerous precedent for discontinuing PPE use.²³

In view of the COVID-19 pandemic, information on the prevalence of SARS-CoV-2 and factors associated with infection are important for protecting both professionals and patients in health services. Although the use of PPE and measures to isolate infected patients reduces infections caused by the novel coronavirus, the real effects of these measures are still unknown.^21,24-26 Thus, we evaluated the seroprevalence of antibodies against SARS-CoV-2 and its association with the degree of exposure and use of PPE by healthcare professionals dedicated to the treatment of patients with flu-like illnesses in the emergency room of Sírio-Libanês Hospital in São Paulo, Brazil.

Methods

We conducted a mixed-methods study with healthcare professionals working in the emergency department of the Sírio-Libanês Hospital, a philanthropic tertiary hospital in São Paulo, Brazil.

Participant Selection

Medical professionals, nurses, and nursing technicians who had been caring for patients with flu-like illnesses since February 25, 2020 and were working in the emergency room of Sírio-Libanês Hospital were asked to participate. Participants were included only after they received a clear explanation of the proposal and provided signed informed consent forms. For the second stage, those who did not have previous diagnostic confirmation of COVID-19 by reverse transcription polymerase chain reaction (RT-PCR) were evaluated for possible behavioral changes associated with previous infection.

Sample Collection

Participants underwent peripheral blood collection via venipuncture between May 28 and June 26, 2020. After processing, the samples were frozen at -80 °C for subsequent analysis. Serological tests were performed to search for IgA, IgM, and IgG antibodies against SARS-CoV-2 using enzyme-linked immunosorbent assay (ELISA), in addition to the dosage of neutralizing antibodies in the samples.²⁷

Questionnaire

At the time of sample collection, participants each completed a questionnaire. Data collection was performed anonymously to ensure participants' privacy. The questionnaire consisted of occupational information, medical health history, and factors associated with community and occupational exposure to the novel coronavirus. The frequency of handwashing, apron use, and mask use was reported as “most of the time” or “all the time.” Participants were questioned about the type of mask worn, with the option to choose “N95 only” or “surgical or N95.” The use of eye protection was evaluated as goggles plus face shields, face shields only, or goggles only. All participants underwent a PPE training program based on World Health Organization recommendations²⁸ and were monitored by a supervisor to review the adequacy of the procedures. Participants were evaluated based on their confidence in putting on and removing PPE, according to the previous training program. Participants without a previous diagnosis of COVID-19 were asked if they believed they had previously been infected with SARS-CoV-2.

Data Analysis

Quantitative data were expressed as mean and standard deviations, and qualitative data were expressed as counts and frequencies (percentages). The software used to conduct our analysis was R version 4.0.1 (R Foundation for Statistical Computing, Vienna, Austria).

The variables were compared according to the neutralizing antibody result (cytopathic effect-based virus neutralization test [CPE-VNT]), which can be positive or negative if the reference value was less than 20 or at least 20, which is the highest dilution where viral neutralization can be observed (ie, absence of cytopathic effect [VNT100]). Spearman's rank correlation coefficient was used to assess the correlation between the level of neutralizing antibodies and the level of IgG. The Shapiro–Wilk test was used to verify the normality of the sample. Student's t test was used for parametric data, and the Wilcoxon signed-rank test was used for nonparametric data. Categorical variables were evaluated using the chi-square test. Logistic regression was used for statistically significant variables and additional analyses were performed for the significant variables.

Additional analysis for the variables was performed according to the titer result (last sample dilution that still showed visible reactivity), which was expressed as an ordinal variable. Categorical variables were evaluated using the chi-square test for linearity by assigning equally spaced weights. Continuous variables were evaluated using the Jonckheere–Terpstra trend test for statistical analysis. Statistical significance was set at P < .05.

The data analyzed in this study are available on reasonable request from the first author (ricardo.mmadureira@hsl.org.br).

Ethics Approval

The study was approved by the community leaders and the National Research Ethics Committee (Comissão Nacional de Ética em Pesquisa; CAAE:32033520.6.0000.5461), in accordance with the Declaration of Helsinki.

Results

Among the 176 professionals who met the inclusion criteria, 164 agreed to participate in the study. The median age of participants was 37 years (interquartile range: 32 to 41 years); 98 participants (60%) were female, 87 (53%) were nursing technicians, 41 (25%) were nurses, and 36 (22%) were physicians. Pathologies recognized as risk factors for COVID-19 were not reported by 85% of the participants (Table 1).

Table 1.

Key Characteristics of Study Participants (N = 164)

Characteristics	Values
Age, years (95% CI)	37.1 (32 to 41)
Sex
Female	98 (59.8%)
Profession
Nursing technician	87 (53.0%)
Nurse	41 (25.0%)
Physician	36 (22.0%)
Comorbidities that are considered risk factors for COVID-19
Arterial hypertension	10 (6.1%)
Diabetes mellitus	1 (0.6%)
Cardiovascular disease	2 (1.2%)
Asthma/COPD	10 (6.1%)
Immunosuppression	1 (0.6%)
Oncological disease	1 (0.6%)
Previous positive RT-PCR result for SARS-CoV-2
Yes	32 (19.5%)

Abbreviations: COPD, chronic obstructive pulmonary disease; RT-PCR, reverse transcription polymerase chain reaction.

The presence of flu-like symptoms (eg, cough, fatigue, sore throat, fever, runny nose, breathlessness), diarrhea, and changes in the sense of taste and smell were evaluated using data obtained from the questionnaire. Out of the 164 participants, 3 did not complete the flu-like symptoms form, which included previous infection reporting, and were excluded from the flu-like symptoms analysis. At least 1 symptom was reported by 96 of the participants (59.63%) since the beginning of the pandemic (Table 2). Among the 62 participants who declared themselves asymptomatic, 8 had positive serology results (neutralizing antibody and IgG) for SARS-CoV-2, which represents 12.90% of all COVID-19 diagnoses in this group.

Table 2.

Participant Symptoms and Positive RT-PCR and Serology Results (N = 161)^a

Symptoms	Total n (%)	Positive RT-PCR Result (n = 32) n (%)	Positive Serology Result (n = 44) n (%)
Cough	40 (24.8)	20 (64.5)	20 (46.5)
Fatigue	40 (24.8)	18 (58.1)	20 (46.5)
Odynophagia	38 (23.6)	9 (29.0)	11 (25.6)
Loss of taste	26 (16.2)	16 (51.6)	9 (44.2)
Diarrhea	30 (18.6)	9 (29.0)	9 (20.9)
Fever	27 (16.8)	17 (54.8)	19 (44.2)
Dyspnea	16 (9.9)	8 (25.8)	7 (16.3)
Coryza	48 (29.8)	13 (41.9)	16 (37.2)
Anosmia	27 (16.8)	20 (64.5)	23 (53.5)

Three participants did not complete the flu-like symptoms form, which included previous infection reporting, and were, therefore, excluded from the flu-like symptoms analysis.

Abbreviation: RT-PCR, reverse transcription polymerase chain reaction.

Overall, 42 patients (25.6%) had serology results that were compatible with SARS-CoV-2 infection (ie, the presence of neutralizing antibodies and IgG in serum analysis). A total of 32 participants (19.5%) had already been diagnosed with COVID-19 using RT-PCR because of flu-like symptoms. Of these participants, only 1 who was diagnosed using a molecular test had no presence of neutralizing antibodies (CPE-VNT <20), IgA, IgM, or IgG. Four participants showed discordant results between the dosages of neutralizing antibodies and IgG. Of these participants, only 2 had neutralizing antibodies and had already been diagnosed using molecular tests (Table 3).

Table 3.

Serology Pattern From a Previous SARS-CoV-2 PCR Result

Test	Previous Positive SARS-CoV-2 PCR Result (n = 32)	No Previous Infection Report (n = 129)
IgM, mean (SD)	95.3 (377.2)Positive = 11 (34.4%)	12.9 (143.4)Positive = 5 (3.9%)
IgG, mean (SD)	1,478.2 (1,403.0)Positive = 29 (90.6%)	189.0 (661.4)Positive = 14 (10.8%)
IgA, mean (SD)	385.6 (834.1)Positive = 12 (37.5%)	8.6 (95.6)Positive = 4 (3.1%)
CPE-VNT	Positive = 31 (96.9%)	Positive = 13 (10.1%)

Abbreviations: CPE-VNT, cytopathic effect-based virus neutralization test; PCR, polymerase chain reaction.

The comparison between IgG measurements and neutralizing antibody titers demonstrated a strong correlation, with a Spearman's rank correlation coefficient of 0.74 (P < .001), as shown in the Figure. This fact was correlated with antibody kinetics in relation to the moment of illness of each patient²⁹; however, this was not determined in the scope of our study. Qualitative agreement analysis of IgG and neutralizing antibody measurements was 95.45% for positive results, 98.31% for negative results, and 98.15% for the overall analysis. Although little data are provided the literature, these values were higher than those reported by Dolscheid-Pommerich et al in 2022,³⁰ who followed up with healthcare professionals during the pandemic.

Figure.

Correlation of IgG measurements and neutralizing antibody titers.

Healthcare professionals without a previous diagnosis of COVID-19 (n = 129) were included in the analysis of data obtained from the questionnaires. Table 4 shows the distribution of the responses in relation to the presence of neutralizing antibodies.

Table 4.

List of Responses to the Individual Questionnaire and Detection of Neutralizing Antibodies by Participants With No Previous COVID-19 Infection (N = 129)

Characteristics	CPE-VNT Positive (n = 13) n (%)^a	CPE-VNT Negative (n = 116) n (%)^a
Mean age (standard deviation), years	39.1 (7.3)	37.1 (7.2)
Sex
Female	5 (38.5)	74 (63.8)
Male	8 (61.5)	42 (36.2)
Profession
Technician	10 (76.9)	55 (47.4)
Nurse	2 (15.4)	33 (28.5)
Physician	1 (7.7)	28 (24.1)
Workload^b (standard deviation), monthly hours	144.3 (21.7)	124.0 (43.5)
Means of transport
Private	8 (61.5)	66 (56.9)
Collective	4 (30.8)	41 (35.3)
Others	1 (7.7)	9 (7.8)
Works out? – Yes	3 (25.0)	40 (37.7)
Comorbidity asthma – Yes	1 (7.7)	8 (6.9)
Lives with children? – Yes	10 (76.9)	64 (55.2)
Lives with older people? – Yes	6 (46.1)	23 (20.2)
Lives with symptomatic people? – Yes	3 (23.1)	41 (35.3)
Agglomeration visit – Yes	2 (15.4)	38 (33.0)
Apron use
All the time	8 (61.5)	98 (85.2)
Most of the time	5 (38.5)	17 (14.8)
Mask use^b
All the time	8 (61.5)	107 (93.0)
Most of the time	5 (38.5)	8 (7.0)
Which mask?
N95 only	10 (76.9)	99 (86.1)
Surgical or N95	3 (23.1)	16 (13.9)
Which protection?
Goggles plus face shield	7 (53.8)	64 (55.6)
Face shield only	4 (30.8)	37 (32.2)
Goggles only	2 (15.4)	14 (12.2)
Handwashing
All the time	12 (92.3)	105 (91.3)
Most of the time	1 (7.7)	10 (8.7)
PPE training – Yes	12 (92.3)	108 (93.9)
Confident about PPE use^b – Yes	7 (53.8)	104 (90.4)
OTI participation – Yes	7 (53.8)	52 (46.0)
Believes to be infected – Yes	6 (46.1%)	29 (25.2)

Unless other value is indicated; ^bP < .05. Abbreviations: CPE-VNT, cytopathic effect-based virus neutralization test; OTI, orotracheal intubation; PPE, personal protective equipment.

Data analysis showed a positive correlation with duration of work, safety in wearing and reusing PPE, and presence of anosmia; however, there was a negative relationship with time of mask use. Other variables, such as sex, profession, lifestyle, and means of transport, were not significantly correlated with the detection of neutralizing antibodies. Symptoms other than anosmia and self-perception of having already been infected were also unreliable predictors (Table 4).

The logistic regression model was used for statistically significant variables in the univariate analysis and variables that corresponded to occupational information, including workload, duration of mask use, and confidence in the use of PPE. In this model, only being confident in the use of PPE remained significant (Table 5).

Table 5.

Logistic Regression Model for Positive Serology

Predictors	B	SE(B)	OR (95% CI)
Mask use	1.74	1.95	5.53 (1.18 to 25.86)
Confident about PPE use^a	2.22	0.79	8.93 (2.09 to 38.2)
Workload	-0.02	0.74	0.98 (0.95 to 1.01)

P < .05. Abbreviation: PPE, personal protective equipment.

The characteristics of the participants and their confidence in PPE use are shown in Table 6. Only the type of face protection used was found to be statistically significant (P < .05), with the participants most confident in PPE use indicating they used goggles plus face shields (n = 65, 58.6%) or face shields only (n = 30. 27.0%).

Table 6.

Characteristics of Participants and Confidence About PPE Use by Participants With No Previous COVID-19 Infection (N = 129)

Characteristics	Confident About PPE Use (n = 111) n (%)^a	Not Confident About PPE Use (n = 18) n (%)^a
Mean age (standard deviation), years	37.3 (7.2)	37.8 (7.7)
Sex
Female	65 (58.6%)	14 (77.8%)
Male	46 (41.4%)	4 (22.2%)
Profession
Technician	56 (50.5%)	9 (50.0%)
Nurse	31 (27.9%)	4 (22.2%)
Physician	24 (21.6%)	5 (27.8%)
Workload (standard deviation), monthly hours	128.0 (40.3)	113.9 (52.6)
Means of transport
Private	61 (55.0%)	13 (72.2%)
Collective	41 (36.9%)	4 (22.2%)
Others	9 (8.1%)	1 (5.6%)
Works out? – Yes	37 (36.3%)	6 (37.5%)
Comorbidity asthma – Yes	8 (7.2%)	1 (5.5%)
Lives with children? – Yes	66 (59.5%)	8 (44.4%)
Lives with older people? – Yes	25 (22.9%)	4 (22.2%)
Lives with symptomatic people? – Yes	38 (34.2%)	6 (33.3%)
Agglomeration visit – Yes	37 (33.3%)	3 (17.6%)
Apron use
All the time	95 (85.6%)	11 (64.7%)
Most of the time	16 (14.4%)	6 (35.3%)
Mask use
All the time	102 (91.9%)	13 (76.5%)
Most of the time	9 (8.1%)	4 (23.5%)
Which mask?
N95 only	94 (84.7%)	15 (88.2%)
Surgical or N95	17 (15.3%)	2 (11.8%)
Which facial protection?
Goggles plus face shield	65 (58.6%)	6 (35.3%)
Face shield only	30 (27.0%)	0
Goggles only	16 (14.4%)	11 (64.7%)
Handwashing
All the time	101 (91.0%)	16 (94.1%)
Most of the time	10 (9.0%)	1 (5.9%)
PPE training – Yes	105 (94.6%)	15 (88.2%)
OTI participation – Yes	51 (46.4%)	8 (50.0%)
Believes to be infected – Yes	28 (25.4%)	7 (38.9%)

Unless other value is indicated. Abbreviations: OTI, orotracheal intubation; PPE, personal protective equipment.

Discussion

The results showed that approximately a quarter of frontline healthcare professionals who participated in the study had evidence of SARS-CoV-2 infection within the first 3 months of the pandemic. During this period, sanitary protocols were implemented to control the pandemic and reduce infection among health teams.

There are questions regarding the role of face masks in reducing the spread of SARS-CoV-2 infection.³¹ N95 mask use became mandatory in care units for symptomatic respiratory patients based on prospective studies, indicating that its use reduces the risk of infection.^23,32-34 Similarly, our study showed a negative correlation between the duration of mask use and the propensity to become infected with viruses.

Seminars on staff attire and safety training regarding aerosol-spraying procedures were provided, in addition to visits, to assess the suitability of protocols by the care teams. Our data showed that participants who felt confident about PPE use were more likely to use goggles plus face shields or face shields alone, as instructed during the PPE training program, suggesting that they followed the training protocol more properly. The data also showed that being confident about the use of PPE was a less significant predisposing factor to infection during the pandemic. The education program on PPE use may reduce COVID-19 infection risk by reinforcing the importance of correctly wearing PPE and highlighting contamination risks related to its manipulation. Therefore, as suggested in another study,³⁵ investment in staff training is associated with a lower rate of infection within the health environment.

Despite the consistent association between confidence in the use of PPE and the presence of positive serology, the number of participants who did not feel confident was relatively low (n = 18, 14% of 129 participants who did not have documented history of previous COVID-19 infection). The reasons for lack of confidence were not within the scope of this study; however, a recent review by Houghton et al³⁶ lists the following as the main reasons for lack of confidence within a team: use of guidelines that are too long or difficult to follow, dissociation between the current protocol and national or international guidelines, and frequent changes in guidelines related to PPE use protocols.

Among all the participants, 96 (58.5%) reported at least 1 flu-like symptom. This represents more than twice the number of individuals with positive serology. This is probably because they represent nonspecific symptoms that are easily confused with symptoms of other health conditions or even other respiratory diseases. In this sense, self-reported symptoms, as a single parameter, may not be appropriate for diagnostic monitoring of COVID-19 among healthcare workers. A 2018 qualitative study conducted in the context of measures to prevent and control pulmonary tuberculosis indicated that the presence of infected coworkers contributes to increased adherence to PPE use.³⁷ Our study was conducted using services with records of infrastructure and knowledge inadequacies. When we conducted the study, approximately 21% of the workforce had already been diagnosed with COVID-19 using RT-PCR to detect SARS-CoV-2. Over time, living with infected coworkers and the knowledge of possible asymptomatic infections began to generate the belief that those who were not infected had already been asymptomatic carriers and were immune to the disease. Paradoxically, this belief could potentially act as an impediment to the proper use of PPE. Among the 35 healthcare professionals without a documented history of previous infection and who had this belief, only 17% (n = 6) were correct. These data prove that the human perception of exposure is flawed in the context of COVID-19 and that surveillance must remain constant.

Another important factor was that the isolation of patients was not systematic, even though this issue was later amended. With the spread of the virus, emergency services have isolated the place of care for patients with respiratory symptoms on the basis of recommendations and national and international experiences. Physical isolation of suspected patients has a decisive effect on infection among healthcare professionals.^38-40

Our study had some limitations. First, it was a single-center study and participation was voluntary. Second, the participants' characteristics were collected via a self-report survey, resulting in the possibility of self-report bias.

Conclusion

Our data show that healthcare professionals' perception of symptoms is not a good screening parameter for the diagnosis of an infectious disease with respiratory symptoms, such as COVID-19. The main influencing factor for the control of infection is the elaboration of workflows and safety protocols based on simple and clear rules and investments in team training.

References

Gereffi

. What does the COVID-19 pandemic teach us about global value chains? The case of medical supplies. J Int Bus Policy. 2020; 3(3):287-301.

World Health Organization (WHO). The Impact of COVID-19 on Health and Care Workers: A Closer Look at Deaths. Geneva, WHO; 2021. Accessed July 12, 2022. https://apps.who.int/iris/handle/10665/345300

Bandyopadhyay

, Baticulon

, Kadhum

, et al. Infection and mortality of healthcare workers worldwide from COVID-19: a systematic review. BMJ Global Health. 2020; 5(12):e003097.

Schwartz

, King

C-C

, Yen

M-Y

. Protecting health care workers during the COVID-19 coronavirus outbreak: lessons from Taiwan's SARS response. Clin Infect Dis. 2020; 71(15):858-860.

Zhou

, Huang

, Xiao

, Huang

, Fan

. Protecting Chinese healthcare workers while combating the 2019 novel coronavirus. Infect Control Hosp Epidemiol. 2020; 41(6):745-746.

Chan

Y-J

, Lee

C-L

, Hwang

S-J

, et al. Seroprevalence of antibodies to pandemic (H1N1) 2009 influenza virus among hospital staff in a medical center in Taiwan. J Chin Med Assoc. 2010; 73(2):62-66.

Chen

MIC

, Lee

VJM

, Lim

W-Y

, et al. 2009 influenza A(H1N1) seroconversion rates and risk factors among distinct adult cohorts in Singapore. JAMA. 2010; 303(14):1383-1391.

Bandaranayake

, Huang

, Bissielo

, et al. Risk factors and immunity in a nationally representative population following the 2009 influenza A(H1N1) pandemic. PLoS One. 2010; 5(10):e13211.

Broberg

, Nicoll

, Amato-Gauci

. Seroprevalence to influenza A(H1N1) 2009 virus—where are we?. Clin Vaccine Immunol. 2011; 18(8):1205-1212.

10.

Chokephaibulkit

, Assanasen

, Apisarnthanarak

, et al. Seroprevalence of 2009 H1N1 virus infection and self-reported infection control practices among healthcare professionals following the first outbreak in Bangkok, Thailand. Influenza Other Respir Viruses. 2013; 7(3):359-363.

11.

Miller

, Hoschler

, Hardelid

, Stanford

, Andrews

, Zambon

. Incidence of 2009 pandemic influenza A H1N1 infection in England: a cross-sectional serological study. Lancet. 2010; 375(9720):1100-1108.

12.

Huff

, Singh

. Asymptomatic transmission during the coronavirus disease 2019 pandemic and implications for public health strategies. Clin Infect Dis. 2020; 71(10):2752-2756.

13.

Huang

, Zhang

, et al. Rapid asymptomatic transmission of COVID-19 during the incubation period demonstrating strong infectivity in a cluster of youngsters aged 16-23 years outside Wuhan and characteristics of young patients with COVID-19: a prospective contact-tracing study. J Infect. 2020; 80(6):e1-e13.

14.

, Xu

, Rong

, et al. Delivery of infection from asymptomatic carriers of COVID-19 in a familial cluster. Int J Infect Dis. 2020; 94:133-138.

15.

Kursumovic

, Lennane

, Cook

. Deaths in healthcare workers due to COVID-19: the need for robust data and analysis. Anaesthesia. 2020; 75(8):989-992.

16.

Chou

, Dana

, Buckley

, Selph

, Fu

, Totten

. Epidemiology of and risk factors for coronavirus infection in health care workers: a living rapid review. Ann Intern Med. 2020; 173(2):120-136.

17.

Remuzzi

, Remuzzi

. COVID-19 and Italy: what next?. Lancet. 2020; 395(10231):1225-1228.

18.

Equipo profesional de Red Nacional de Vigilancia Epidemiologica (RENAVE). Análisis de los casos de COVID-19 en personal sanitario notificados a la RENAVE hasta el 10 de mayo en España. Madrid: Instituto de Salud Carlos III; 2020. Accessed August 4, 2022. Accessed">https://www.isciii.es/QueHacemos/Servicios/VigilanciaSaludPublicaRENAVE/EnfermedadesTransmisibles/Documents/INFORMES/Informes%20COVID-19/COVID-19%20en%20personal%20sanitario%2029%20de%20mayo%20de%202020.pdf">Accessed August 4, 2022. https://www.isciii.es/QueHacemos/Servicios/VigilanciaSaludPublicaRENAVE/EnfermedadesTransmisibles/Documents/INFORMES/Informes%20COVID-19/COVID-19%20en%20personal%20sanitario%2029%20de%20mayo%20de%202020.pdf

19.

Self

, Tenforde

, Stubblefield

, et al. Seroprevalence of SARS-CoV-2 among frontline health care personnel in a multistate hospital network — 13 academic medical centers, April–June 2020. MMWR Morb Mortal Wkly Rep. 2020; 69(35):1221-1226.

20.

Buonafine

, Paiatto

BNM

, Leal

, et al. High prevalence of SARS-CoV-2 infection among symptomatic healthcare workers in a large university tertiary hospital in São Paulo, Brazil. BMC Infect Dis. 2020; 20(1):917.

21.

Martin-Delgado

, Viteri

, Mula

, et al. Availability of personal protective equipment and diagnostic and treatment facilities for healthcare workers involved in COVID-19 care: a cross-sectional study in Brazil, Colombia, and Ecuador. PLoS One. 2020; 15(11):e0242185.

22.

Grifoni

, Weiskopf

, Ramirez

, et al. Targets of T cell responses to SARS-CoV-2 coronavirus in humans with COVID-19 disease and unexposed individuals. Cell. 2020; 181(7):1489-1501.e15.

23.

Reddy

, Valderrama

, Kuhar

. Improving the use of personal protective equipment: applying lessons learned. Clin Infect Dis. 2019; 69(suppl 3):S165-S170.

24.

CDC COVID-19 Response Team. Characteristics of health care personnel with COVID-19 — United States, February 12–April 9, 2020. MMWR Morb Mortal Wkly Rep. 2020; 69(15):477-481.

25.

Tian

, Stedman

, Whyte

, Anderson

, Thomson

, Heald

. Personal protective equipment (PPE) and infection among healthcare workers – what is the evidence?. Int J Clin Pract. 2020; 74(11):4-6.

26.

Gómez-Ochoa

, Franco

, Rojas

, et al. COVID-19 in healthcare workers: a living systematic review and meta-analysis of prevalence, risk factors, clinical characteristics, and outcomes. Am J Epidemiol. 2020; 190(1):161-175.

27.

Wendel

, Kutner

, Machado

, et al. Screening for SARS-CoV-2 antibodies in convalescent plasma in Brazil: preliminary lessons from a voluntary convalescent donor program. Transfusion. 2020; 60(12):2938-2951.

28.

World Health Organization (WHO). How to Put On and Take Off Personal Protective Equipment (PPE). Geneva: WHO; 2008. Accessed July 12, 2022. https://apps.who.int/iris/handle/10665/70066

29.

Sun

, Tang

, Bai

, et al.

The kinetics of viral load and antibodies to SARS-CoV-2.

Clin Microbiol Infect. 2020;26(12):1690.e1-1690.e4.

30.

Dolscheid-Pommerich

, Bartok

, Renn

, et al. Correlation between a quantitative anti-SARS-CoV-2 IgG ELISA and neutralization activity. J Med Virol. 2022; 94(1):388-392.

31.

Ippolito

, Vitale

, Accurso

, et al. Medical masks and respirators for the protection of healthcare workers from SARS-CoV-2 and other viruses. Pulmonology. 2020; 26(4):204-212.

32.

Dehaghi

, Ghodrati-Torbati

, Teimori

, Ghavamabadi

, Jamshidnezhad

. Face masks vs. COVID-19: a systematic review. Invest Educ Enferm. 2020; 38(2):155-162.

33.

Bartoszko

, Farooqi

MAM

, Alhazzani

, Loeb

. Medical masks vs N95 respirators for preventing COVID-19 in healthcare workers: a systematic review and meta-analysis of randomized trials. Influenza Other Respir Viruses. 2020; 14(4):365-373.

34.

Friese

, Veenema

, Johnson

, Jayaraman

, Chang

, Clever

. Respiratory protection considerations for healthcare workers during the COVID-19 pandemic. Health Security. 2020; 18(3):237-240.

35.

Gholami

, Fawad

, Shadan

, et al. COVID-19 and healthcare workers: a systematic review and meta-analysis. Int J Infect Dis. 2021; 104:335-346.

36.

Houghton

, Meskell

, Delaney

, et al. Barriers and facilitators to healthcare workers' adherence with infection prevention and control (IPC) guidelines for respiratory infectious diseases: a rapid qualitative evidence synthesis. Cochrane Database Syst Rev. 2020; 4(4):CD013582.

37.

Zinatsa

, Engelbrecht

, van Rensburg

, Kigozi

. Voices from the frontline: barriers and strategies to improve tuberculosis infection control in primary health care facilities in South Africa. BMC Health Serv Res. 2018; 18:269.

38.

Park

J-H

, Lee

S-G

, Ahn

, et al. Strategies to prevent COVID-19 transmission in the emergency department of a regional base hospital in Korea: from index patient until pandemic declaration. Am J Emerg Med. 2021; 46:247-253.

39.

Chung

, Lee

, Kim

, et al. Revised triage and surveillance protocols for temporary emergency department closures in tertiary hospitals as a response to COVID-19 crisis in Daegu Metropolitan city. J Korean Med Sci. 2020; 35(19):e189.

40.

Choi

, Kim

, Cho

, Noh

, Park

, Chung

. Efficacy of a four-tier infection response system in the emergency department during the coronavirus disease-2019 outbreak. PLoS One. 2021; 16(8):e0256116.