SARS-CoV-2 seroprevalence among patients with severe mental illness: A cross-sectional study.

Patients with severe mental illness (SMI) i.e. schizophrenia, schizoaffective disorder, and bipolar disorder are at increased risk of severe outcomes if infected with coronavirus disease 2019 (COVID-19). Whether patients with SMI are at increased risk of COVID-19 is, however, sparsely investigated. This important issue must be addressed as the current pandemic could have the potential to increase the existing gap in lifetime mortality between this group of patients and the background population. The objective of this study was to determine whether a diagnosis of schizophrenia, schizoaffective disorder, or bipolar disorder is associated with an increased risk of COVID-19. A cross-sectional study was performed between January 18th and February 25th, 2021. Of 7071 eligible patients with schizophrenia, schizoaffective disorder, or bipolar disorder, 1355 patients from seven psychiatric centres in the Capital Region of Denmark were screened for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) antibodies. A total of 1258 unvaccinated patients were included in the analysis. The mean age was 40.5 years (SD 14.6), 54.3% were female. Fifty-nine of the 1258 participants had a positive SARS-CoV-2 antibody test, corresponding to a adjusted seroprevalence of 4.96% (95% CI 3.87-6.35). No significant difference in SARS-CoV-2-risk was found between female and male participants (RR = 1.32; 95% CI 0.79-2.20; p = .290). No significant differences in seroprevalences between schizophrenia and bipolar disease were found (RR = 1.12; 95% CI 0.67-1.87; p = .667). Seroprevalence among 6088 unvaccinated blood donors from the same region and period was 12.24% (95% CI 11.41-13.11). SARS-CoV-2 seroprevalence among included patients with SMI was significantly lower than among blood donors (RR = 0.41; 95% CI 0.31-0.52; p < .001). Differences in seroprevalences remained significant when adjusting for gender and age, except for those aged 60 years or above. The study is registered at ClinicalTrails.gov (NCT04775407). https://clinicaltrials.gov/ct2/show/NCT04775407?term=NCT04775407&draw=2&rank=1.

Introduction

The pandemic of coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has caused health problems worldwide [1]. More than 250 million cases have been confirmed globally [2]. As the pandemic has evolved, it has become increasingly evident that individuals are disproportionately affected by the disease, e.g. patients with comorbid diabetes, obesity, and cardiovascular diseases have been reported to correlate with a worse outcome [3-8]. The same somatic conditions are overrepresented in patients with a severe mental illness (SMI) [9,10]. In addition, patients suffering from SMI are at increased risk for infectious diseases, have lower hospitalization rates, and have limitations in access to healthcare [11-13]. Thus, patients with SMI are possibly at increased risk of severe outcomes of COVID-19. These concerns have been confirmed in several studies [14-20]. However, whether patients with SMI are more likely to be infected with SARS-CoV-2 is not clear [14-16,21,22].

There is a need for studies to elucidate if this patient group is at increased risk of contracting the virus and whether the COVID-19 pandemic increases the existing health inequalities between this vulnerable patient group and the general population. The aim of the present study was to determine the seroprevalence of SARS-CoV-2 antibodies in patients with a diagnosis of schizophrenia, schizoaffective disorder, or bipolar affective disorder receiving in-patient or out-patient care via mental health services in the Capital Region of Denmark and to compare these data with the seroprevalence among Danish blood donors as a proxy for the general population in the Capital Region of Denmark. Additionally, we aimed to examine possible risk factors that might be associated with SARS-CoV-2 infection.

Methods

Study overview

This cross-sectional study was conducted at seven psychiatric centres in the Capital Region of Denmark. The Scientific-Ethical Committee of the Capital Region of Denmark (H-20037171) and the Danish Data Protection Authorities (P-2020-1037) in the Capital Region of Denmark approved the study. The study was carried out in accordance with the Declaration of Helsinki 1964 and with national laws and Regulations for Clinical Research. The study is registered at ClinicalTrails.gov: NCT04775407.

Participants

Eligible patients were adults aged 18 or above, diagnosed with SMI i.e. schizophrenia, schizoaffective disorder, or bipolar affective disorder according to the criteria of the International Classification of Diseases, World Health Organization (WHO), and treated in the Capital Region of Denmark. To secure sufficient inclusion of patients, all seven psychiatric centres in The Mental Health Services, Capital Region of Denmark, provided the following information (from EPIC/Medical Record System); name, social security number, contact information, diagnoses, and name on the psychiatric department. Patients who were found too vulnerable for participation by their treating psychiatrists were excluded.

Study design

Between January 18th and February 19th, 2021 all eligible patients were invited to serological screening for SARS-CoV-2 antibodies. All patients who fulfilled the inclusion criteria received an invitation and written study information through the Danish governmental, personal, secure email system (E-Boks) or by letter. All patients were then contacted via phone and oral study information was provided. Contact via phone was attempted up to three times. Patients whose phone number was not available through EPIC/Medical Record System, received another E-Boks message with contact information to study investigators. All patients were offered up to 48 hours to consider participation.

The serological screening took place from February 1st, through February 25th, 2021, at 20 different locations, associated with seven psychiatric centres throughout the Capital Region of Denmark. Patients were free to choose test day and location. On testing day, patients who wanted to participate signed written consent. All patients were informed that they could withdraw from the study at any time without providing a reason and without any consequences for their current or future psychiatric treatment. Patients who did not manage to show up for the scheduled blood test were offered a new appointment.

A maximum of 6 mL blood was collected per patient, and serum was tested for total SARS-CoV-2 immunoglobulins by use of WANTAI SARS-CoV-2 antibody ELISA [23]. Analyses were performed at Statens Serum Institut in Copenhagen, Denmark, and carried out according to the manufacturer’s recommendations (Beijing Wantai Biological Pharmacy Enterprise Co., Ltd.). The manufacturer reported a sensitivity of 94.5% and a specificity of 100% [23]. All samples were analysed within 8 weeks and destroyed immediately afterward. Test results were categorized as positive if they showed the presence of SARS-CoV-2 immunoglobulins.

Results from included patients with SMI were compared with data from Danish blood donors in the Capital Region of Denmark obtained from the same period. Data extracted from the database of Danish blood donors, included results of SARS-CoV-2 antibody tests and demographic data, i.e. age and gender. These data were used as a proxy for the general population in the Capital Region of Denmark. Inconclusive test results and results from study participants and blood donors who had received at least one dose of any vaccine against SARS-CoV-2 were excluded from the statistical analysis.

Statistical analysis

Seroprevalences were presented as percentages. Demographic data were presented descriptively, e.g. mean, standard deviation (SD), and percentages and compared (seropositive vs. seronegative and participants vs. non-participants) using unpaired t-tests and Fisher’s exact tests respectively. P-values for Fisher’s exact tests were two-sided. Comparisons between included patients and blood donors were explored through relative risks (RRs). RRs were calculated as the probability of having a positive test for SARS-CoV-2 antibodies among included patients compared with the probability among blood donors. We calculated RRs for gender and age groups to rule out the risk of these factors being confounders. Further, RRs were calculated to explore associations between SARS-CoV-2 serology and certain risk factors, such as gender, diagnosis, geographical area, and age <30 years. RRs and seroprevalences were calculated using the “twoby2” function in the R-package “Epi”. The 95% confidence intervals (CIs) of the probabilities were computed using the normal approximation to the log-odds as implemented in the “Epi” package. All seroprevalences were adjusted for test sensitivity and specificity, using the Rogan and Gladen [24] method through the R package “epiR”. Results were presented with 95% CIs. All p-values below 0.05 were considered statistically significant. Statistical analyses were calculated using R (http://www.R-project.org/).

Results

Trial population and baseline characteristics

Out of a total population of 7310 patients, 238 patients were found too vulnerable for participation. Of the remaining, 1355 patients (19.2%) signed the consent. Of these, 41 (3.0%) of consenters had received at least one dose of a COVID-19 vaccine. Blood samples from another 56 (4.1%) patients were not included in the analysis due to various reasons (Fig 1). A total of 1258 patients (17.2% of the total study population) were included in the final analysis, of which 54.3% were female. The mean age was 40.5 years (SD 14.6). Seropositive patients were significantly younger than seronegative patients (mean age (SD): 35.7 (13.5) vs. 40.8 (14.6); p = .006). Most of the included patients (57.8%) were diagnosed with schizophrenia. Baseline characteristics of the included patients are presented in Table 1. Participants were significantly younger (mean age (SD): 40.5 (14.6) vs. 45.0 (16.4); p < .001) and more participants were females (54.3% vs. 46.8%; p < .001) and diagnosed with bipolar disorder (38.5% vs. 24.5%; p < .001) compared with SMI patients who did not participate in the study (Table 2).

Fig 1

Flowchart of patients with severe mental illness.

All unvaccinated patients with severe mental illness who fulfilled all criteria to be enrolled in the study, and with a positive or negative SARS-CoV-2 antibody test result, were included in the statistical analysis.

Table 1

Baseline characteristics of study population according to SARS-CoV-2 serology.

Characteristic	Total (N = 1258)	Positive (n = 59)	Negative (n = 1199)	P valueb
Age, mean (SD)	40.5 (14.6)	35.7 (13.5)	40.8 (14.6)	.006
Gender, n (%)				.349
Female	683 (54.3)	36 (61.0)	647 (54.0)
Male	575 (45.7)	23 (39.0)	552 (46.0)
Diagnosis, n (%)				.316
Schizophrenia	727 (57.8)	37 (62.7)	690 (57.5)
Schizoaffective disorder	47 (3.7)	0	47 (3.9)
Bipolar disorder	484 (38.5)	22 (37.3)	462 (38.5)
Geographical location, n (%)				.002.10
Copenhagen municipalityc	711 (56.5)	45 (76.3)	666 (55.5)
Suburban municipalitiesd	547 (43.5)	14 (23.7)	533 (44.5)

aAll patients with severe mental illness who underwent SARS-CoV-2 antibody testing and were included in the statistical analysis.

bComparisons between seropositive and seronegative participants calculated using Fisher’s exact test and unpaired t-test. Proportions may not be equal to 100% due to rounding error.

cIncluding psychiatric centres of Copenhagen and Amager.

dIncluding psychiatric centres of Ballerup, Bornholm, Glostrup, Nordsjælland, and Sct. Hans.

Table 2

Baseline characteristics of total study population by study participation.

Characteristic	Total (N = 7310)	Participants (n = 1258)	Non-participants (n = 6052)	P valueb
Age, mean (SD)	44.2 (16.2)	40.5 (14.6)	45.0 (16.4)	< .001
Gender, n (%)				< .001
Female	3516 (48.1)	683 (54.3)	2833 (46.8)
Male	3794 (51.9)	575 (45.7)	3219 (53.2)
Diagnosis, n (%)				< .001
Schizophrenia	5025 (68.7)	727 (57.8)	4298 (71.0)
Schizoaffective disorder	316 (4.3)	47 (3.7)	269 (4.4)
Bipolar disorder	1969 (26.9)	484 (38.5)	1485 (24.5)
Geographical location, n (%)				< .001
Copenhagen municipalityc	3604 (49.3)	711 (56.5)	2893 (47.8)
Suburban municipalitiesd	3706 (50.7)	547 (43.5)	3159 (52.2)

aAll patients treated for either schizophrenia, schizoaffective disorder, or bipolar disorder in The Mental Health Services, Capital Region of Denmark.

bComparisons between seropositive and seronegative participants calculated using Fisher’s exact test and unpaired t-test. Proportions may not be equal to 100% due to rounding error.

cIncluding psychiatric centres of Copenhagen and Amager.

dIncluding psychiatric centres of Ballerup, Bornholm, Glostrup, Nordsjælland, and Sct. Hans.

SARS-CoV-2-seroprevalence

Fifty-nine of 1258 included patients had a positive SARS-CoV-2 antibody test, corresponding to a adjusted seroprevalence of 4.96 (95% CI 3.87–6.35). No significant difference was found between the participating females and males (RR = 1.32; 95% CI 0.79–2.20; p = .290). The diagnose of schizophrenia were not associated with a higher risk of COVID-19 infection, compared with bipolar disease (RR = 1.12; 95% CI 0.67–1.87; p = .667). No cases (95% CI 0.00–8.00) of SARS-COV-2 antibodies were detected among patients with schizoaffective disorder. Patients from districts located in Copenhagen municipality were at higher risk of COVID-19 compared with patients from suburban psychiatric centres (RR = 2.47; 95% CI 1.37–4.46; p = .002). Also, participants younger than 30 years were at significantly higher risk of being seropositive compared with those aged 30 or above (RR = 1.70; 95% CI 1.02–2.82; p = .042) (Table 3).

Table 3

Associations between SARS-CoV-2 antibodies and certain risk factors among study participants.

Risk factor	Seroprevalences, % (95% CI)a		Relative Risk (95% CI)	P-value
	Risk factor group	Control group
Femaleb	5.58 (4.05–7.63)	4.23 (2.84–6.28)	1.32 (0.79–2.20)	.290
Schizophreniac	5.39 (3.93–7.34)	4.81 (3.20–7.18)	1.12 (0.67–1.87)	.667
Copenhagen municipalityb Calibri (Brødtekst)	6.70 (5.04–8.85)	2.71 (1.62–4.50)	2.47 (1.37–4.46)	.002
Age <30b	7.08 (4.76–10.40)	4.17 (3.02–5.72)	1.70 (1.02–2.82)	.042

aSeroprevalences were adjusted for test sensitivity and specificity.

bControl group = Remaining participants.

cControl group = Participants with bipolar disease.

In the same period, 6088 (50.0% female, mean age 40.0 years (SD 13.7)) unvaccinated blood donors from the Capital Region of Denmark were tested for SARS-CoV-2 antibodies. A total of 704 blood donors were seropositive, corresponding to a adjusted seroprevalence of 12.24 (95% CI 11.41–13.11). Seroprevalence among participants in the present study was significantly lower than among blood donors (RR = 0.41; 95% CI 0.31–0.52; p < .001) (Fig 2).

Fig 2

Seroprevalences for included patients with severe mental illness compared with blood donors of the Capital Region of Denmark.

Seroprevalences adjusted for test sensitivity and specificity with 95% confidence intervals among all included patients with severe mental illness (n = 1258), according to ICD-10 diagnosis (schizophrenia n = 725, bipolar disorder n = 484) and blood donors (n = 6088).

When comparing gender, significant differences between included patients and blood donors remained (female: RR = 0.43; 95% CI 0.31–0.60; p < .001, male: RR = 0.37; 95% CI 0.24–0.56; p < .001) (Table 4). As Table 4 reveals, the seroprevalence was significantly lower for patients with SMI across all age groups except for patients aged 60 years or above (<30 years: RR = 0.41, 95% CI 0.27–0.62; p < .001, 30–39 years: RR = 0.49; 95% CI 0.30–0.79; p = .003, 40–49 years: RR = 0.32, 95% CI 0.16–0.64; p = .001, 50–59 years: RR = 0.38; 95% CI 0.18–0.81; p = .010, ≥60 years: RR = 0.51; 95% CI 0.18–1.42; p = .194).

Table 4

Seroprevalences and RRs for the total group and according to gender and age group for study participants and blood donors.

Characteristic	Seroprevalence, % (95% CI)a		Relative Risk (95% CI)	P-value
	SMI patients	Blood donors
Total	4.96 (3.87–6.35)	12.24 (11.41–13.11)	0.41 (0.31–0.52)	< .001
Female	5.58 (4.05–7.63)	13.02 (11.84–14.31)	0.43 (0.31–0.60)	< .001
Male	4.23 (2.84–6.28)	11.45 (10.33–12.67)	0.37 (0.24–0.56)	< .001
Age <30	7.08 (4.76–10.40)	17.15 (15.47–18.97)	0.41 (0.27–0.62)	< .001
Age: 30–39	5.92 (3.73–9.28)	12.20 (10.41–14.25)	0.49 (0.30–0.79)	.003
Age: 40–49	3.25 (1.65–6.28)	10.30 (8.65–12.22)	0.32 (0.16–0.64)	.001
Age: 50–59	3.65 (1.78–7.35)	9.57 (7.99–11.44)	0.38 (0.18–0.81)	.010
Age: ≥60	2.92 (1.14–7.28)	5.72 (4.08–7.97)	0.51 (0.18–1.42)	.194

aSeroprevalences were adjusted for test sensitivity and specificity.

Discussion

The present study investigated the seroprevalence of SARS-CoV-2 antibodies in patients with SMI compared with the prevalence among blood donors in the same region and period. We found that the risk of SARS-CoV-2 seropositivity was significantly lower among patients with SMI than among blood donors. The risk of COVID-19 remained significantly lower among the patients when comparing across gender and age groups apart from for those aged 60 or above. The seroprevalence was significantly higher in SMI patients younger than 30 and patients from psychiatric centres located in Copenhagen municipality with no difference between diagnoses or gender. This is in congruence with the patterns of SARS-CoV-2 seroprevalence among the general Danish population [25].

The significantly lower seroprevalence of SARS-CoV-2 antibodies in patients with SMI may be a consequence of a lower number of social contacts for patients with psychotic disorders and patients with increased anxiety and depression symptoms related to the pandemic [26-28]. While isolation due to lock-down has imposed profound changes on many people’s private and working life, the changes for people with SMI may have been less profound. In accordance with our findings, Van der Meer et al. have reported, that patients with psychiatric disorders were more frequently SARS-CoV-2-negative than people undergoing testing without such conditions. The population in this study, however, was small and not fully representative [22]. On the contrary, other studies have reported an association between SMI and increased risk of contracting COVID-19 [14,16,21].

In our study, we included seroprevalence from unvaccinated blood donors. Limitations using blood donors as a proxy for the general population must be considered [29]. A study has proposed that seropositivity might be lower in blood donors than in the general Danish population, as they represent a healthy group, screened for behavioral risks [30,31]. However, it has been estimated that 30% of COVID-19 transmission in Denmark occurs at workplaces [32]. One could argue that blood donors due to their better health conditions, could constitute a larger proportion of the labor market and thus at higher risk of SARS-CoV-2.

In a recent National Danish surveillance study, a 2 percentage points lower seroprevalence (10.69% (95% CI 9.74–11.73)), than for our blood donors samples, was found among 3919 participants from the Capital Region of Denmark. The surveillance study was conducted over 3 weeks following the termination of our own study [25]. This suggests that the seroprevalence among danish blood donors could be slightly higher than in the general Danish population.

The majority of previously published studies report an increased risk of SARS-CoV-2 infection among patients with SMI [14,16,21], an association strongest for schizophrenia [14,21]. In our study, we found a significantly low seroprevalence independently of mental disorder. These contradictory results can be attributed to various reasons. To the best of our knowledge, all previous studies were designed as retrospective cohorts with results based on database registered PCR tests which induce a risk of selection bias; In some countries, infection symptoms have been a prerequisite for COVID-19-testing, why asymptomatic cases would not be captured. In a meta-analysis, Buitrago-Garcia et al. [33] estimated that approximately 20% of people with COVID-19 infection remain asymptomatic. Additionally, it is well recognised that patients with SMI find it difficult to report physical health needs [34,35]. Furthermore, there may be COVID-19-testing fees that deter people from being tested as well as the use of other non-national test forms (e.g. pop-up sites with antigen rapid tests) which may not be registered in the national electronic health records. Due to our study design with the use of serological antibody screening, selection bias as a consequence of various test strategies has been avoided. A variety of other factors are also likely to contribute to conflicting results regarding the risk of COVID-19 among patients with SMI. Findings might be influenced by the characteristics of the Health Care System. Appraising health care information can be challenging for patients with SMI [36]. Muruganandam et al. reported that knowledge about COVID-19 infection is reduced in this patient population compared with the general population, with two-thirds of the SMI patients not having adequate knowledge about precautionary measures [37]. All patients, regardless of the type of SMI, receiving in-patient or out-patient care via one of the psychiatric centres in the Capital Region of Denmark have a personal health care professional who during lockdown has been able to provide COVID-19 related information during the COVID-19 lockdown. The close interaction between patients in our study population and health care professionals might have played an important role in optimizing the patients understanding of the pandemic and thereby minimizing their risk of infection. Socioeconomic differences (e.g. education level and housing conditions) for patients with SMI between countries must also be considered as part of the explanation [38]. As an example, homelessness in Denmark is associated with a higher risk of SARS-CoV-2 infection [39]. In Denmark however, all patients with SMI are offered social assistance and housing. Homelessness as a risk factor for patients with SMI may play a larger role in countries with a less comprehensive social welfare system.

Out of the total number of eligible patients in the study population, 1355 patients chose to participate. Serological samples from 1258 unvaccinated patients were included in the final analysis. The participation rate in our study was relatively high considering the vulnerable patient population (17.2% of the total patient population). Still, it must be considered if study participants are representative of patients with SMI in general. There was a significantly larger proportion of females and patients with bipolar disorder among participants compared with non-participants. This might have biased our results, as previous studies have found an increased risk of SARS-CoV-2 infections among patients with schizophrenia compared with other psychiatric conditions [14]. However, the fact that we found no differences in seroprevalence between diagnoses and gender makes this unlikely. We excluded serological results from patients who had received at least one dose of any COVID-19 vaccine. At this stage in Denmark, only patients expected to be at high risk of a severe outcome of COVID-19 were offered vaccination [40]. This may have biased our results if this vulnerable group of patients, e.g. psychotic or manic patients, were at higher risk of infection. Nevertheless, if vaccinated patients were included in the analysis, the seroprevalence (7.90%) for our study participants remains significantly lower than for unvaccinated blood donors (RR = 0.65; 95% CI 0.53–0.79; p < .001). The number of COVID-19-related deaths among Danish patients with schizophrenia or bipolar affective disorder is relatively low [20] and therefore unlikely to be a reason for the difference in seroprevalence.

Several studies have addressed that patients with SMI, are at increased risk of severe outcomes of COVID-19 [14-20] and common comorbid risk factors among patients with psychiatric disorders such as chronic obstructive pulmonary disease, obesity, and cardiovascular disease, are associated with severe COVID-19 outcomes [9,10]. A recently published study reported a 2 fold higher risk of severe outcomes for patients with schizophrenia and bipolar disorder, respectively across all regions in Denmark [20]. Consequently, the current pandemic has the potential to increase the already existing gap in lifetime mortality between this group of patients and the background population. Our results suggest that patients with SMI might be less exposed to COVID-19, but the fact that these patients are at higher risk of severe COVID-19 outcomes emphasizes the importance of continued focus on this vulnerable patient group during the pandemic.

Limitations and strengths

This study has several strengths. Our study design allowed us to compare seroprevalences, for two demographically similar populations in the exact same period. Although our study population comprises a mentally vulnerable group of patients, our participation rate is comparable with those of the nationwide SARS-CoV-2 prevalence studies of the general Danish population [25,41]. To our knowledge, this is the first study systematically screening patients diagnosed with schizophrenia, schizoaffective disorder, and bipolar disorder for SARS-CoV-2 antibodies. By testing for SARS-CoV-2 antibodies, we secured that individuals having had asymptomatic infections with SARS-CoV-2 were also included. Finally, all blood samples were analysed using ELISA, which has high sensitivity and specificity and therefore a low risk of false-positive or false-negative test results. The present study also has limitations that must be considered. Patients who previously had been tested positive by PCR or antigen rapid tests might have declined to participate thereby underestimating our results. To minimize this bias, however, the importance of study participation regardless of former infection was emphasized to invited patients when recruited to the study. Furthermore, it is worth mentioning, that the same bias exists in the previously mentioned National Danish Surveillance Study. Due to the cross-sectional design, it was not possible to determine when the participants were infected. Also, recently infected patients may not have been detected, as antibodies are detectable around one week after contamination [42]. Out of the total study population, 238 patients were found too vulnerable to participate by their treating psychiatrist. Furthermore, patients on closed wards or under involuntary commitment were not allowed to participate. Both groups might be at higher risk of SARS-CoV-2 infection. Study participants were, nevertheless, younger and more comprised of patients from the psychiatric centres in Copenhagen municipality compared with non–participants in the study population—both factors which were associated with higher seroprevalence in our study participants.

In conclusion our study documents a significantly lower seroprevalence of SARS-CoV-2 antibodies among patients with SMI compared with the seroprevalence of SARS-CoV-2 in blood donors. Further studies should be carried out to elucidate the different impacts of COVID-19 in patients with SMI between countries to secure national health care strategies.

12 Oct 2021

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,

Ray Borrow, Ph.D., FRCPath

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

“This study was funded by Mental Health Services, the Capital Region of Denmark.”

We note that you have provided funding information within the 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:

“The study was funded by Mental Health Services, the Capital Region of Denmark. The funders played no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.”

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

3. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

[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: Yes

Reviewer #3: Partly

Reviewer #4: Yes

**********

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

Reviewer #1: No

Reviewer #2: I Don't Know

Reviewer #3: Yes

Reviewer #4: 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: Yes

Reviewer #3: Yes

Reviewer #4: 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: Yes

Reviewer #3: Yes

Reviewer #4: 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: The manuscript entitled ‘The SARS- CoV-2 seroprevalence among patients with severe mental illness: a cross-sectional study’ with the aim to determine whether a diagnosis of schizophrenia, schizoaffective disorder, or bipolar disorder is associated with an increased risk of COVID-19.

This is quite an interesting study but the manuscript can be further improved.

Abstract

Page 3 Line 47, the comma to indicate 95% CIs could be omitted e.g 95% CI, 0.79-2.20 can be presented as 95% CI 0.79-2.20 or 95% CI: 0.79-2.20.

Page 3 Line 49, comparison of gender to be highlighted.

Methods

The type of SMI included in the study to be clearly stated.

Statistical analyses

Line 121, the word respectively to be added after tests.

Line 121, unpaired t-tests and Pearson’s chi-square tests to be written as unpaired t-test and Pearson’s chi-square test. Likewise with Line 154-155, 161-162.

Line 124, the sentence ‘to adjust for gender and age, we calculated RRs for gender and age groups’ requires revision. The word adjust not clear.

Line 129, the word adjust to be replaced with another word. Likewise with Line 187.

Results

Table 1, the empty column to be removed.

Line 143-144, the sentence requires revision.

Line 151, the title for Table 1 is too short.

Table 1, decimal point for p value to be standardized. N to be used for overall sample size (Total). No.(%) to be replaced with n(%).

Line 166-175, Line 181-189, data/results to be presented in table form. The analysis whether crude or adjusted to be clearly stated. Multivariable statistical test could be explored.

The cited reference to be spaced out from the previous word throughout the manuscript e.g. COVID-19(14,16,21). [ ] to be used instead of ( ).

List of references to conform with PLoS ONE format.

Reviewer #2: Line 99 and Figure 2 – While Blood donors were used as a comparison, is there an estimate of the average population rate of COVID positivity in Denmark in February 2021? And if so, is this different? Blood donors may not be representative of the general population and this should be addressed in the text. The authors should discuss more why the patients in this study had a lower than population rate of positivity but, as noted in the Discussion, probably a higher rate of hospitalization due to COVID-19.

In general the results suggest that mental health patients are significantly less likely to have contracted COVID-19 is a surprising finding and contrary to most other studies. Clearly the study is limited by the patients who chose to be screened and may have not agreed to screening because they knew they had had COVID-19. This complicates the interpretation of the results unless there is another reason for the difference. Enhanced discussion is needed

Reviewer #3: The manuscript is well written and studies an underserved population (severe mentally ill). The focus on COVID in this population is highly timely. Investigation of such a population is challenging for many reasons and the authors have done an excellent job describing their approach and success.

A major strength of the study is reliance on serology for estimates of prevalence, which is not subject to many bias and other potential confounding factors related to medical care, patient perception, isolation, and testing of the mentally ill.

The study findings of a low prevalence of COVID (~4%) in the mentally ill population is remarkable and worth recognition. The factors that contribute to the low prevalence are important to recognize. This reviewer feels the low prevalence rate alone should be a major focus, rather than comparison to another group.

The major weakness of the study is the attempt to compare their study population to a reference group. While of interest, the rate of blood bank donors is a poor comparison group. The rate of similarly institutionalized individuals would make a much better comparison group, of which, admittedly there are few. One suggestion might be prisoners.

Another analytic approach might be to compare the mentally ill patients to multiple other groups that may have been published, such as local COVID rates in Denmark among citizens, or specialized groups (health care workers). The limitation of the comparison group is redundantly addressed in the discussion and the limitations sections.

Reviewer #4: This study is original as it examines the association between severe mental disorders and risk of COVID-19. The author uses a cross-sectional study on psychiatric centers with blood donors as control. The study is of interest even if the results are not significant.

I have some questions for this manuscript including the following (organized by manuscript section).

Methods - Statistical analysis

Due to the small number of expected subjects (<5) in some categories, and the highly unbalanced marginal sums of the actual data set, Pearson’s chi-square test cannot be applied. It is better to use Fisher’s exact test.

Results

All seroprevalences in the text are crude prevalences and not corrected prevalences as suggested by Rogan et al. 1978 and indicated in the method paragraph. So give the corrected prevalences even if there are not significantly different from the crude prevalence. Do the same in the abstract.

I don’t find the same prevalences than authors for male and female:

Crude prevalence = 4.69% for female (32/683) instead of 5.27% in the text

Corrected prevalence= 4.97% for female (0.0469+1-1)/(0.945+1-1) according to Rogan et al. 1978)),

Crude prevalence = 4.70% for male (27/575) instead of 4.00% in the text

Corrected prevalence= 4.97% for male

Even if there are no cases of SARS-COV-2 antibodies among patients with schizoaffective disorder, it is possible to calculate the 95% CI.

For mental health diagnosis, it will be better to calculate prevalence for each diagnosis separately, and relative risk as compared to schizophrenia which is the largest group, and not as compared to remaining included patients with SMI which include people with different prevalences.

For the different districts of Copenhagen, I would separate central part and Amager from the other districts which have higher corrected prevalences of respectively 6.67 and 6.78 while others have corrected prevalences of 2.74.

Even if it seems evident, gives the reference in the sentence 174-175 about age. It could be another group than ≥ 30 years.

RR between participants and blood donors is 0.41 instead of 0.40 with approximation (0.405).

Table 1: there is an extra empty column. Thank you to delete it. I will be interested to have age by group and not as a continued variable, the mean age being included in the text.

Discussion

The paragraph lines 218- must be just after lines 211.

I will be interesting to know if there are some difference in the literature according severe mental health diagnosis and to discuss it and not only according sex differences.

It would also be interesting to discuss the difference between neighbourhoods. Is it differences in socio-economic level, in population characteristics?

Problems at the end of line 259 for reference 20.

**********

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: No

Reviewer #2: No

Reviewer #3: No

Reviewer #4: 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.

12 Nov 2021

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)

We thank the Reviewers for taking the time to carefully read our submission to PLOS ONE and appreciate the positive and constructive comments. In the following, please find the specific comments from Reviewers and the corresponding responses from the authors. Thank you.

Reviewer #1: The manuscript entitled ‘The SARS- CoV-2 seroprevalence among patients with severe mental illness: a cross-sectional study’ with the aim to determine whether a diagnosis of schizophrenia, schizoaffective disorder, or bipolar disorder is associated with an increased risk of COVID-19.

This is quite an interesting study but the manuscript can be further improved.

We do appreciate the constructive and positive comments by Reviewer 1 and appreciate hers/his time reviewing our manuscript.

Abstract

Page 3 Line 47, the comma to indicate 95% CIs could be omitted e.g 95% CI, 0.79-2.20 can be presented as 95% CI 0.79-2.20 or 95% CI: 0.79-2.20.

Author response: Revised accordingly. Thank you for your comment.

Page 3 Line 49, comparison of gender to be highlighted.

Authors response: Please see lines 44-45: “No significant difference in SARS-CoV-2-risk was found between female and male participants (RR=1.32; 95% CI 0.79-2.20; p=.29).” We hope that you will find this adequate.

Methods

The type of SMI included in the study to be clearly stated.

Authors response: Please see lines 32 and 82. We hope that you will find this adequate.

Statistical analyses

Line 121, the word respectively to be added after tests.

Author response: Thank you. Revised accordingly.

Line 121, unpaired t-tests and Pearson’s chi-square tests to be written as unpaired t-test and Pearson’s chi-square test. Likewise with Line 154-155, 161-162.

Author response: Thank you. Revised accordingly.

Line 124, the sentence ‘to adjust for gender and age, we calculated RRs for gender and age groups’ requires revision. The word adjust not clear.

Author response: The sentence has been rewritten: “We calculated RRs for gender and age groups to rule out the risk of these factors being confounders.”

Line 129, the word adjust to be replaced with another word. Likewise with Line 187.

Author response: Replaced with “corrected”.

Results

Table 1, the empty column to be removed.

Author response: You are completely right. Thank you for your comment, the empty column has been removed.

Line 143-144, the sentence requires revision.

Author response: Thank you. The sentence has now been rephrased: “Participants were significantly younger (mean age (SD): 40.5 (14.6) vs. 45.0 (16.4); p<.001) and more participants were females (54.3% vs. 46.8%; p<.001) and diagnosed with bipolar disorder (38.5% vs. 24.5%; p<.001) (Table 2) compared with SMI patients who did not participate in the study.”

Line 151, the title for Table 1 is too short.

Author response: Good point. The title has been changed from: “Baseline characteristics” to “Baseline characteristics of patients according to SARS-CoV-2 serology”.

Table 1, decimal point for p value to be standardized. N to be used for overall sample size (Total). No.(%) to be replaced with n(%).

Author response: Corrected. P-values �  .01 are presented with 2 decimals, p-values between .001 and .01 are presented with 3 decimals and p-values below .001 are presented as p<.001.

Line 166-175, Line 181-189, data/results to be presented in table form. The analysis whether crude or adjusted to be clearly stated. Multivariable statistical test could be explored.

Author response: Results are now presented in table forms with corrected seroprevalences. Unfortunately, it was not possible to do multivariable tests, as data from blood donors were sent categorically in tables and not from each individual.

The cited reference to be spaced out from the previous word throughout the manuscript e.g. COVID-19(14,16,21). [ ] to be used instead of ( ). List of references to conform with PLoS ONE format.

Author response: We apologize and have carefully revised the manuscript, and the references do now confirm with PLOS ONE format.

Reviewer #2: Line 99 and Figure 2 – While Blood donors were used as a comparison, is there an estimate of the average population rate of COVID positivity in Denmark in February 2021? And if so, is this different? Blood donors may not be representative of the general population and this should be addressed in the text.

Author response: Thank you for your time and effort. We agree that this comparison would be very relevant. Unfortunately, data for the average population rate of COVID positivity in Denmark in February 2021 does not exist. Instead, we have added to the manuscript a comparison of our results with data from a National Danish Surveillance study conducted over 3 weeks right after termination of our own study. Please see lines 220-224. The reviewer is completely right in his/her comment about the limitations of using blood donors as a reference group and we believe to have discussed this thoroughly. Please see lines 214-219.

The authors should discuss more why the patients in this study had a lower than population rate of positivity but, as noted in the Discussion, probably a higher rate of hospitalization due to COVID-19.

Author response: Somatic conditions associated with a worse outcome of COVID-19 (e.g. hospitalization) are overrepresented in patients with SMI. Reasons for this are enhanced in the discussion lines 391-399. Also, please see see introduction lines 58-61. We believe that our contradictory results with a low seroprevalence among patients with SMI can be attributed to various reasons; Isolation (lines 205-209), selection bias in testing strategies (lines 236-237), characteristics of the Health Care System (lines 242-248), Socioeconomic differences (lines 248-252). Thank you.

In general the results suggest that mental health patients are significantly less likely to have contracted COVID-19 is a surprising finding and contrary to most other studies. Clearly the study is limited by the patients who chose to be screened and may have not agreed to screening because they knew they had had COVID-19. This complicates the interpretation of the results unless there is another reason for the difference. Enhanced discussion is needed.

Author response: This is a good point and a possible bias that also exists in the National Surveillance studies. We have enhanced our discussion please see lines 290-293. Thank you.

Reviewer #3: The manuscript is well written and studies an underserved population (severe mentally ill). The focus on COVID in this population is highly timely. Investigation of such a population is challenging for many reasons and the authors have done an excellent job describing their approach and success.

A major strength of the study is reliance on serology for estimates of prevalence, which is not subject to many bias and other potential confounding factors related to medical care, patient perception, isolation, and testing of the mentally ill.

The study findings of a low prevalence of COVID (~4%) in the mentally ill population is remarkable and worth recognition. The factors that contribute to the low prevalence are important to recognize. This reviewer feels the low prevalence rate alone should be a major focus, rather than comparison to another group.

The major weakness of the study is the attempt to compare their study population to a reference group. While of interest, the rate of blood bank donors is a poor comparison group. The rate of similarly institutionalized individuals would make a much better comparison group, of which, admittedly there are few. One suggestion might be prisoners. Another analytic approach might be to compare the mentally ill patients to multiple other groups that may have been published, such as local COVID rates in Denmark among citizens, or specialized groups (health care workers). The limitation of the comparison group is redundantly addressed in the discussion and the limitations sections.

Author response: We thank you for your positive and constructive comments. We do agree that our finding of a low prevalence in the population of SMI patients is important and the key point in our publication. The perspective of our approach, using a comparison group, is to highlight this result. Our rationale with a comparison of the low prevalence to another group is exactly to make the common reader see, how surprisingly low the prevalence of COVID-19 among patients with SMI is. We do agree that it would have been preferable to have a comparison group without limitations. To this end, we have now included a reference from a National Danish Surveillance study (please see lines 220-224) conducted over 3 weeks in the extension of our study. Comparing our results to the data from participants in the National Danish surveillance study still results in a significantly lower seroprevalence in our patient group. Patients participating in this study are not institutionalized as prisoners but part of the general community. We have included a reference to a study that found that homelessness in Denmark is associated with a higher risk of COVID-19 (please see lines 249-252). In Denmark with a comprehensive social welfare system, however, this group would not be a preferable group for comparison, due to socioeconomic differences (SMI patients in Denmark have normally their own apartment. All people have free access to the hospital system). The data from blood donors were collected in the exact same period and region as blood samples were drawn from patients participating in the study. We, therefore, of the possible reference groups, do believe to have chosen the best group for comparison to the general population to highlight our result of a low seroprevalence among patients with SMI.

Reviewer #4: This study is original as it examines the association between severe mental disorders and risk of COVID-19. The author uses a cross-sectional study on psychiatric centers with blood donors as control. The study is of interest even if the results are not significant.

I have some questions for this manuscript including the following (organized by manuscript section).

We do appreciate the constructive and positive comments by Reviewer 4 and appreciate hers/his time reviewing our manuscript.

Methods - Statistical analysis

Due to the small number of expected subjects (<5) in some categories, and the highly unbalanced marginal sums of the actual data set, Pearson’s chi-square test cannot be applied. It is better to use Fisher’s exact test.

Author response: Fisher’s exact test could not be made with too many groups or observations. To make Fisher’s exact test possible, the psychiatric centres are now separated in those located in Copenhagen municipality and those located elsewhere. Thank you for the comment.

Results

All seroprevalences in the text are crude prevalences and not corrected prevalences as suggested by Rogan et al. 1978 and indicated in the method paragraph. So give the corrected prevalences even if there are not significantly different from the crude prevalence. Do the same in the abstract.

Author response: All seroprevalences are now presented as corrected prevalences. Thank you.

I don’t find the same prevalences than authors for male and female:

Crude prevalence = 4.69% for female (32/683) instead of 5.27% in the text

Corrected prevalence= 4.97% for female (0.0469+1-1)/(0.945+1-1) according to Rogan et al. 1978)),

Crude prevalence = 4.70% for male (27/575) instead of 4.00% in the text

Corrected prevalence= 4.97% for male

Author response: Thank you for this observation. You are completely right. The seroprevalences for males and females were correct but the numbers in Table 1 were not. They are now corrected.

Even if there are no cases of SARS-COV-2 antibodies among patients with schizoaffective disorder, it is possible to calculate the 95% CI.

Author response: The 95% CI of 0.00-8.00 is now added to the manuscript.

For mental health diagnosis, it will be better to calculate prevalence for each diagnosis separately, and relative risk as compared to schizophrenia which is the largest group, and not as compared to remaining included patients with SMI which include people with different prevalences.

Author response: Results are now presented according to each diagnosis, while relative risks are calculated between schizophrenia patients and patients with bipolar disease.

For the different districts of Copenhagen, I would separate central part and Amager from the other districts which have higher corrected prevalences of respectively 6.67 and 6.78 while others have corrected prevalences of 2.74.

Author response: Thank you for your observation. Districts are now separated in psychiatric centres located in Copenhagen municipality (Copenhagen and Amager), and those located in the suburbs (Ballerup, Bornholm, Glostrup, Nordsjælland, and Sct. Hans).

Even if it seems evident, gives the reference in the sentence 174-175 about age. It could be another group than ≥ 30 years.

Author response: Good point. We have now added: ”compared with those aged 30 or above”.

RR between participants and blood donors is 0.41 instead of 0.40 with approximation (0.405).

Author response: Rounding error is corrected. Thank you.

Table 1: there is an extra empty column. Thank you to delete it. I will be interested to have age by group and not as a continued variable, the mean age being included in the text.

Author response: Thank you for your comment, the empty column is now deleted. Age by group in table 1 and table 2 would make it impossible to do comparison tests (Fisher’s exact tests), as there would be too many groups/observations.

Discussion

The paragraph lines 218- must be just after lines 211.

Author response: Thank you. This has now been corrected

It will be interesting to know if there are some difference in the literature according severe mental health diagnosis and to discuss it and not only according sex differences. It would also be interesting to discuss the difference between neighbourhoods. Is it differences in socio-economic level, in population characteristics?

Author response: Thank you for the interesting perspective. We have enhanced our discussion according to severe mental health diagnosis (please see lines 225-226 + 258-260). We have added that in Denmark all patients with severe mental illness, regardless of the type of mental diagnosis, are offered social assistance and housing. This could explain why we see no difference in seroprevalence between the groups of patients with SMI (i.e schizophrenia, -schizoaffective disorder, and -bipolar disorder) in our population. The seroprevalence was significantly higher among patients from Psychiatric Centres located in Copenhagen municipality which is the area in Denmark with the highest population density. This is in congruence with the patterns of SARS-CoV-2 seroprevalence among the general Danish population due to data from a National Danish Surveillance Study (please see lines 220-224). The offer of social services to patients with SMI does not differ between neighborhoods in Denmark. Unfortunately, however, we do not have any exact knowledge or numbers on socio-economic differences between the groups in our population. Hopefully, we can explore this in future studies.

Problems at the end of line 259 for reference 20.

Author response: Thank you. This has now been corrected.

________________________________________

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: No

Reviewer #2: No

Reviewer #3: No

Reviewer #4: 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: PLOS ONE_response to reviewers.docx

Click here for additional data file.

12 Jan 2022

2 Feb 2022

We thank all Reviewers for once again carefully revising our manuscript and for raising some important and relevant questions. Please find our specific comments below. Thank you.

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: The word adjusted could still be used as alternative to the word corrected.

Author response: revised accordingly. Thank you for the comment.

For Table 1 & 2, chi-squares test where applicable could still be utilized provided the assumptions of the statistical test are fulfilled/met. Please check the assumptions and also recheck the p values. If Fisher's Exact test is used, one sided or two sided p value to be stated.

Author response: Good point. We have now added the sentence: “P-values for Fisher’s exact tests were two-sided”.

Reviewer #2: Revisions were adequate Revisions were adequate Revisions were adequate Revisions were adequate Revisions were adequate

Reviewer #3: (No Response)

Reviewer #4: I thank the authors for their responses to comments. Nevertheless, I have some other comments.

Introduction

If you want to use the acronym SMI, it is best to associate it with severe mental illness as in the abstract. Otherwise use SMD for severe mental disorders.

Author response: Thank you for your observation. You are completely right. Severe mental disorders have now been changed to severe mental illness.

Methods - Statistical analysis

Due to the small number of expected subjects (<5) in some categories, and the highly unbalanced marginal

sums of the actual data set, Pearson’s chi-square test cannot be applied. It is better to use Fisher’s exact test.

Author response: Fisher’s exact test could not be made with too many groups or observations.

Reviewer comment : The extension of Fisher's exact test to the case where the two variables have any finite number of modalities, but greater than two, was first performed by G. H. Freeman and J.H. Halton in 1951. This test is sometimes referred as the Freeman-Halton or Fisher-Freeman-Halton test. So it is possible to calculate Fisher’s exact test with more than 2 modalities in one variable.

It is written in Table 1 that Ficher’s exact test was used. I think this is not the case for Diagnosis according the author response.

Author response: We thank you for the comment and apologize for the misunderstanding regarding the statistical analyses used in the manuscript. After the first comment wisely suggesting Fisher’s exact test as test when comparing categorical demographics (in table 1 and table 2), the authors applied this test together with an unpaired t-test (for continuous demographics). The authors recognize that this was not clearly stated in our responses to the review comments. As written in table 1 and table 2, and the statistical analyses section, Fisher’s exact tests were used.

Results

It is possible to have standardized decimal point in p value.

0.35 Could be witten 0.350.

Author response: All p-values are now standardized to contain three decimals.

There is a confusion in the article between total population (for me the 7310 subjects) and the study population (for me the 1258 subjects

Table 1: change the Table title by « Baseline characteristics of study population according to SARS-Cov-2 serology

Table 2: change the Table title by « Baseline characteristics of total population by study participation

Author response: Thank you. This has now been corrected as suggested.

Submitted filename: Rebuttal letter 02.02.22.docx

Click here for additional data file.

9 Feb 2022

SARS-CoV-2 seroprevalence among patients with severe mental illness: a cross-sectional study

PONE-D-21-26611R2

Dear Dr. Sass,

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,

Ray Borrow, Ph.D., FRCPath

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

21 Feb 2022

PONE-D-21-26611R2

SARS-CoV-2 seroprevalence among patients with severe mental illness: a cross-sectional study

Dear Dr. Sass:

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

Prof. Ray Borrow

Academic Editor

PLOS ONE