Serology suggests adequate safety measures to protect healthcare workers from COVID-19 in Shiga Prefecture, Japan.

Healthcare workers (HCWs), especially frontline workers against coronavirus disease 2019 (COVID-19), are considered to be risky because of occupational exposure to infected patients. This study evaluated the correlation between seroprevalence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) antibodies among HCWs and the implementation of personal protective equipment (PPE) & infection prevention and control (IPC). We recruited 1237 HCWs from nine public COVID-19-designated hospitals in Shiga Prefecture, central Japan, between 15-26 February 2021. All participants answered a self-administered questionnaire and provided blood samples to evaluate SARS-CoV-2 antibodies. A total of 22 cases (1·78%) were seropositive among the 1237 study participants. An unavoidable outbreak of SARS-CoV-2 had occurred at the terminal care unit of one hospital, before identifying and securely isolating this cluster of cases. Excluding with this cluster, 0·68% of HCWs were suspected to have had previous SARS-CoV-2 infections. Binomial logistic regression from individual questionnaires and seropositivity predicted a significant correlation with N95 mask implementation under aerosol conditions (p = 8.63e-06, aOR = 2.47) and work duration in a red zone (p = 2.61e-04, aOR = 1.99). The institutional questionnaire suggested that IPC education was correlated with reduced seropositivity at hospitals. Seroprevalence and questionnaire analyses among HCWs indicated that secure implementation of PPE and re-education of IPC are essential to prevent SARS-CoV-2 infection within healthcare facilities. Occupational infections from SARS-CoV-2 in healthcare settings could be prevented by adhering to adequate measures and appropriate use of PPE. With these measures securely implemented, HCWs should not be considered against as significantly risky or dirty by local communities.

Introduction

Since the start of the coronavirus disease 2019 (COVID-19) pandemic in the Shiga Prefecture, Central Japan, in April 2020, a total of 2,466 cases were recorded as of February 2021 (Fig 1). The number COVID-19 cases recorded in the prefecture was about half of the mean of cases in Japan, i.e., from April 2020 to February 2021, the average number of COVID-19 cases per 100,000 population was 175 in the Shiga Prefecture and 342 nationwide [1]. During the pandemic, healthcare workers (HCWs) have been treating patients with COVID-19 in hospitals, and have restricted various aspects of their own daily lives, so as to not spread the disease from hospitals to the general population. During the several months of this pandemic, HCWs struggled to treat COVID-19 patients even with the shortage of personal protective equipment (PPE). Meanwhile, the Shiga Prefectural administration had to prepare hospitals designated specifically for COVID-19 rapidly. Regardless of such efforts, HCWs, especially frontline workers designated for COVID-19 areas, named the red zone areas, have been considered to be at increased risk of the disease, owing to their occupational exposure to infected patients. In Japan, the seroprevalence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) was reported to be relatively higher in HCWs [2, 3]. Because of these, the HCWs and their relatives have often been considered against as being dirty or risky according to the Japanese local communities.

Fig 1

Numbers of new and cumulative cases of COVID-19 in Shiga Prefecture.

New and cumulative cases of COVID-19 are indicated as bars and lines, respectively (https://covid2019.fa.xwire.jp/#japan_prefecture). The investigation and blood sampling were performed 15–26 February 2021, at the terminal timing of the third wave of the pandemic in Shiga Prefecture. The numbers and arrows below the graph indicate the estimated cases and timing of HCWs exposed to SARS-CoV-2. Four and 18 cases presumably suffered from SARS-CoV-2 at the second and third waves, respectively. Eight patients were likely exposed during a period of less than three months. Note. COVID-19, coronavirus disease 2019; HCW, healthcare workers; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.

In order to evaluate whether HCWs have been at increased risk of COVID-19, how the PPE usage and infection prevention and control (IPC) guidelines were invaded by SARS-CoV-2 in healthcare facilities, and what should be implemented for better protection of HCWs from the infection, this survey was conducted during 15–26 February 2021, approximately 11 months after the pandemic started to affect Shiga Prefecture. Via the serological surveillance of SARS-CoV-2 antibodies among HCWs, we speculated whether the seroprevalence of HCWs was higher and whether the timing of HCWs’ exposure to SARS-CoV-2 was earlier than those of the general population in Japan. Additionally, via the questionnaires’ analyses from individual HCWs and representatives of healthcare institutions, we evaluated whether HCWs had securely implemented PPE and IPC.

Methods

Study design and participants

A cross-sectional study of HCWs was performed as a prefectural administrative investigation in nine public hospitals designated for COVID-19 in Shiga Prefecture, central Japan. Under the health & infection managements of Shiga Prefecture administration, the red & green zone management and questionnaire investigation were similarly conducted in all the participated hospitals. The investigations took place during 15–26 February 2021, which corresponded to the period between the third and fourth waves of COVID-19 in Japan. The seroprevalence of SARS-CoV-2 antibodies at the timing of this survey was estimated to be about 1–2%, and on that basis, a survey of 1000–2000 participants had been planned in order to allow evaluating at least two variables in a logistic regression analysis. Thus, this survey recruited 1237 HCWs from nine public hospitals. Participation in the study was voluntary. HCWs were invited by advertisements and/or internal announcements to participate in the study. Those interested in the study were asked to contact the study team for an appointment. If the participant had a history of working in the ward designated for COVID-19 patients, he or she was defined as a red zone worker and included in the study. However, to securely assess the risk between the red zone workers and the general (green zone) workers in the hospital workplaces, quota sampling was applied in each hospital. The ratio between the red and green zone workers was approximately 1:1.

Data collection

Each consenting participant was given a self-administered questionnaire (S1 File) to capture the implementation of PPE, adherence with recommended IPC measures, and history of exposure to SARS-CoV-2 during the previous 11 months from April 2020 to February 2021. A representative of each institution was provided with another questionnaire (S2 File) to capture the practices of PPE usage and IPC measures in their respective hospitals.

These questionnaires originated from the protocol ‘Assessment of potential risk factors or 2019-novel coronavirus (2019-nCoV) infection among HCWs in a healthcare setting’, published by the World Health Organization [4] and were designed for HCWs in Japan, referring also to the checklists of Japanese society for infection prevention and control [5] and of national institute of infectious diseases [6]. From each participant, 5 mL of peripheral venous blood was collected for serological testing of SARS-CoV-2 antibodies.

Serological tests of SARS-CoV-2 antibodies

The blood samples were separated by centrifugation, and serum was frozen until antibody evaluation. After all the study samples were collected, the serum samples were defrosted and detection of SARS-CoV-2 antibodies was conducted using the Roche Cobas® 8000 (Roche, Basel, Switzerland) and the Abbott ARCHITECT® i1000SR (Abbott, Chicago, IL) platforms at the departments of Clinical Laboratory Medicine and Pharmacy, respectively, at the Shiga University of Medical Science Hospital. Roche Cobas® was used to measure serum antibodies specific to SARS-CoV-2 nucleocapsid (Elecsys® Anti-SARS-CoV-2 RUO), and to the receptor binding domain (RBD) of spike protein (Elecsys® Anti-SARS-CoV-2 S RUO) [7-11]. Abbott ARCHITECT® assays were used to measure specific immunoglobulin (Ig)M to spike protein (SARS-CoV-2 IgM), IgG to nucleocapsid (SARS-CoV-2 IgG), and IgG to RBD of spike protein (SARS-CoV-2 IgG II Quant) [7, 8, 11, 12]. The measured values adjusted with the manufacturers’ calibrators/standards were interpreted as positives, with a cut-off index (COI) of ≥1·0, ≥0·8 U/mL, ≥1·0, ≥1·4, and ≥50·0 AU/mL, respectively. On the post-infectious Day 14, the sensitivity & specificity of each kit were 100% & 99.81%, 98.8% & 99.98%, 95% & 100%, 100% & 99.63%, and 100% & 99.9%, respectively [13, 14]. Individuals who were serologically positive for antibodies specific to the RBD of spike protein in both Elecsys® Anti-SARS-CoV-2 S RUO (Roche) and ARCHITECT® SARS-CoV-2 IgG II Quant (Abbott) were considered to have been previously infected with COVID-19 during the 11 months prior to entering the study.

Statistical analyses

To securely assess the risk between red and green zone workers in the hospitals, chi-square or Fisher’s exact tests were applied. To evaluate whether the questionnaire responses for each individual HCW were predictive of SARS-CoV-2 antibody seropositivity, analyses of binomial logistic regression following univariate correlation were performed from each question in the questionnaire to the seropositivity. Among variables, excluding those with variance inflation factor (VIF) >10 from the model due to the risk of multicollinearity, and including those indicated with p<0.05 by Fisher’s exact test, binomial logistic regression was initially performed using the stepwise variable reduction method using p-Value. Individual questionnaires mainly included questions for the implementation of PPE and adherence to recommended IPC measures. In this cohort, referring to the seroprevalence and number of participants, the confirmatively regression analysis was conducted with only 2 variables, N95 mask implementation under possible aerosol conditions and working period in the red zone. To evaluate whether the questionnaires from each institute’s representative were predictive of hospitals where there were seropositive HCWs, that is, where the protection barrier breakage had incidentally occurred in the hospitals designated for COVID-19, univariate and Spearman’s rank correlation coefficient analyses were applied to the questions in the institutional questionnaire. The institutional questionnaire mainly asked representatives about the practical implementation of PPE and IPC educational programs. In this study, only 9 institutes were included, and both the seropositive case numbers and rates of each institute were out of normal distribution through Shapiro-Wilk normality tests (p = 6.66e-05 and 7.80e-05, respectively). Thus, each institutional difference was evaluated using the Kruskal-Wallis test followed by Holm’s post-hoc significance test. Fisher’s exact test and Spearman’s rank correlation coefficient were also conducted to select and evaluate correlating factors in the institutional questionnaire to the seropositive hospital cases. The analysis was performed using Easy R software version 4·1·0 [15].

Ethical statement

Written informed consent were obtained from all study participants. During the process of obtaining consent, all participants were informed of the need to publish the results. All participants’ personal information was anonymised during the write-up, and no participant is identifiable in the publication. This study was approved as an administrative investigation of Shiga Prefecture by the Research Review Board of Shiga University of Medical Science (No. RRB20-032). Findings from the study have been disseminated to study participants, HCWs, through a newsletter and each institute representative that represent HCWs’ communities.

Results

Seroprevalence of total antibodies against SARS-CoV-2 and the presumed exposure history

The study population of 1237 HCWs comprised 257 (20·8%) medical doctors, 817 (66·0%) nurses, 67 (5·4%) office workers, and 96 others (7·8%) across the nine study hospitals. The other characteristics of the present study participants couldn’t be precisely clarified, because individual questionnaire didn’t include such questions. However, the second serological survey of HCWs was similarly conducted December 2021 in Shiga prefecture. In the second survey, 1600 HCWs was characterized with age of 42.1 ±16.4 (indicating mean ±S.D. below) years old, body mass index of 23.4 ±8.7 kg/m2, and 66.2% of female predominance. Thus, the present study was presumably composed with similar character population.

A total of 22 out of 1237 HCW samples were categorised as serologically positive, meaning that 22 HCWs had been previously infected with COVID-19 during the previous 11 months. Although IgM specific to the spike protein can only be detected 2–3 months post infection, the IgG counterpart is continuously detected between 8 months and 1 year from the initial infection of SARS-CoV-2 [9, 11, 13, 16–20]. IgG specific to the SARS-CoV-2 nucleocapsid shows serological positivity for approximately 3–6 months [8, 20–22]. Using these information, we approximately estimated the timing of exposure to SARS-CoV-2 in 22 serologically positive cases. Thus, four HCWs with serological detection of only IgG specific to the spike protein were infected with COVID-19 during the second wave of the pandemic in Shiga Prefecture, and the other 18 individuals were likely infected during Shiga Prefecture’s third wave (Fig 1 and S1A Fig). Eight of the 18 individuals with serologically detected IgM specific to the spike protein, in addition to the nucleocapsid-specific IgG and the spike protein-specific IgG (S1B Fig), were likely exposed to SARS-CoV-2 1–3 months prior to this investigation. The proportion of COVID-19 cases in the second and third waves in Shiga Prefecture was similar for HCWs and the general public (4 vs. 18, and 450 vs. 2011 cases, respectively; Fig 1).

Risk comparison between red zone and green zone hospital workers

The total seroprevalence of SARS-CoV-2 antibodies was 1·78% (22/1237 cases). Twenty individuals out of 611 (3·27%) from the red zone and two out of 626 (0·32%) from the green zone were previously infected with COVID-19, with the difference between both zones being highly significant (red zone, p = 0·0000485; Table 1). However, one hospital indicated the highly significant prevalence of SARS-CoV-2 positive cases in comparison with the other eight hospitals (Kruskal-Wallis test followed by Holm’s post-hoc significance, p<0·0001; S2 Fig). In fact, an inevitably silent outbreak of SARS-CoV-2 had occurred among patients with primary and metastatic lung cancers at the terminal care unit of one hospital. In this specific hospital, the HCWs could not completely implement their PPE protocols before identifying and securely isolating this cluster of cases. Therefore, there was an abnormally high prevalence of HCWs with SARS-CoV-2 antibodies at one of the study’s hospitals. Thus, in evaluating the general seroprevalence of HCWs in Shiga Prefecture, we additionally considered another model excluded this specific hospital (n = 212 HCWs) and evaluated the risks of 500 red zone workers compared with those of 525 green zone workers, and found that the general seroprevalence of SARS-CoV-2 antibodies was 0·68% (7/1025 cases) in the model, and that the exposed risks of SARS-CoV-2 were not highly significant in red zone workers in Shiga Prefecture (p = 0·0633; Table 1).

Table 1

Risk comparison between red zone and green zone workers in Shiga Prefecture.

Working zone		Antibody against SARS-CoV-2		Total
		Negative (-)	Positive (+)	n	p-Value
	Red	591	20	611	4.85e-5
	Green	624	2	626
	General seroprevalence 1·78% *
	Red	494	6	500	0.0633
	Green	524	1	525
	General seroprevalence 0·68% #

Note. SARS-CoV-2, severe acute respiratory syndrome coronavirus 2; Models including (*) or excluding (#) with a hospital which the cluster happened.

Secure implementation of PPE was required for protecting HCWs against COVID-19

Univariate analysis (Table 2) between SARS-CoV-2 seropositivity and individual characteristics of the self-administered questionnaire indicated possible correlations between HCWs with close contact with COVID-19 patients, close contact with the other workers in the hospital, everyday changing of the hospital uniform, working experience in the red zone section, and everyday changing of the red zone uniform. However, binomial logistic regression analysis among such significant variables could indicate only 3 factors; infected history of COVID-19 (adjusted odds ratio [aOR] 319; 95% confidence interval [CI] 22·2–4600; p = 0·0000227), N95 mask implementation under possible aerosol conditions (aOR 2·69; 95% CI 1·61–4·49; p = 0·000148), and working period in the red zone (aOR 2·06; 95% CI 1·04–4·08; p = 0·0377), as highly significant factors of SARS-CoV-2 seropositivity (Table 2). In this cohort, considering the seroprevalence and number of participants (1·78% and 1237 HCWs, respectively) and also the predictively important factors, we excluded the previously infected history of COVID-19, and conducted the subsequently binomial logistic regression analysis with only 2 variables, N95 mask implementation and working period in the red zone, whose aOR indicated 2·47 and 1·99 (p = 8·63e-06 and 2·61e-04; Table 3), respectively. In the confirmative model using only these 2 factors, these VIFs indicated 1.21 and 1.21, respectively, in which case multicollinearity was likely very little. In the model, the area under the receiver operating characteristic curve (AUC) was 0·807 (95% CI 0·707–0·907; Fig 2).

Table 2

Individual characteristics and seroprevalence of SARS-CoV-2.

Feature		Antibody against SARS-CoV-2			Binomial Logistic Regression
		Negative (-)	Positive (+)	p-Value (Fisher’s exact)	aOR	(95% CI)	p-Value
Red-zone working period: Days
<1 week		395	1	3.56e-6	2.06	(1.04–4.08)	3.77e-2
1<2 weeks		272	0
2<3 weeks		98	6
3<4 weeks		81	4
4< weeks		254	8
Previously suffered COVID-19: Infected History
No infection		1198	14	1.24e-12	319	(22.2–4600)	2.27e-5
Infected		4	8
Close contact to patients: Pt_Contact
No contact		543	5	0.0289	1.47	(0.283–7.64)	0.646
Close contact		588	16
N95 mask implementation under aerosol situations: Q5
No situation		332	1	3.78e-4	2.69	(1.61–4.49)	1.48e-4
Applied:	1	716	12
	2	59	1
	3	31	0
Non-applied:	4	67	7
Never talked with others in the hospital: Q12
Applied:	1	1069	18	0.0451	0.751	(0.268–2.10)	0.585
2	71	0
3	37	3
Non-applied:	4	31	1
Everyday changing the hospital uniform: Q29
Applied:	1	371	6	0.0205	0.948	(0.540–1.66)	0.851
	2	205	9
	3	140	3
Non-applied:	4	422	3
Everyday changing the red-zone uniform: Q43
Applied:	1	403	8	1.32e-4	1.30	(0.670–2.53)	0.435
	2	62	10
	3	37	1
Non-applied:	4	88	1

Note. SARS-CoV-2, severe acute respiratory syndrome coronavirus 2; COVID-19, coronavirus disease 2019

aOR, adjusted odds ratio; CI, confidence interval.

Table 3

The subsequently binomial logistic regression analysis with 2 variables.

Feature		Antibody against SARS-CoV-2			Binomial Logistic Regression
		Negative (-)	Positive (+)	p-Value (Fisher’s exact)	aOR	(95% CI)	p-Value
Red-zone working period: Days
<1 week		395	1	3.56e-6	1.99	(1.38–2.89)	2.61e-04
1<2 weeks		272	0
2<3 weeks		98	6
3<4 weeks		81	4
4< weeks		254	8
N95 mask implementation under aerosol situations: Q5
No situation		332	1	3.78e-4	2.47	(1.66–3.67)	8.63e-06
Applied:	1	716	12
	2	59	1
	3	31	0
Non-applied:	4	67	7

Note. SARS-CoV-2, severe acute respiratory syndrome coronavirus 2; COVID-19, coronavirus disease 2019

aOR, adjusted odds ratio; CI, confidence interval.

Fig 2

Area under the receiver operating characteristic curve (AUC) for the prediction of SARS-CoV-2 seropositivity of healthcare workers.

The model was composed of only two factors of N95 mask implementation under possible aerosol conditions and working period in the hospital red zone section. AUC was 0·807 (95% CI 0·707–0·907). Note. AUC, area under the receiver operating characteristic curve; CI, confidence interval.

Re-education of IPC were suggested to protect the hospital barrier against SARS-CoV-2

SARS-CoV-2 seropositive cases were detected in four of the nine hospitals investigated. These four hospitals were compared to the other five in terms of questions on the institutional questionnaire. Next, among the 54 questions for practical implementation of IPC and PPE usage, only re-education practice of IPC for HCWs suggested a highly significant correlation to protect the hospital barrier against COVID-19 (Table 4 and S3 Fig). In addition to Fisher’s exact test, IPC re-education was strongly correlated with SARS-CoV-2 HCW seropositivity cases at each hospital on Spearman’s rank correlation coefficient analysis (ρ = 0.949, p = 0.0000958; S3A Fig). Additionally, together with the numbers of both participating individuals in this survey and hospitalised COVID-19 patients in each hospital, multiple regression analysis of IPC re-education was performed to predict an association with SARS-CoV-2 seropositive cases (S3B Fig).

Table 4

Institutional comparison for SARS-CoV-2 seropositivity of hospital.

Feature	Seropositivity of hospital		Fisher’s exact test
	Negative (-)	Positive (+)	p-Value
IPC re-education: Q50
Performed (Yes)	5	0	7.94e-3
Not (No)	0	4

Note. SARS-CoV-2, severe acute respiratory syndrome coronavirus 2; IPC, infection prevention and control.

Discussion

This study has indicated that red zone HCW with adequate implementation of PPE and IPC was not a highly significant risk of COVID-19, and should not have been considered against. If HCWs had had an increased risk of SARS-CoV-2 exposure, they would have been affected by COVID-19 earlier or more than the general population. In fact, SARS-CoV-2 seroprevalence and the presumed timing of those infected were similar to those of the general population (Fig 1), even though the PPE shortage caused HCWs to struggle throughout several months of the pandemic in Shiga Prefecture. SARS-CoV-2 seroprevalence in HCWs in Shiga Prefecture (1·78–0·68%) was similar to that of the general population in December 2020 across various parts of Japan (Tokyo 1·35%, Aichi 0·71%, Osaka 0·69%, Fukuoka 0·42%, and Miyagi 0·14%) [23]. Theoretically calculated from each prefecture’s population number and antibody prevalence, the seropositive population numbers of SARS-CoV-2 had been approximately 3–5 times more than the number of COVID-19 cases diagnosed by polymerase chain reaction (PCR) and/or antigen tests at that time [1]. From these theoretical numbers and together with PCR and/or antigen-diagnosed COVID-19 numbers in Shiga Prefecture, we were able to calculate the seroprevalence rate of the general population of Shiga Prefecture at the time of our investigation. In doing so, a hypothetical 1–0·3% prevalence was calculated in the general population, and the seroprevalence of HCWs (1·78–0·68%) was not so higher than that of the general population in the Shiga Prefecture. In addition, HCWs’ seroprevalence in the Shiga Prefecture wasn’t so higher than that of another prefecture’s hospital workers (1·1%) in the same time frame of February to April 2021 [24]. In Japan, previous investigations had reported that the seroprevalence was higher in HCWs [2, 3], and the data may have misled the local communities into recognizing the HCWs and their relatives as being significantly dirty or risky. However, at least in Shiga Prefecture, occupational infections from SARS-CoV-2 in healthcare settings weren’t so higher than those of the generals, and we believe that HCWs should not be considered against as significantly risky or dirty by local communities.

As a matter of course, recognition of previous COVID-19 infection, that is, any symptomatic and/or diagnosed history of COVID-19 prior to the survey, indicated a maximum aOR to SARS-CoV-2 seropositivity (Table 2). However, about two-thirds of the seropositive cases (14/22 cases) could not recognise any symptoms or diagnoses in their previous histories. This reflects the difficulty of dealing with the silent invasion of SARS-CoV-2 into healthcare facilities. In all medical institutes, HCWs were checked every day for their subjective symptoms, such as fever, cough, sore throat, general malaise, dyspnoea, and abnormal taste or smell, and were under adequate IPC. However, it may be difficult to completely prevent SARS-CoV-2 invasion even in hospitals. In addition to securing IPC, if hospitals are not able to perform regular screening of HCWs using PCR and/or antigen tests, silent outbreaks of SARS-CoV-2 may incidentally occur even in healthcare facilities. In fact, among investigated hospitals, 15 cases of HCWs could not implement their PPE protocols before identifying and securely isolating SARS-CoV-2 cases at the terminal care unit of lung cancers, and an unavoidable outbreak occurred. This conformed that the greatest risk to HCWs may be their own colleagues or patients in the early stages of unsuspected infections rather than red zone working [25]. It’s also in line with that HCWs working in general, ophthalmology, and respiratory departments were prone to risk compared with HCWs working in the infection department [26].

Apart from recognition of previous COVID-19 infections, the binomial logistic regression analysis from the individual questionnaire to SARS-CoV-2 seropositivity identified two highly significant factors: N95 mask implementation under possible aerosol conditions and working period (in days) in the red zone section. Among PPE, N95 mask implementation should be especially required under possible aerosol conditions in hospitals. Even if red zone HCWs could be safely protected with PPE and IPC, the working duration in the red zone should be as distributed as far as possible. This is comparable hat more prolonged contact with COVID-19 patients remains a crucial risk factor for SARS-CoV-2 [27]. In order to reduce the opportunities of SARS-CoV-2 exposure, we should shorten the working periods in the red zone for HCWs, as far as possible. The efforts for two highly significant aspects should be continuously performed in healthcare facilities.

The analysis from the institutional questionnaire suggested a correlation between practical re-education of IPC for HCWs and SARS-CoV-2 seropositive hospital cases. There were few participating hospitals in this investigation, and we could not indicate the absolute significance of IPC re-education, which is possibly most important for implementing IPC and PPE. If the number of individuals participating in this survey and/or hospitalised COVID-19 patients in each hospital were to increase, SARS-CoV-2 seropositive HCWs could be increased in each hospital. Therefore, we performed multiple regression analysis of IPC re-education with these numbers to predict an association with SARS-CoV-2 seropositive cases (S3B Fig). The analysis indicated a weak association with IPC re-education even without its statistical significance, so IPC re-education may still give each hospital the capacity to accept more than 100 hospitalised COVID-19 patients.

Although this study provides interesting correlation between SARS-CoV-2 seroprevalence and the implementation of PPE & IPC in HCWs, it is important to note that there are some limitations. The survey was performed before alpha-variant predominant expansion in Japan, and the results might not reflect that in the predominant period of alpha, delta, or omicron variant, etc. Nevertheless, the implementation procedure of PPE and IPC is likely similar to that of the original SARS-CoV-2, so the study data will contribute to the preventive measures of the present and future variants in healthcare facilities. Meanwhile, to evaluate the statistical correlation to multi-variables, the sample size of this study might not be enough. The cross-sectional nature of this study also doesn’t allow to obtain conclusive causal evidence. Thus, referring to the seroprevalence and number of participants, we have focused on 2 factors, N95 mask implementation under possible aerosol conditions and working period in the red zone, in this correlation analysis of Shiga Prefecture, central Japan. The occupational infection of SARS-CoV-2 in healthcare settings could be prevented by adherence to adequate measures and appropriate use of PPE like N95 mask. Still, the working period in the red zone should be as distributed for HCWs as far as possible, to reduce the occupational opportunities of SARS-CoV-2 exposure. The analysis also indicated an association between SARS-CoV-2 seropositivity and IPC re-education, so practical re-education of IPC for HCWs may contribute to accept hospitalised COVID-19 patients into hospitals. The study findings should be confirmed with more cases of participants, and PPE & IPC strategy should be adapted to predominant variants of SARS-CoV-2 in future. Nevertheless, this survey indicated that red zone HCWs with adequate implementation of PPE and IPC were not at high risk of SARS-CoV-2 exposure, and HCWs should not be considered against as being risky or dirty by local Japanese communities.

Conclusions

This study indicated that secure implementation of PPE and re-education of IPC were essential to prevent SARS-CoV-2 infection within healthcare facilities. HCWs with adequate implementation of PPE and IPC should not be considered against as significantly risky or dirty by local communities.

Serological values of SARS-CoV-2 antibodies in 22 positive cases.

(TIFF)

Click here for additional data file.

Comparison for SARS-CoV-2 seropositivity of healthcare workers among nine hospitals designated for COVID-19 in Shiga Prefecture.

(TIFF)

Click here for additional data file.

Infection prevention control re-education was likely correlated with SARS-CoV-2 seropositivity of healthcare workers at each hospital.

(TIFF)

Click here for additional data file.

Individual questionnaire.

(XLSX)

Click here for additional data file.

Institutional questionnaire.

(XLSX)

Click here for additional data file.

Individual data in Shiga Prefecture.

(XLSX)

Click here for additional data file.

Institute data in Shiga Prefecture.

(XLSX)

Click here for additional data file.

5 Jan 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Gabriel O Dida, PhD

Academic Editor

PLOS ONE

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf  and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match.

When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

3. Thank you for stating the following in the Funding and Acknowledgments Section of your manuscript:

“Funding:

The study cost was mainly funded as an administrative investigation by the Shiga Prefecture Governor. Laboratory tests were partly funded from Shiga University of Medical Science Hospital budget. The fund numbers are not applicable. The funders did not have any role in the study design, data collection, data analysis, interpretation, or report writing.

Acknowledgement:

We would also like to thank Shiga Prefecture Governor, Taizo Mikazuki, for his funding decision and permission to publish this study”

We note that you have provided information within the Funding and Acknowledgements Section. Please note that funding information should not appear in the Acknowledgments section or other areas of your manuscript. We will only publish funding information present in the Funding Statement section of the online submission form.

Please remove any funding-related text from the manuscript and let us know how you would like to update your Funding Statement. Currently, your Funding Statement reads as follows:

“No - The study cost was mainly funded as an administrative investigation by the Shiga Prefecture Governor. Laboratory tests were partly funded from Shiga University of Medical Science Hospital budget. The fund numbers are not applicable. The funders did not have any role in the study design, data collection, data analysis, interpretation, or report writing.”

Please include your amended statements within your cover letter; we will change the online submission form on your behalf.

4. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide.

5. We note that you have included the phrase “data not shown” in your manuscript. Unfortunately, this does not meet our data sharing requirements. PLOS does not permit references to inaccessible data. We require that authors provide all relevant data within the paper, Supporting Information files, or in an acceptable, public repository. Please add a citation to support this phrase or upload the data that corresponds with these findings to a stable repository (such as Figshare or Dryad) and provide and URLs, DOIs, or accession numbers that may be used to access these data. Or, if the data are not a core part of the research being presented in your study, we ask that you remove the phrase that refers to these data.

6. Your ethics statement should only appear in the Methods section of your manuscript. If your ethics statement is written in any section besides the Methods, please delete it from any other section.

7. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: General comments

Dear authors, thank you for the opportunity to read your work. This paper investigates the possibility that HCWs are a source of risk for the population given their constant exposure to COVID-19 patients. Preventive safety measures are also investigated. The survey was conducted between 15 and 26 February 2021 in Shiga, Japan. Questionnaires, blood sample collection and regressive and correlational analyzes were used to find a conclusion. However, at present, I believe that there are some critical issues to be addressed.

=============

Major comments

1) Abstract. “From these, binomial logistic regression from individual questionnaires and seropositivity predicted a significant correlation with N95 mask implementation under aerosol conditions and work duration in a red zone.” Please, specify P-value ​​and the regression coefficient.

2) Methods.

2.1. “Binomial logistic regression analysis was applied from the individual questionnaire to SARS-CoV-2 seropositivity [...]” The use of logistic regression models requires the verification of some assumptions (https://www.statisticssolutions.com/free-resources/directory-of-statistical-analyses/assumptions-of-logistic-regression/, https://pubmed.ncbi.nlm.nih.gov/21996075/#:~:text=Basic%20assumptions%20that%20must%20be,lack%20of%20strongly%20influential%20outliers). Please, detail in this section how these have been verified.

2.2. “Each institutional difference was evaluated using the Kruskal-Wallis test followed by Holm’s post-hoc significance test.” and “[...] Fisher’s exact test and Spearman’s rank correlation coefficient were conducted to select and evaluate correlating factors in the institutional questionnaire to the seropositive hospital cases.” Generally, the use of non-parametric measures is appropriate when the dataset is not normally distributed (In some cases, some authors support the adoption of parametric tests even for non-normal datasets, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3445820/). On the other hand, other authors point out that the difference in power between parametric and non-parametric tests is small (e.g., https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2743502/). Other approaches have also been proposed (e.g., https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1310536/). Therefore, I suggest that the authors briefly address this aspect in the manuscript, motivating their choices to use non-parametric tests. In this regard, the verification of the shape of the datasets through tests for normality (e.g., Shapiro-Wilk + Q-Q plots) could be conclusive.

2.3. Dear authors, I ask if the possibility to test for internal consistency of questionnaires has been evaluated.

3) Results.

3.1. P-values ​​should be used, at best, as graded measures of the strength of evidence against the null hypothesis (https://pubmed.ncbi.nlm.nih.gov/28698825/, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4877414/). Therefore, the adoption of a simple threshold can be misleading. Hence, I suggest speaking of "more and less significant" or "high and low significance" rather than "significant and non-significant."

3.2. Since P-values measure the statistical significance but not the effect size, I suggest commenting on the results also based on the intensity of the phenomenon (e.g., strong, moderate, weak associations, etc.)

4) Discussion.

4.1. I suggest making a comparison with the literature published on this topic (also relating to other countries). This serves to contextualize and better understand the relevance of the results found (e.g., https://pubmed.ncbi.nlm.nih.gov/33115772/, https://pubmed.ncbi.nlm.nih.gov/33003634/, https://pubmed.ncbi.nlm.nih.gov/33140084/).

4.2. I suggest clearly specifying the limitations of the study. In particular, i) the sample size prevents these results from being generalizable, and ii) the fundamentally cross-sectional nature of the study and the search for correlations do not allow to obtain conclusive causal evidence.

=============

Minor comments

m1) References. Regarding websites, I suggest specifying the name of the source and providing an access date.

Reviewer #2: Reviewer’s report

Title: Serology suggests adequate safety measures to protect healthcare workers from

COVID-19 in Shiga Prefecture, Japan

Version: Date: 9th December 2021

Reviewer: Samuel Bosomprah

Reviewer’s report

The authors stated that this was a seroprevalence study aimed “to estimate the timing of HCWs’ exposure to SARS-CoV-2 and to analyse whether the HCWs had been exposed to SARSCoV-2 earlier than the general population as well as to determine whether the seroprevalence of HCWs was higher than that of the general population in Japan”

General comment:

The manuscript did not read well as a research document. The language is not consistent with a research paper write up. For example, the authors used terms such as “noted” in the introduction instead of, say, cases were “recorded” etc. Another term was “clarify” in method section “To clarify whether the questionnaire responses for each individual HCW were predictive of SARS-CoV-2 antibody seropositivity…” and “To clarify whether the questionnaires from each institute’s representative were predictive of hospitals where there were seropositive HCWs…” These are not appropriate statistical terminologies.

Introduction

• I would have expected the authors to begin the introduction with clear statement of the problem before elaborating on it in terms of the magnitude of the problem and which group is unfairly or disproportionately affected.

• The authors did not provide compelling argument for why the study is important (i.e., justification/significance/rationale) and what the knowledge/research gap is, which they are seeking to fill in

Methods

• The method section is not well structured. The authors should consider renaming “study setting” as “Study design and participants” and integrate the “study participants” texts.

• Authors should avoid the phrase “…investigation prospectively recruited..” for a cross-sectional study design

• Authors should consider integrating the subheadings “Evaluation of individual personal protective equipment (PPE)” and “Evaluation for institutional infection prevention and control (IPC)” into “Data analysis” and edit the language to read well.

• Authors did not provide the assumptions on arriving at a sample size of 1237. They should consider including a section on “sample size consideration”, which should describe how the sample size of 1237 was arrived at.

• The section “Patient and public involvement” should be renamed as “Ethical statement” and should be edited accordingly.

Findings

• “Findings” should be edited to “Results”

• The analysis and presentation of results did not meet the statistical rigour required of a seroprevalence study. The tables were poorly presented – only frequencies were presented; no corresponding proportions (seroprevalence) were presented (See tables 1, 2 and 3). P-values were inappropriately presented. It is recommended to present p-values to 3 decimal places.

Discussion

• The authors were inappropriately referencing tables and figure as if they are writing up results

• The authors should focus on interpreting the results instead of rehashing same in the discussion.

• Should discuss the strength and limitations of the study.

• They should also discuss the impact of the findings as well as prescription of future work.

• No apparent conclusion

Reviewer #3: - Despite the fact that the blood sample was collected in prospective manner, the seropositivity/ seroprevalence detection is actually (or at least what it is believed to) a measure of past historical record of covid exposure. In the absence of any baseline blood samples, a strong assumption that exposure was solely determined by the seropositivity is needed. This along with its justification should be made clear in the article. The interpretation of such data should be done carefully. Do we have any supporting evidence on how the seropositivity behaves over time? Can exposure to covid19 be perfectly covered by the seropositivity?

- Related to the above question, what is the sensitivity and specificity of the kit used wrt to past covid exposure? This should be explained in the article.

- Covid19 may have a big impact on how the disease spreads and hence in determining the green or red area. These variants are also time dependent. Can this be taken into account in the current model? If not, how do you explain the validity of the model when lacking such data?

- Removal of the terminal care unit of one hospital in which the outbreak took place from the analysis is problematic and potentially raises some questions on the definition used to determine what is the red zone in this research. If the removal is done post-hoc, i.e. upon collecting and inspecting the data, this may potentially introduce some biases at the very least. If an outbreak disqualified a red zone hospital from analysis, what will constitute a red-zone? Has this disqualification of data being defined in the research protocol or SAP prior to conducting the analysis?

- Secure implementation of PPE was required …, Pg 10 and 11, and table 1: it is not clear whether the fitted binomial logistic regression analysis was a multivariable model? If so this should be made clear and should be explained what method was used to retain those 3 variables. Was it simply variables retained in the model which had significant p-values? Was there kind or model selection such as backward, forward or stepwise attempted?

- The article should highlight some restrictions of the study as highlighted above. This should be done in the discussion.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Alessandro Rovetta

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: Reviewer_comment.docx

Click here for additional data file.

5 Apr 2022

Response to Reviewers

Reviewer #1

Major comments

1) Abstract.

Please specify P-value and the regression coefficient.

Following your suggestion, we’ve added p- & beta-Values of each variable, additionally into Table 2-3.

2) Methods.

2.1. The use of logistic regression models requires the verification of some assumptions.

As you have suggested, if the seroprevalence has been 1-2% and we've analyzed 8 variables, the logistic regression will require about 4000-8000 cases. Thus, binomial logistic regression was initially performed using the stepwise variable reduction method using p-Value. However, in this cohort, referring to the seroprevalence and number of participants (1.78% and 1237 cases), the confirmatively regression analysis was conducted with only 2 variables, N95 mask implementation and working period in the red zone (Page 7, Line 137 - 141).

This study limitation has also been described in the revised manuscript (Page 18, Line 371 - 382).

2.2. I suggest that the authors briefly address this aspect in the manuscript, motivating their choices to use non-parametric tests.

Only 9 institutes were included in this study, meaning small number. In addition, both the seropositive case numbers and rates of each institute were out of normal distribution through Shapiro-Wilk normality test (p=0.00006657 and 0.00007803), respectively. Thus, in order to evaluate the difference and questionnaire of each institution, we applied nonparametric measures such as the Kruskal-Wallis test and Spearman’s rank correlation coefficient, respectively, to the analyses.

These are described in the revised manuscript (Page 7 - 8, Line 141 - 147).

2.3. Dear authors, I ask if the possibility to test for internal consistency of questionnaires has been evaluated.

Under the health & infection management of Shiga prefecture administration, the zone management and questionnaire investigation were similarly conducted in all the participated hospitals (Page 4, Line 80 - 83).

In addition, these questionnaires were designed, referring also to the checklists of Japanese society for infection prevention and control and of national institute of infectious diseases (Page 5, Line 99 - 101).

3) Results.

3.1. I suggest speaking of "more and less significant" or "high and low significance" rather than "significant and non-significant."

3.2. I suggest commenting on the results also based on the intensity of the phenomenon.

Following your suggestion, we have modified our descriptions into "highly significant" and "weak association," etc.

4) Discussion.

4.1. I suggest making a comparison with the literature published on this topic.

Based on your suggestion, we’ve compared with the other literatures and produced more fruitful discussion (Page 16 -17, Line 344 - 350 and 357 - 358).

4.2. I suggest clearly specifying the limitations of the study.

According to your suggestion, we have described some limitations of this study, and produced a more balanced work (Page 18, Line 371 - 389).

Minor comments

m1) References. Regarding websites, I suggest specifying the name of the source and providing an access date.

According to your suggestion, we have added the information into References.

Reviewer #2

General comment: The language is not consistent with a research paper write up.

Base on your suggestion, we have used adequate terminologies in the sections of “Introduction” and “Methods”.

Introduction

• The authors to begin the introduction with clear statement of the problem.

• The authors did not provide compelling argument for why the study is important

Considering your suggestion, we have reconstructed the “Introduction” section better.

Methods

• “Study design and participants”

• Authors should avoid the phrase “…investigation prospectively recruited...”

• “Data analysis”

• “Ethical statement”

Following your suggestion, we have reconstructed the “Methods” section better.

• Authors did not provide the assumptions on arriving at a sample size of 1237.

We appreciate your meaningful suggestion. As you have suggested, if the seroprevalence has been 1-2% and we've analyzed 8 variables, the logistic regression will require about 4000-8000 cases. Thus, binomial logistic regression was initially performed using the stepwise variable reduction method using p-Value. However, in this cohort, referring to the seroprevalence and number of participants (1.78% and 1237 cases), the confirmatively regression analysis was conducted with only 2 variables, N95 mask implementation and working period in the red zone (Page 7, Line 137 - 141).

In addition, the study limitation has been described in the revised manuscript (Page 18, Line 371 - 382).

Findings

• should be edited to “Results”

• Tables were poorly presented – only frequencies were presented; no corresponding proportions (seroprevalence) were presented (See tables 1, 2 and 3). P-values were inappropriately presented. It is recommended to present p-values to 3 decimal places.

We appreciate your meaningful suggestion, and have reconstructed all the tables.

Discussion

• The authors were inappropriately referencing tables and figure as if they are writing up results

• The authors should focus on interpreting the results instead of rehashing same in the discussion.

• Should discuss the strength and limitations of the study.

• They should also discuss the impact of the findings as well as prescription of future work.

• No apparent conclusion

Base on your fruitful suggestions, we have reconstructed the “Discussion” section better. We have also described the study limitations and prescription in future (Page 18, Line 371 - 382 and 387 - 389).

We’ve added apparently the “Conclusion” section in the revised manuscript.

Reviewer #3

- What is the sensitivity and specificity of the kit used to past covid exposure?

In this context, referring to the manufacturer's’ specification documents and previously reports, we have clarified them in the “Method” section (Page 6, Line 116 - 118).

- Do we have any supporting evidence on how the seropositivity behaves over time? Can exposure to covid19 be perfectly covered by the seropositivity?

- How the seropositivity behaves over time? Can exposure to covid19 be perfectly covered by the seropositivity?

In a healthy adult population (, not in children), the seropositivity has almost completely reflected the previous infections. Additionally, the time courses of each antibody titer decay have been in line with the previous reports. In the revised version, we have described them in the “Result” section (Page 9, Line 164 - 167).

- Covid19 variants are also time dependent. Can this be taken into account in the current model? If not, how do you explain the validity of the model when lacking such data?

The survey was performed before alpha-variant predominant expansion in Japan, and the results might not reflect that in the predominant period of alpha, delta, or Omicron variant, etc. Nevertheless, the implementation procedure of PPE and IPC is likely similar to that of the original SARS-CoV-2, so the study data will contribute to the preventive measures of the present and future variants in healthcare facilities.

We have described this issue in the study limitations (Page 18, Line 371 - 382).

- Removal of the terminal care unit of one hospital in which the outbreak took place from the analysis is problematic.

Following your suggestion, we have evaluated even both models included with or without the specific hospital where the outbreak took place (Table 1; Page 10, Line 194 - 205).

- It is not clear whether the fitted binomial logistic regression analysis was a multivariable model? If so this should be made clear and should be explained what method was used to retain those 3 variables. Was it simply variables retained in the model which had significant p- values? Was there kind or model selection such as backward, forward or stepwise attempted?

As you suggested, we clarified that binomial logistic regression was initially performed using the stepwise variable reduction method using p-Value. In this cohort, referring to the seroprevalence and number of participants (1.78% and 1237 cases), the confirmatively regression analysis was conducted with only 2 variables, N95 mask implementation and working period in the red zone.

These are described in the "Statistical analysis” of the “Method” section (Page 7, Line 137 - 141).

- The article should highlight some restrictions of the study as highlighted above. This should be done in the discussion.

As you suggested, we have highlighted the study limitations in the “Discussion” section (Page 18, Line 371 - 382).

Submitted filename: PtoP_PLosOne.docx

Click here for additional data file.

10 May 2022

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Gabriel O Dida, PhD

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide

Please reconsider the use of the term "discrimination" and the presentation of your conclusions. This term implies the unjust or prejudicial treatment, and your conclusions appear to suggest (by inverse of their arguments) that it would be appropriate to subject healthcare workers to such prejudicial treatment had occupational infections been detected.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: (No Response)

Reviewer #2: (No Response)

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: No

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: No

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Dear authors, thank you for your revisions and responses. Almost all points have been adequately addressed. However, I suggest one final important check.

1) Logistic regression assumptions are still not discussed, e.g., little or no multicollinearity, linearity of independent variables and log odds, and others (please, see the references I suggested in the previous round). This aspect is essential for validating the statistical results and should be added to the publication (this could be done in a supplementary file as well). In particular, it is necessary to test all the assumptions and report the results to consent the reader to interpret them independently. Thank you.

Reviewer #2: Reviewer’s report

Title: Serology suggests adequate safety measures to protect healthcare workers from

COVID-19 in Shiga Prefecture, Japan

Version 2: Date: 30th April 2022

Reviewer: Samuel Bosomprah

General comment:

The authors addressed some of my earlier comments. But there remains some major work especially presentation of results and tables. The specifics are stated below

Methods

• Authors should state the eligibility (inclusion and exclusion) criteria under the subsection “Study design and participants”

• “Data analyses” and Statistical analysis” sections should be integrated into one subheading called “Statistical analysis”.

• Authors did not provide the assumptions on arriving at a sample size of 1237. They should consider including a section on “sample size consideration”, which should describe how the sample size of 1237 was arrived at.

Results

• The results section should start with the subheading “Characteristics of study participants”. This section should have a table and comments describing who these participants were: example, distribution of sex, age, and other sociodemographic characteristics…

• The analysis and presentation of results did not meet the statistical rigour required of a seroprevalence study. The tables were poorly presented. P-values were inappropriately presented. It is recommended to present p-values to 3 decimal places.

• The betas in the logistic regression table are uninterpretable. Odds ratio would have been better…

Discretionary Revisions: Reject

Level of interest: An article of importance in its field

Quality of written English: Authors should proofread and correct for grammatical errors and use appropriate research language

Statistical review: No, I am a statistician and have reviewed the statistical methods used.

Declaration of competing interests: I declare that I have no competing interest

Reviewer #3: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Alessandro Rovetta

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: Reviewer_comment_round 2.docx

Click here for additional data file.

27 May 2022

For Academic Editor

• Please reconsider the use of the term "discrimination" and the presentation of your conclusions.

According to your suggestion, we've changed the word “discriminate” to “consider” or “recognize”.

Response to Reviewers

Reviewer #1

1) Logistic regression assumptions are still not discussed; multicollinearity should be added to the publication.

We’ve taken your advice and clarified variance inflation factor (VIF) of each variable has been used for evaluating the risk of multicollinearity in "Statistical analyses" section (Page 7, Line 132-133).

Additionally, in the binomial logistic regression model with only 2 variables, N95 mask implementation and working period in the red zone, we’ve clarified those VIFs (Page 12, Line 233-234).

Meanwhile, also referring the suggestion of Reviewer 2, in the re-revise version of our manuscript, Tables 2-3 have been re-summarized with odds ratio rather than beta, etc.

Reviewer #2

Methods

• Authors should state the eligibility (inclusion and exclusion) criteria under the subsection “Study design and participants”.

We’ve taken your advice and added the rationale for the provision for red zone workers into the "Study design and participants" section (Page 5, Line 91-93).

• “Data analyses” and Statistical analysis” sections should be integrated into one subheading called “Statistical analysis”.

Following your suggestion, we have integrated these two into one section " Statistical analysis".

• Authors should provide the assumptions on arriving at a sample size of 1237.

We’ve taken your advice and added the rationale for the size of the recruited participants into the "Study design and participants" section (Page 5, Line 85-88).

Results

• The results section should start with the subheading “Characteristics of study participants”. This section should have a table and comments describing who these participants were: example, distribution of sex, age, and other sociodemographic characteristics…

In fact, the other characteristics of the present study participants couldn't be precisely clarified, because individual questionnaire didn't include such questions. However, the second serological survey of HCWs was similarly conducted December 2021 in Shiga prefecture. In the second survey, 1600 HCWs was characterized with age of 42.1 ±16.4 (indicating mean ±S.D. below) years old, body mass index of 23.4 ±8.7 kg/m2, and 66.2% of female predominance. Thus, the present study was presumably composed with similar character population.

These have been described in the re-revised manuscript (Page 8, Line 164 – Page 9, Line 170).

• The analysis and presentation of results did not meet the statistical rigor required of a seroprevalence study. The tables were poorly presented. P-values were inappropriately presented. It is recommended to present p-values to 3 decimal places.

• The betas in the logistic regression table are uninterpretable. Odds ratio would have been better…

According to your suggestions, we have summarized and refined Tables 2-3.

Submitted filename: PtoP_PLosOne2.docx

Click here for additional data file.

9 Jun 2022

Serology suggests adequate safety measures to protect healthcare workers from COVID-19 in Shiga Prefecture, Japan

PONE-D-21-35904R2

Dear Dr. Chano,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Gabriel O Dida, PhD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Dear authors, thank you for your final revisions. I wish you the best for this and your future research.

Reviewer #3: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Alessandro Rovetta

Reviewer #3: No

**********

14 Jun 2022

PONE-D-21-35904R2

Serology suggests adequate safety measures to protect healthcare workers from COVID-19 in Shiga Prefecture, Japan

Dear Dr. Chano:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Gabriel O Dida

Academic Editor

PLOS ONE