Risk of COVID-19 infection among frontline healthcare workers during the COVID-19 pandemic

Abstract

BACKGROUND

In the initial stages of the coronavirus disease 2019 (COVID-19) pandemic, healthcare workers (HCWs) who were immunologically naive to COVID-19, were exposed to a highly transmissible virus.

AIM

To compare infection risk among HCWs in high-risk (HR) and low-risk (LR) areas.

METHODS

Data on reverse transcriptase-polymerase chain reaction confirmed clinical infection and samples for nucleocapsid, and spike protein antibodies were collected at five time-points (T1 to T5) from HCWs in the emergency department and intensive care unit (HR group) and pre-clinical and para-clinical areas (LR). For the sero-study, only participants who provided at least one baseline sample and one during the second wave (T4 or T5) were analysed. Since Covishield^TM elicits only spike protein antibodies, subclinical infection was diagnosed if asymptomatic unvaccinated and Covishield^TM vaccinated individuals tested positive for nucleocapsid antibody.

RESULTS

Overall, by T5, clinical infection rate was similar in the HR (120/366, 32.8%) and LR (22/82, 26.8%) groups (P = 0.17). However, before vaccination (T3), more HCWs in the HR group developed COVID-19 infection (21.9% vs 8.8%, P = 0.046). In the sero-study group, clinical infection occurred in 31.5% (45/143) and 23.7% (14/59) in the HR and LR groups respectively (P = 0.23). Spike antibody was detected in 140/143 (97.9%) and 56/59 (94.9%) and nucleocapsid antibody was positive in 95/143 (66.4%) and 35/59 (59.3%) in the HR and LR groups respectively (P = 0.34). Subclinical infection rate (HR 34.9%, LR 35.6%, P = 0.37) and hospitalization rate were similar. There was no mortality.

CONCLUSION

Before vaccination, HCWs in HR areas had a higher risk of infection. Seroprevalence studies suggest that sub-clinical infection was not uncommon.

Key Words: COVID-19 pandemic; Seroprevalence; Healthcare workers; SARS-CoV-2 antibodies; Nucleocapsid antibody; Spike protein antibody

Core Tip: This study tracked clinical infection and seroprevalence to severe acute respiratory syndrome-coronavirus 2 virus during the two waves of the pandemic in India and compared infection risk between health care workers (HCWs) in high-risk (HR) areas (emergency department, critical care) and low-risk (LR) non-clinical areas. The seroprevalence rate of 1.1% at the start of the pandemic increased to 34.1% and 60.1% during the first and second waves respectively. Prior to vaccination, more HCWs in HR areas developed clinical infection. Following vaccination, clinical infection rates and seropositivity were similar in HR and LR groups. About 1/3^rd had evidence of subclinical infection.

INTRODUCTION

The rapid spread of the coronavirus disease 2019 (COVID-19) pandemic constrained medical services worldwide. As of May 2023, there were over 700 million confirmed cases of COVID-19 and 6.5 million deaths[1]. The pandemic raised concerns regarding the safety of healthcare workers (HCWs) who were in close contact with infected patients[2-4]. This led to anxiety, mental stress, and decreased morale that resulted in absenteeism[5,6]. Governments resorted to nationwide lockdowns and infection control measures were strictly enforced.

Infection control measures were put in place during the pandemic which included surveillance, personal protection equipment (PPE), and environmental measures[7]. Vaccination against COVID-19 was considered to be the panacea and was rapidly developed[8-10]. Frontline workers such as HCWs were preferentially vaccinated in order to protect them against serious illness including intensive care unit (ICU) admission and death[11,12].

Studies during the H1N1 pandemic showed that HCWs were at a higher risk of contracting influenza when compared to the general population[13]. Seroprevalence studies showed that sub-clinical infection was common[14]. However, positivity rates were similar among HCWs working in high-risk (HR) and low-risk (LR) areas[15].

In the context of the COVID-19 pandemic, it was important to evaluate the risk of COVID-19 infection among HCWs, who were immunologically naive to COVID-19 and exposed to a highly transmissible virus at the workplace. It was also essential to track infection and seroconversion rates as the pandemic evolved and vaccination became available. This study was thus undertaken to evaluate clinical and sub-clinical infection among HCWs during different time-points in the pandemic and to assess if the risk of COVID-19 infection was higher among HCWs working in HR areas such as the emergency department (ED) and the ICU where they are in direct contact with COVID-19-infected patients when compared with LR areas such as pre-clinical and para-clinical departments of a medical institution, where HCWs do not come in direct contact with patients.

MATERIALS AND METHODS

Participants and setting

This prospective single-center observational study was conducted in a large tertiary care referral hospital in India between June 1, 2020, to August 31, 2021 (15 months), and included HCWs from various departments. HCWs working in HR areas included those working in the ED, ICU, and general medical wards, and HCWs working in LR areas included the departments of anatomy, physiology, biochemistry, and community medicine. HCWs were excluded if they did not consent to participate in the study.

Demographic and outcome data

Demographic data including age, gender, and co-morbidities were recorded. Serum samples were collected at five time-points for the seroprevalence component of the study (Figure 1). The first wave of the COVID-19 pandemic in India was witnessed between June 2020 and September 2020 while the second wave started in March 2021 and started waning by June 2021. The first sample was taken in April 2020 (T1), after the lockdown was initiated and after the first COVID-19 positive patient was diagnosed in the institution on 28^th March 2020. The second sample was collected in June 2020-July 2020 (T2) as the cases started increasing during the first wave of the pandemic. The first health care worker (HCW) in our institution was tested positive for COVID-19 on 5^th July 2020. The third sample was collected in October 2020 (T3) after the first wave of the pandemic waned and before vaccination became available in the country. The fourth sample was collected in March 2021 (T4) about 2-months after vaccination was rolled out in the country and as the country witnessed the start of the second wave of the pandemic, while the final sample was collected in June 2021 (T5) as the second wave of the pandemic waned and when two doses of vaccination among HCWs was completed in a large proportion of HCWs. For the seroprevalence component of the study, only those who had given at least one sample in the initial period of the pandemic (T1 to T3) and at least one sample during the second wave of the pandemic (T4 or T5) were included.

Figure 1 The timeline of the study. T1 to T5 reflects the five time points when sampling was done for the seroprevalence study. The first coronavirus disease 2019 (COVID-19) case was reported in India on 27^th January 2020. The index case of COVID-19 was reported in our institution on 28^th March 2020 while the first confirmed COVID-19 infection in a healthcare worker (HCW) was reported on 5^th July 2020. The period T1 to T3 denotes the first wave of the pandemic during which time vaccine was not available. Vaccination for COVID-19 started in India for HCWs on January 16^th, 2021. T4 and T5 denote the period when vaccination was rolled out in a phased manner and also reflects the period when the second wave of the pandemic occurred in India. COVID-19: Coronavirus disease 2019; HCW: Healthcare worker.

At each time point, when samples were collected for the seroprevalence part of the study, participants were asked if they had COVID-19 infection in the preceding period. COVID-19 infection was diagnosed when severe acute respiratory syndrome-coronavirus 2 (SARS-CoV-2) RNA was detected in respiratory samples of symptomatic individuals using reverse transcriptase-polymerase chain reaction assays (Altona RealStar^® SARS-CoV-2) and/or Cartridge Based Nucleic Acid Amplification Test (CBNAAT) assay (Xpert^® Xpress SARS-CoV-2). Vaccination status and the details of vaccination were recorded. In India, the Oxford-Astra Zeneca’s chimpanzee adenovirus vectored ChadOx1 vaccine, Covishield^TM, and the Bharat biotech’s whole virion inactivated BBV152, Covaxin^® were available[16]. Since Covishield^TM was available much earlier than Covaxin^®, more HCWs received Covishield^TM; Sputnik V (also known as Gam-COVID-Vac), an adenovirus viral vector vaccine, was available at a much later stage and only one HCW included in the study received it.

The duration of work in COVID-19 areas was documented for all the HCWs. Since the healthcare system was overwhelmed during the second wave of the pandemic, some of the HCWs working in LR areas were posted for a limited time to work in COVID-19 areas. Data on number of HCWs who developed COVID-19 infection, need for home isolation or hospitalization, requirement of oxygen therapy, ventilation and hospital outcome were recorded.

The primary outcome of the study was to determine if HCWs working in HR areas were at a higher risk of developing COVID-19 infection when compared with HCWs working in LR areas at different time-points of the pandemic. The study also looked at seroconversion of HCWs to COVID-19 at five time points to evaluate the proportion of HCWs who had subclinical infection. Since Covishield^TM elicits only spike protein antibodies, subclinical infection was diagnosed if asymptomatic HCWs who were either unvaccinated or were vaccinated with Covishield^TM tested positive for nucleocapsid antibody. Covaxin^® elicits both spike protein and nucleocapsid antibodies. This would have made it difficult to differentiate antibody response to clinical or subclinical infection vs vaccination. Hence Covaxin^®-vaccinated individuals were excluded for the determination of subclinical infection among HCWs. For the same reason, seroprevalence data among HCWs during the 5 time periods was also ascertained based on positivity to the nucleocapsid protein.

Analysis of samples

Blood samples were drawn for in vitro qualitative detection of antibodies against SARS-CoV-2 virus; both nucleocapsid (N) and spike (S) protein antibodies were estimated using Elecsys^® Anti-SARS-CoV-2, Roche diagnostics. This employs the chemiluminescent immunosorbent assay method which is more sensitive and specific than other available serological methods[17]. The antibody levels were considered positive if either the N or S antibody levels were above the cut-off index (COI) reference limits. Nucleocapsid antibody was considered positive if COI was ≥ 1.0 U/mL, while for the spike antibody, a COI of ≥ 0.8 U/mL was considered positive.

Statistical analysis

The proportion of influenza-like illness in high and LR HCWs was assumed to be 10% and 1% respectively, with an anticipated odds ratio of 1.72. The study was powered to 80% with a type 1 error of 0.05. The calculated sample size worked out to 150 with 75 in each arm.

Data was analyzed using the Statistical Package for the Social Sciences for Windows software released in 2015, version 23.0, Armonk, New York, United States. Mean and SD were used for continuous variables and percentages for categorical and nominal variables. Seropositivity and clinical outcomes were compared between the HR and LR using student’s t-test or Fisher’s exact test as appropriate. For all tests, a 2-sided P-value less than 0.05 was considered statistically significant.

Ethics

The study was approved by the Institutional Review Board And Ethics Committee (No. 12751, dated 1 May, 2020).

RESULTS

Between June 2020 and August 2021, 448 HCWs consented to participate. A majority (n = 366, 81.7%) worked in HR areas while the remaining worked in LR areas (n = 82, 18.3%). Clinical infection during the study period was assessed in all HCWs, while sero-study outcomes were available on 143 HCWs in the HR group and 59 HCWs in the LR group (Figure 2).

Figure 2 The seropositivity outcomes among healthcare workers working in high-risk and low-risk areas in our hospital. High-risk areas refer to those working in the emergency department and/or the medical intensive care unit; low-risk (LR) areas refer to those working in paraclinical areas such as anatomy, physiology, biochemistry, or community medicine. Although healthcare workers (HCWs) in LR areas were not supposed to work in coronavirus disease 2019 (COVID-19) wards, they were requested to work in COVID-19 wards for short periods during the peak of the 2^nd wave when the health system was overwhelmed with many patients. Clinical infection occurred among 120 of the 366 HCWs who were working in high-risk (HR) areas. Sero-study was done in 143 of the 366 HCWs in HR areas. At the end of the study 66.4% were positive for nucleocapsid antibodies and 97.9% were positive for spike antibodies. Among HCWs working in LR areas, 22 of the 82 HCWs developed clinical infection. At the end of the study 59.3% were positive for nucleocapsid antibodies and 94.9% were positive for spike antibodies. ¹Only 202 HCWs who gave at least one sample for antibodies during the second wave (T4 or T5) and at least one baseline or T2 or T3 sample, and those who were unvaccinated or were vaccinated with COVISHIELD™ were included for the analysis on sero-study outcome. HCWs who were vaccinated with COVAXIN^® were excluded from the sero-study analysis as it triggers nucleocapsid and spike antibodies and mimics infection. Clinical outcomes were followed up for all 448 HCWs.

Demographic characteristics, duration of work in COVID areas and vaccination status are summarized in Table 1. The mean (SD) age was significantly lower in HCWs working in HR areas when compared with those working in LR areas [31.7 (8.3 years) years vs 39.9 (10.4 years) years, P < 0.001]. A higher proportion of males worked in HR areas (54.9% vs 39.1%, P = 0.01). The frequency of co-morbidities was low (13.4%) and similar in the two groups (P = 0.27). The duration of work in COVID-19 areas was significantly higher among HCWs working in HR areas when compared with LR areas [13 (4.6 months) months vs 1.8 (1.2 months) months, P < 0.001]. Four hundred and two HCWs (89.7%) had received at least one dose of COVID-19 vaccine by the end of the study period. A majority were vaccinated with Covishield^TM vaccine (83.7%) while a small proportion (5.8%) received Covaxin^®; one HCW received Sputnik V.

Table 1 Characteristics of the healthcare workers at baseline, n (%).

Characteristic	High-risk areas (n = 366)	Low-risk areas (n = 82)
Age, mean (SD) years	31.7 (8.3)	39.9 (10.4)
Gender, male sex	201 (54.9)	32 (39.1)
Comorbidities
Presence of at least one co-morbid illness	49 (13.4)	49 (13.4)
Diabetes mellitus	17 (4.6)	8 (9.7)
Hypertension	14 (3.8)	10 (12.2)
Obstructive airway disease	31 (8.5)	9 (10.9)
Duration of work in COVID-19 areas, mean (SD) months1	13 (4.6)	1.8 (1.2)
Vaccination details2
Vaccinated against (COVID-19)2	325 (88.8)	77 (93.9)
Received two doses	288 (78.7)	65 (79.3)
Received one dose	37 (10.1)	12 (14.6)
Unvaccinated	41 (11.2)	5 (6.1)
Type of vaccine
COVISHIELD™	306 (94.2)	69 (84.4)
COVAXIN™	18 (5.5)	8 (10.4)
Sputnik V	1 (0.3)	0 (0)

¹In the low-risk areas, healthcare workers were requested to work in coronavirus disease 2019 (COVID-19) wards during the peak of the 2^nd wave when the health system was overwhelmed with a large number of patients. However, there was a significant (P < 0.001) difference of 11.3 months between the two groups: 95%CI: 10.3-12.3 months.

²Vaccination for COVID-19 started in India for healthcare workers on January 16^th, 2021 (7 months after the study started), and the status is given as of August 31^st, 2021.

All values are expressed as numbers (n) and percentages unless specified otherwise. COVID-19: Coronavirus disease 2019.

Overall, 144 (32.1%) HCWs developed COVID-19 infection during the study period (Table 2). There was no difference between the HR and LR groups in terms of clinical infection (32.8% vs 26.8%, P = 0.17), hospital admission (44.2% vs 54.5%, P = 0.58), home isolation (55.8% vs 45.5%, P = 0.51), and length of hospitalization [7.4 (3.6) vs 6.5 (4), P = 0.22]. Only one HCW required oxygen therapy. There was no mortality.

Table 2 Clinical outcomes among healthcare workers who tested positive for coronavirus disease 2019 infection during the study period, n (%).

Parameter	COVID-19-positive in high-risk areas (n = 120)	COVID-19-positive in low-risk areas (n = 22)	P value
COVID-19 positivity rate	120/366 (32.8)	22/82 (26.8)	0.17
Treated as home isolation	67 (55.8)	10 (45.5)	0.51
Hospital admission	53 (44.2)	12 (54.5)	0.58
Admission duration, mean (SD), days	7.4 (3.6)	6.5 (4)	0.22
Oxygen requirement	1 (0.01)	0	Not applicable
Ventilation (invasive or non-invasive)	0	0	-
Mortality	0	0	-

All values are expressed as number (n) and percentage, the number of patients who were admitted is high since government directives mandated admission of even mildly symptomatic patients, particularly during the first wave of the pandemic in India. COVID-19: Coronavirus disease 2019.

Table 3 summarizes the cross-sectional data of clinical infection and seropositivity for the nucleocapsid and spike proteins at five time points (T1-T5). The seroprevalence rate among HCWs, based on the nucleocapsid antibodies, was 1.1% (T1), 17.4% (T2), 34.1% (T3), 44.9% (T4) and 60.1% (T5) at the different time points.

Table 3 Cross-sectional data of clinical infection and seropositivity rates among healthcare workers at various study time periods.

Time point	Clinical infection				Nucleocapsid (N) antibody				Spike (S) antibody
	Overall	HR	LR	P value	Overall	HR	LR	P value	Overall	HR	LR	P value
Baseline
T1 (n = 283) [HR (n = 204); LR (n = 79)]	0/283 (0)	0/204 (0)	0/79 (0)	-	3/283 (1.1)	3/204 (1.5)	0/79 (0)	-	0/283 (0)	0/204 (0)	0/79 (0)	-
First wave
T2 (n = 168) [HR (n = 121); LR (n = 47)]	5/168 (3.0)	3/121 (2.5)	2/47 (4.3)	0.56	30/168 (17.9)	21/121 (17.4)	9/47 (19.1)	0.82	13/168 (7.7)	10/121 (8.3)	3/47 (6.4)	0.70
T3 (n = 214) [HR (n = 146); LR (n = 68)]	38/214 (17.8)	32/146 (21.9)	6/68 (8.8)	0.046	73/214 (34.1)	54/146 (37)	19/68 (27.9)	0.36	59/214 (27.6)	46/146 (31.5)	13/68 (19.1)	0.15
Second wave
T4 (n = 156) [HR (n = 110); LR (n = 46)]	6/156 (3.9)	5/110 (4.5)	1/46 (2.2)	0.50	70/156 (44.9)	51/110 (46.4)	19/46 (41.3)	0.72	138/156 (88.5)	98/110 (89.1)	40/46 (86.9)	0.92
T5 (n = 175) [HR (n = 122); LR (n = 53)]	29/175 (16.6)	21/122 (17.2)	8/53 (15.1)	0.77	106/175 (60.1)	77/122 (63.1)	29/53 (54.7)	0.60	172/175 (98.3)	121/122 (99.1)	51/53 (96.2)	0.11
T51 (n = 166) [HR (n = 118); LR (n = 48)]	24/166 (14.5)	17/118 (14.4)	7/48 (14.6)	0.78	100/166 (60.2)	73/118 (61.9)	27/48 (56.3)	0.42	164/166 (98.8)	117/118 (99.2)	47/48 (97.9)	0.52

¹T5 excludes healthcare workers (HCWs) who were vaccinated with COVAXIN^®.

All values are expressed as numbers (n) and percentages and indicate the number of HCWs; the number of participants varies in the different time periods since not all HCWs provided samples for antibodies at all the time points. T1 to T5 indicates the time point at which seroprevalence was assessed and clinical data was collected. Only individuals who had given at least two antibody samples, including one between T1 to T3 and the second sample at T4 or T5 were included in this analysis. Data on clinical infection represents the number of HCWs who developed clinical infection during that period. COVID-19 antibody levels for Nucleocapsid (N) and Spike (S) proteins were measured using Elecsys^® anti-severe acute respiratory syndrome-coronavirus 2. The antibody levels were considered positive if either N or S antibody levels were above the cut-off index (COI) reference limits. The nucleocapsid antibody was considered positive if COI was ≥ 1.0 U/mL and the spike antibody was considered positive if COI was ≥ 0.8 U/mL. HR: High-risk; LR: Low-risk.

The T1 to T3 period represented the first COVID-19 wave when vaccination was still unavailable in India, during which time seropositivity increased from 1.1% to 34.1%. The first case of COVID-19 infection was reported in India on 27^th January 2020. At T1, which was in April 2020, no clinical infections were reported among HCWs in the institution; however, 3 HCWs in the HR area and none in the LR area tested positive for the nucleocapsid antibody but not for the spike protein. During the second sampling (T2) which was from June 2020 to July 2020, 5 HCWs reported clinical infection in the preceding period (HR 2.5% vs LR 4.3%, P = 0.56). COVID-19 antibodies for the nucleocapsid protein were positive in 17.4% and 19.1% in HR and LR groups respectively (P = 0.82) and the spike protein in 8.3% and 6.4% respectively (P = 0.70).

Between T2 and T3, when the first wave of COVID-19 peaked, HCWs in the HR area had a 2.48 times higher risk (21.9% in HR vs 8.8% in LR, P = 0.046) of developing COVID-19 infection. However, there was no demonstrable difference in seroconversion rates between the HR and LR groups by this time-point (Table 3).

During the second wave (T4 and T5), when the majority of the HCWs were vaccinated, the proportion of HCWs who developed clinical infection was similar in the HR and LR groups. Antibodies to spike protein were evident in 99.2% in the HR group and 97.9% in the LR group (P = 0.52), suggesting good response to vaccination, while 61.9% in the HR group and 56.3% in the LR group respectively had antibodies to nucleocapsid protein (P = 0.42), suggesting prior clinical or subclinical COVID-19 infection.

When the sero-study component was assessed (Table 4), out of the 202 patients who had given at least one sample in the T1 to T3 period and one sample in T4 or T5, clinical infection occurred in 31.5% (45/143) and 23.7% (14/59) in the HR and LR groups respectively (P = 0.23). Those vaccinated with Covaxin^® (11/213, 5.1%) and Sputnik V were excluded from the final sero-study analysis. Given that nucleocapsid antibody was positive in 66.4% in the HR group and 59.3% in the LR group (Table 4), the subclinical infection rates worked out to 34.9% in the HR group and 35.6% in the LR group (P = 0.37).

Table 4 Seropositivity among COVISHIELD™ vaccinated and unvaccinated healthcare workers at T4 or T5, n (%).

Antibody positive		High-risk areas (n = 143)	Low-risk areas (n = 59)	Inference1
Nucleocapsid +	Spike +	95 (66.4)	34 (57.6)	Infection with vaccination
Nucleocapsid -	Spike +	45 (31.5)	22 (37.3)	Vaccination
Nucleocapsid +	Spike -	0 (0)	1 (1.7)	Infection
Nucleocapsid -	Spike -	3 (2.1)	2 (3.4)	Not infected or vaccinated

¹The inference must be made cautiously since all infections and vaccinations do not necessarily produce an antibody response.

All values are expressed as number (n) of healthcare workers (HCWs) and percentage, (+) indicates positive for nucleocapsid or spike antibody while (-) indicates negative for nucleocapsid or spike antibody. Coronavirus disease 2019 (COVID-19) antibody levels for Nucleocapsid (N) and Spike (S) proteins were measured using Elecsys^® anti-severe acute respiratory syndrome-coronavirus 2. The antibody levels were considered positive if either N or S antibody levels were above the cut-off index (COI) reference limits. The nucleocapsid antibody was considered positive if COI was ≥ 1.0 U/mL and the spike antibody was considered positive if COI was ≥ 0.8 U/mL. Since COVAXIN^® can elicit nucleocapsid and spike antibodies and COVISHIELD™ elicits antibodies only to the spike protein, HCWs who were vaccinated with COVAXIN^® were excluded from this analysis to distinguish the antibody response between natural infection and vaccination. Only those who provided at least two samples for antibodies with the 2^nd sample given during the 2^nd wave of COVID-19 (T4 or T5) were included in this analysis. The 11/143 (7.7%) and 5/59 (8.5%) from high-risk (HR) and low-risk (LR) groups were unvaccinated. In the sero-study group, clinical infection occurred in 31.5% (45/143) and 23.7% (14/59) in the HR and LR groups respectively (P = 0.23) during the entire study period.

DISCUSSION

Our study, involving 448 HCWs, tracked seroprevalence to the SARS-CoV-2 virus during the first two waves of the pandemic at five time-points and provided valuable insights into the virus's impact and vaccination efficacy. The seroprevalence rate among HCWs was 1.1% at the start of the pandemic and increased to 34.1% during the first wave and to 60.1% during the second wave of the pandemic. Prior to vaccination, during the first wave of the pandemic, a significantly higher proportion of HCWs in HR areas developed clinical infection when compared with HCWs working in LR areas (21.9% vs 8.8%, P = 0.046); seropositivity was, however, similar in both groups (37% vs 27.9%, P = 0.36). Following vaccination, during the second wave of the pandemic, clinical infection rates (17.2% vs 15.1%, P = 0.77) and seropositivity (63.1% vs 54.7%, P = 0.60) were similar in HR and LR groups. Subclinical infection rates were 34.9% in the HR group and 35.6% in the LR group (P = 0.37).

Table 5 presents the seroprevalence data among HCWs from studies across the globe[18-33]. At the initial stages of the pandemic, when the population was immunologically naïve to the novel COVID-19 pathogen, as expected, only 1.1% of the HCWs were seropositive in the current study in April 2020. These results are comparable with the seropositivity of 0.3% from Japan in 2020[18] and the 0.5% in March 2020 from Italy[19]. The seroprevalence data from Brazil (5.5%, June 2020)[20], the United States (6%, April 2020 to June 2020)[21], Switzerland (9.6%, April 2020)[22], Netherlands (13.9%, June 2020 to July 2020)[23], Sweden (19.1%, April 2020 to May 2020)[24], and the United Kingdom (24.4%, April 2020)[25] probably correspond to the seroprevalence of 17.9% during the upslope of the first wave of the pandemic in our study (T2 sampling, June 2020 to July 2020).

Table 5 Studies on seroprevalence among healthcare workers.

Ref.	Country	Period	Sample size	Sero-prevalence
Current study	India	April 2020 (T1)	283	1.1%
		June 2020 to July 2020 (T2)	168	17.9%
		October 2020 (T3)	214	34.1%
		March 2021(T4)	156	44.9%
		June 2021(T5)	175	60.1%
Milazzo et al[19]	Italy	March 2020	697	0.5%
		May 2020		5.4%
Piccoli et al[22]	Switzerland	April 16th to April 30th 2020	4726	9.6%
Shields et al[25]	United Kingdom	April 2020	545	24.4%
Rudberg et al[24]	Sweden	April 2020 to May 2020	2149	19.1%
Self et al[21]	United States	April 2020 to June 2020	3248	6%
Airoldi et al[27]	Italy	May 2020 to June 2020	2252	17.1%
Oliveira et al[20]	Brazil	June 2020	1996	5.5%
Recanatini et al[23]	Netherlands	June 2020 to July 2020	2328	13.9%
Kataria et al[28]	United States	July 2020	1743	5.5%
Prakash et al[26]	India	August 2020	1710	23.65%
Sonmezer et al[30]	Turkey	October 2020	1974	19%
Halili et al[29]	Kosovo	October 2020 to December 2020	647	17.5%
Wiggen et al[31]	United States	November 2020 to February 2021 December 2020 to February 2021	459	9.47% 17.7%
Ige et al[32]	Nigeria	December 2020 to July 2021	413	30.9%
Kanamori et al[18]	Japan	2020	3788	0.3%
		2021		1.6%
		2022		17.7%
Taher et al[33]	Saudi Arabia	June 2022 to September 2022	404	94%

The presence of nucleocapsid (N) antibodies has been taken into account for the seroprevalence across the different time periods (T1 to T5).

As the first wave of the pandemic progressed, seroprevalence rates increased as more HCWs were exposed for a longer time to patients with infection. During the first wave of the pandemic in India, the seroprevalence was reported to be 23.65% in August 2020 in a cohort of 1710 HCWs[26]. The seroprevalence of 34.1% in the current study in October 2020 (T3) is consistent with the trajectory of seroconversion as the pandemic evolved in India. Other centers from across the world reported[27-30] much lower seroprevalence rates ranging from 5.5%[28] to 19%[30] during the period May to December 2020.

The seroprevalence rate in the current study in June 2021 of 60.1% (T5) was much higher than the seroprevalence rates for 2021 from the United States (17.7%, December 2020 to February 2021)[31], and Nigeria (30.9%, December 2020 to July 2021)[32]. Seroprevalence rates in 2022 were 17.7% in Japan (annual checkup of HCWs in 2022)[18] and very high rates of 94% in Saudi Arabia (June 2022 to September 2022)[33].

These results highlight significant regional and temporal disparities in seroprevalence. These variations in seroprevalence rates among HCWs reflect the pandemic's complexity and emphasize factors such as geographic location and vaccination timing. The prevalence of SARS-COV-2 antibodies among HCWs can differ significantly based on local epidemiological conditions and workplace settings.

Our study showed that HCWs working in HR areas were at a higher risk of developing clinical infection during the initial period of the pandemic when vaccination was not available. This can be attributed to the disease's novelty, lack of vaccines, and PPE shortages. Another possible contributing factor could be the significant differences in work duration between HR and LR areas, as illustrated in Table 1. However, clinical infection risks became comparable by the end of the second wave (T5), with over 96% of HCWs in both groups showing seroconversion. This shows the evolving nature of the pandemic and the protective effects of vaccination over time. In our institution, which had around 11000 workers including HCWs and support staff at the time of the pandemic, only 2 HCWs succumbed to COVID-19 infection. Vaccination has a positive impact on the severity of illness and mortality with full vaccination conferring a substantially higher protective effect over partial vaccination[16]. A study by Murhekar et al[34] showed that the seroprevalence status among HCWs and the general population who received two doses of COVID-19 vaccination was comparable (89.8% vs 88.6%) confirming the efficacy of infection control policies and vaccination strategies.

Our study shows that roughly half of the COVID-19 infections among HCWs were symptomatic, highlighting the need to address both symptomatic and asymptomatic HCWs from spreading the virus. These findings align with similar findings in other studies, reinforcing that many infections may present with milder symptoms or be asymptomatic, underscoring the need for vigilant surveillance and infection control measures[35,36]. Although routine screening of asymptomatic HCWs for COVID-19 infection was undertaken in several centers, its role in prevention of transmission of infection to patients is unclear[37]. Lastly, our study reveals that a notable percentage of HCWs were hospitalized, with some admissions being mandated by government directives, which highlights the seriousness of COVID-19 infections and their immense pressure on healthcare systems. Similar trends were seen in other research, underlining the need for prepared healthcare infrastructure and safeguarding frontline HCWs[38].

A few limitations merit mention. This study was conducted in a single center and hence this limits generalizability to other regions. Not all HCWs provided samples at all time points given that some HCWs, particularly postgraduate trainees, joined after the initial sampling or left the institution on completion of their course before study completion. However, it must be highlighted that the proportion of HCWs with clinical infection was similar in the entire cohort and the subgroup on whom seroprevalence data was analyzed. It is likely that the subset reflects the characteristics of the entire cohort of HCWs. Covaxin^® elicits antibodies to nucleocapsid and spike proteins as a natural infection does. This would have made it difficult to differentiate the antibody response to vaccination versus infection (clinical and subclinical). Hence, Covaxin^®-vaccinated individuals were excluded from the seroprevalence analysis.

CONCLUSION

This study demonstrates an increase in seroprevalence to COVID-19 infection among HCWs as the pandemic evolved, from 1.1% at the start of the pandemic to 60.2% by the end of the second wave of the pandemic, reflecting the progression of the pandemic among a population that was immunologically naïve to COVID-19. HCWs working in HR areas were more prone to develop clinical infection in the pre-vaccination period. About 1/3^rd of HCWs had evidence of sub-clinical infection. By the end of the study period, 98.8% were positive for spike antibody suggesting good response to vaccination.