Factors associated with psychological symptoms in hospital workers of a French hospital during the COVID-19 pandemic: Lessons from the first wave.

PURPOSE: The COVID-19 pandemic has put hospital workers around the world in an unprecedented and difficult situation, possibly leading to emotional difficulties and mental health problems. We aimed to analyze psychological symptoms of the hospital employees of the Paris Saint-Joseph Hospital Group a few months after the 1st wave of the pandemic. PARTICIPANTS AND METHODS: From July 15 to October 1, 2020, a cross-sectional survey was conducted among hospital workers in the two locations of our hospital group using the Hospital Anxiety and Depression Scale (HADS) and Post-Traumatic Stress Disorder (PTSD) Checklist (PCL) to measure anxiety, depression, and PTSD symptoms. Factors independently associated with these symptoms were identified.
RESULTS: In total, 780 participants (47% caregivers, 18% health administrative workers, 16% physicians, and 19% other professionals) completed the survey. Significant symptoms of anxiety, depression, and PTSD were reported by 41%, 21%, and 14% of the participants, respectively. Hierarchical regression analysis showed a higher risk of having psychological symptoms among those (1) who were infected by SARS-CoV-2 or had colleagues or relatives infected by the virus, (2) who retrospectively reported to have had an anxious experience during the first wave, and (3) with a previous history of burnout or depression. In contrast, job satisfaction appeared to be a protective factor. Overall, hospital workers showed the statistically same levels of anxiety, depression, and PTSD symptoms, regardless of their profession and whether they had worked in units with COVID-19 patients or not.
CONCLUSIONS: Our cross-sectional survey of 780 hospital workers shows that after the first wave, hospital employees had a significant burden of mental health symptoms. Specific preventive measures to promote mental well-being among hospital workers exposed to COVID-19 need to be implemented, first among particularly vulnerable staff, and then, for all hospital staff for whom anxiety is detected early, and not only those who were directly exposed to infected patients.

Introduction

On January 20, 2020, the World Health Organization (WHO) declared the disease caused by the novel coronavirus, severe acute respiratory syndrome coronavirus 2 (SARS-CoV2), called coronavirus disease 2019 (COVID-19), to be a public health emergency. Soon afterwards, as COVID-19 started to show the characteristics of rapid and wide transmission, the WHO re-classified the problem as a pandemic. With more than six million confirmed cases worldwide and more than 350,000 deaths declared between February and May 2020, this pandemic emerged as an unprecedented healthcare crisis.

Previous research has shown that, historically, exposure to pandemics is accompanied by mental health problems in the general population and healthcare workers, such as symptoms of post-traumatic stress disorder (PTSD), anxiety, and depression [1]. Such psychiatric symptoms among healthcare workers have been reported during the outbreak, but also away from the epidemic [2, 3].

WHO interim guidance from March 2020 [4] emphasized that the COVID-19 pandemic inevitably places healthcare workers at risk. While healthcare professionals must cope with the same societal shifts and emotional stressors faced by all people, they also face a greater risk of exposure, extreme workloads, moral dilemmas, and a rapidly evolving practice environment that differs greatly from that with which they are familiar.

In a meta-analysis of 13 studies (33,062 participants) evaluating the mental health of healthcare workers during the COVID-19 pandemic, the prevalence rate of anxiety and depression symptoms was 23.2% and 22.8%, respectively [5]. Unlike certain countries, France has not had to deal with other recent epidemics, such as SARS-1, Middle East Respiratory Syndrome, or Ebola. A French study, carried out among critical care clinicians managing patients with COVID-19 during the first wave, estimated that the prevalence of anxiety, depression, and peri-traumatic dissociation symptoms was 50.4%, 30.4%, and 32.0%, respectively [6].

Many hospital employees shared risk factors with healthcare workers because of their direct exposure to infected patients, the limited availability of protective equipment, and the increased workload related to the pandemic. It is, thus, likely that non-clinical support staff, such as administrative assistants, security, and environmental services personnel, were also subject to the same effects on their mental health. Specific literature on this subject is, however, scarce.

Our hospital group (Groupe Hospitalier Paris Saint-Joseph) is composed of two hospitals (Hospital Paris Saint-Joseph and Hospital Marie-Lannelongue). During the first wave (between March 1 and May 29, 2020) 1,177 patients were managed in the emergency ward, 834 were hospitalized, 132 needed intensive care, and 100 died.

Here, we aimed to determine the risk factors associated with mental health symptoms (anxiety, depression, and PTSD) after the first wave of the COVID-19 pandemic among all components of the workforce in our hospital group. The findings of such a study could be informative for decision making during future waves of the pandemic and potential pandemics caused by other agents that may follow.

Methods

The study sponsor was the “Groupe Hospitalier Paris Saint-Joseph”. The protocol was performed in accordance with the Declaration of Helsinki and approved by the institutional ethics committee: the GERM (Groupe Ethique et Recherche Médicale) from the Hospital Paris Saint-Joseph (IRB number IRB00012157 -n° initial agreement 436 and registered on the national institute of health data platform).

No written informed consent was required. The authors guarantee the anonymization of all data collected.

The study started on July 15, 2020, three months after the peak of the pandemic in France, and finished on October 1, 2020. All hospital workers were invited to complete an online survey, with a link sent to the mailing list. The use of a “Microsoft Forms” electronic format allowed us to securely send the survey and store the answers. Informed consent was obtained from all participants on the first page of the survey.

To be included, the only criteria was to have worked in the hospital between March 15 and May 15, 2020.

The survey included five items that were identified from a literature review [3, 7–9] and from interviews with Groupe Hospitalier Paris Saint-Joseph workers. These items were: (the entire questionnaire can be found in the S1 Appendix)

Individual characteristics (demographics, exposure to COVID-19, history of burnout)

Professional experience and fears at the time of the survey

Emotional experience during the first wave and the degree of job satisfaction:

Did the COVID-19 crisis make you anxious? If so: for you, your family, others, or in your work?

Were you afraid of infecting your loved ones?

Currently, do you feel good at work?

The Hospital Anxiety and Depression Scale (HADS). The HADS is a 14-item auto questionnaire that includes seven items about symptoms of anxiety and seven items on symptoms of depression [10]. A cutoff score > 7 was used for each subscale to detect significant symptoms of anxiety or depression [11, 12].

The PTSD Checklist (PCL). The PCL is a 17-item self-reported measure reflecting DSM-IV symptoms of PTSD [13]. A cutoff score > 44 was used to detect significant symptoms of PTSD [14]. It was made clear in the questionnaire that the stressful event to which the respondents had to refer was the health crisis. At the time of the survey, it was thought that the pandemic was over and had been over for more than a month.

Statistical analysis

Data are described as numbers or percentages. Variables were compared using either χ2-test or Fisher exact tests, as appropriate.

Independent predictors for anxiety, depression, and PTSD symptoms were assessed using logistic regression. Odds ratios (ORs) and their 95% confidence intervals (95% CIs) were calculated using univariate and multivariate logistic regression models. We conducted several logistic regressions. We first calculated the unadjusted models and then conducted multiple logistic regression analyses using a hierarchical approach. The first model included socio-demographic variables and the second the same variables plus variables related to the emotional experience during the first wave of the COVID-19 pandemic (personal experience of anxiety, fear of infecting loved ones, fear of being infected by SARS-CoV-2, and job satisfaction). We also assessed potential interactions.

Variables of interest were selected according to their statistical significance in univariate analysis (critical p-value for entry into the model < 0.1) or were forced into the model (age and sex).

Analyses were performed using R software (the R project for statistical computing, https://www.r-project.org/). All tests were two-sided and a p-value < 0.05 was considered statistically significant.

Results

Participant characteristics (Table 1)

Among the 3,272 employees working in the two locations of our hospital group, 780 participated in the survey. Among them, 1,675 worked at the Hospital Paris Saint-Joseph (which has 2,408 employees), and 105 at the Hospital Marie-Lannelongue (which has 864 employees) (Table 1).

Table 1

Respondents’ characteristics and reports about their COVID-19 experience.

Variable (N = 780)		N (%)
Gender (N = 780)
• Women		638 (81.8%)
• Men		142 (18.2%)
Age (years) (N = 780)
• ≤ 41		429 (55.0%)
• > 41		351 (45.0%)
Marital situation (N = 747)
• Single		235 (31.5%)
• Living as a couple		512 (68.5%)
Familial situation (N = 471)
• No children		72 (15.3%)
• One or several children		399 (84.7%)
Location of professional practice (N = 780)
• Hospital Paris Saint-Joseph		675 (86.5%)
• Hospital Marie-Lannelongue		105 (13.5%)
Profession (N = 780)
• Caregiver (nurse, assistant nurse, nurse manager)		367 (47.1%)
• Healthcare administrator		141 (18.1%)
• Medical professional (physician, pharmacologist, biologist)		125 (16.0%)
• Other caregivers (physiotherapist, stretcher-bearer, radiological technologist, psychologist)		72 (9.2%)
• Midwife		21 (2.7%)
• Others		54 (6.9%)
Professional experience in the same hospital unit (years) (N = 780)
	< 5	437 (56.0%)
	≥ 5	343 (44.0%)
History of professional burnout or depression (N = 780)
No		605 (77.6%)
Yes		175 (22.4%)
Duty station during the COVID-19 crisis (N = 780)
• COVID Unit		375 (48.1%)
• Non-Covid Unit		347 (44.5%)
• Other professional activity		143 (18.3%)
• Remote work		105 (13.5%)
COVID-19 patient management (N = 780)
• Frequently / Regularly		361 (46.3%)
• Rarely / Never		419 (53.7%)
Respondents who were infected or with colleagues or relatives who were infected (N = 780)
Yes		466 (59.7%)
No		314 (40.3%)

The largest proportion (47%) of healthcare staff were caregivers (nurses, assistant nurses, nurse managers), followed by 18% healthcare administrators (secretaries, logistic managers, pharmaceutical assistants), and 16% medical doctors (physicians, pharmacists, or medical technicians).

Among the participants, 56% had been in the same workplace for less than five years. A previous history of burnout or depression was reported by 22%. Forty-eight percent reported having worked in a unit with COVID-19 patients during the first wave, 44% in another clinical unit, and the remaining participants were distributed among different types of units.

Almost half of the participants (46%) reported having frequently worked with COVID-19 patients (every working day) or regularly (at least once a week). Nearly 60% reported that a colleague, friend, or close relative had a SARS-CoV-2 infection.

Emotional experience during the COVID-19 first wave

Among the respondents, 62% reported that they were anxious during the period of the first wave; most (86%) were anxious for their family, 57% for themselves, 51% at work, and 42% for others. Participants were more afraid of infecting their loved ones (75%) than being infected themselves by SARS-CoV-2 (45%).

Three quarters of hospital workers indicted they had high job satisfaction at the moment of the survey.

HADS and PCL results

Clinically significant levels of anxiety symptoms (HADS-Anxiety = HADa > 7) were found for 41% of the respondents, depression symptoms (HADS-Depression = HADd > 7) for 21%, and PTSD symptoms (PCL > 44) for 14%.

In unadjusted models, several factors were significantly associated with the odds of probable anxiety, depression, and PTSD symptoms to varying extents (Tables 2–4 and Table A-C in S1 Table). There were fewer men in the group with symptoms of anxiety than in the group without (13% vs 22%, p = 0.001), as well as in the group with symptoms of PTSD than that without (11% vs 19%, p = 0.029). However, there were as many men in the group of workers with symptoms of depression as in the group without (17% vs 19%, p = 0.73). There were fewer workers who were married or partnered in the group with symptoms of PTSD than in the group without (56% vs 71%, p = 0.003). In the group with symptoms of anxiety, there were fewer medical professionals than in the group without (11% vs 20%, p = 0.033). There were fewer healthcare workers working in a COVID-19 area in the group with symptoms of depression than in the group without (40% vs 50%, p = 0.023). There were more employees who were infected or had colleagues or relatives who were infected with SARS-CoV-2 in the group with symptoms of anxiety than in the group without (68% vs 54%, p < 0.001), as well as in the group with symptoms of PTSD relative to the group without (74% vs 57%, p < 0.001).

Table 2

Unadjusted and adjusted logistic regression for symptoms of anxiety (HADa > 7), n = 780.

	Unadjusted		Model 1*		Model 2**
	OR [95%CI]	p	OR [95%CI]	p	OR [95%CI]	p
Gender: Women (vs Men)	1.97 [1.32;2.93]	0.001	1.70 [1.11;2.64]	0.015	1.48 [0.90;2.45]	0.12
Age (years) ≤ 41 (vs > 41)	0.90 [0.67;1.19]	0.45	0.89 [0.65;1.20]	0.44	0.94 [0.66;1.33]	0.72
Marital situation: Living as a couple (vs single)	0.78 [0.57;1.07]	0.12
Familial situation: One or several children vs no children	1.01 [0.61;1.68]	0.98
Profession (vs Healthcare administrator)		0.033
• Medical professional	0.45 [0.27;0.76]		0.56 [0.32;0.96]	0.035	0.54 [0.29;1.00]	0.05
• Caregiver	0.88 [0.60;1.31]		0.87 [0.58;1.31]	0.50	0.76 [0.47;1.21]	0.24
• Other Caregiver	0.74 [0.42;1.33]		0.75 [0.41;1.37]	0.36	0.97 [0.48;1.93]	0.92
• Midwife	0.58 [0.22;1.54]		0.57 [0.20;1.51]	0.27	0.40 [0.12;1.24]	0.12
• Other	1.01 [0.54;1.89]		1.15 [0.60;2.21]	0.67	1.35 [0.63;2.89]	0.43
Location of professional practice: Hospital Marie-Lannelongue (vs Hospital Paris Saint-Joseph)	0.96 [0.63;1.47]	0.86	0.96 [0.61;1.49]	0.85	0.81 [0.48;1.34]	0.41
Professional experience in the same hospital area: > 5 years vs < 5 years	0.94 [0.71;1.25]	0.68
Duty station during the COVID-19 crisis
COVID-area assignment	0.79 [0.60;1.06]	0.11
Non COVID-area assignment	0.97 [0.73;1.29]	0.83
Remote work	1.38 [0.91;2.09]	0.12
Non-clinical professional activity	1.03 [0.71;1.48]	0.89
COVID-19 patient management: regularly / frequently vs never/rarely	0.79 [0.59;1.05]	0.102
Respondents who were infected or with colleagues or relatives who were infected: yes vs no	1.75 [1.30;2.36]	< 0.001	1.63 [1.19;2.23]	0.002	1.55 [1.09;2.23]	0.016
Anxiety during the 1st Wave: yes vs no	6.85 [4.77;9.84]	< 0.001			5.83 [3.89;8.90]	< 0.001
• Anxiety for oneself: yes vs no	1.18 [0.82;1.69]	0.37
• Anxiety for family: yes vs no	1.03 [0.61;1.74]	0.92
• Anxiety for others: yes vs no	0.92 [0.64;1.33]	0.67
• Anxiety at work: yes vs no	1.43 [1.00;2.05]	0.049
Fear of contaminating relatives during the 1st wave: yes vs no	2.50 [1.75;3.58]	< 0.001			1.64 [1.04;2.59]	0.034
History of professional burnout or depression: yes vs no	2.94 [2.08;4.16]	< 0.001	2.79 [1.96;4.00]	< 0.001	2.63 [1.75;3.97]	< 0.001
Date of history of depression or burnout (N = 175): > 3 years vs < 3 years	0.53 [0.28;1.00]	0.05
Current job satisfaction: yes vs no	0.20 [0.14;0.28]	< 0.001			0.19 [0.13;0.29]	< 0.001

Model 1 included socio-demographic variables and model 2 the same variables plus variables related to the emotional experience during the 1st wave of the COVID-19 pandemic (personal experience of anxiety during the 1st wave due to fear of infecting loved ones or being infected by SARS-CoV-2 and job satisfaction).

HADa = HAD-Anxiety.

*Nagelkerke Pseudo-R2: 0.11.

**Nagelkerke Pseudo-R2: 0.37.

Table 4

Unadjusted and adjusted logistic regression for symptoms of post-traumatic stress (PCL > 44), n = 780.

	Unadjusted		Model 1*		Model 2**
	OR [95%CI]	p	OR [95%CI]	p	OR [95%CI]	p
Gender: Women (vs Men)	1.99 [1.06;3.73]	0.029	1.68 [0.91;3.35]	0.12	1.33 [0.69;2.77]	0.41
Age (years) ≤ 41 (vs > 41)	0.77 [0.51;1.17]	0.22	0.85 [0.55;1.31]	0.46	0.83 [0.52;1.31]	0.43
Marital situation: Living as a couple (vs single)	0.53 [0.35;0.81]	0.003	0.55 [0.35;0.84]	0.006	0.50 [0.32;0.80]	0.003
Familial situation: One or several children vs no children	0.97 [0.47;2.01]	0.94
Profession (vs healthcare administrator)		0.12
Medical professional	0.47 [0.21;1.04]
Caregiver	1.10 [0.65;1.87]
Other Caregiver	0.58 [0.24;1.44]
Midwife	1.27 [0.39;4.14]
Other	0.68 [0.26;1.77]
Location of professional practice: Hospital Marie-Lannelongue (vs Hospital Paris Saint-Joseph)	1.19 [0.68;2.10]	0.55	1.41 [0.74;2.55]	0.27	1.29 [0.66;2.43]	0.44
Professional experience in the same hospital unit: > 5 years vs < 5 years	1.01 [0.67;1.51]	0.97
Duty station during the COVID-19 crisis
COVID-area assignment	1.17 [0.78;1.74]	0.46
Non COVID-area assignment	1.07 [0.72;1.60]	0.74
Remote work	0.92 [0.50;1.67]	0.78
Non-clinical professional activity	0.55 [0.30;1.01]	0.05	0.62 [0.32;1.14]	0.14	0.73 [0.36;1.39]	0.36
COVID-19 patients management: Regularly / Frequently vs never/rarely	1.03 [0.69;1.54]	0.90
Respondents who were infected or with colleagues or relatives who were infected (yes vs no)	2.10 [1.34;3.29]	0.001	2.10 [1.33;3.41]	0.002	1.99 [1.22;3.33]	0.007
Anxiety during the 1st Wave: yes vs no	6.67 [3.51;12.66]	<0.001			4.47 [2.35;9.27]	< 0.001
• Anxiety for oneself: yes vs no	1.27 [0.81;1.99]	0.30
• Anxiety for family: yes vs no	1.32 [0.66;2.62]	0.43
• Anxiety for others: yes vs no	1.14 [0.73;1.78]	0.55
• Anxiety at work: yes vs no	1.87 [1.19;2.94]	0.006
Fear of contaminating relatives during the 1st wave: yes vs no	3.15 [1.69;5.88]	< 0.001			2.52 [1.26;5.55]	0.014
History of professional burnout or depression: yes vs no	2.02 [1.31;3.12]	0.001	1.80 [1.13;2.83]	0.011	1.47 [0.89;2.39]	0.12
Date of history of depression or burnout (N = 175): > 3years vs < 3 years	0.68 [0.33;1.40]	0.29
Current job satisfaction: yes vs no	0.24 [0.16;0.36]	< 0.001			0.28 [0.18;0.44]	< 0.001

Model 1 included socio-demographic variables and model 2 the same variables plus variables related to the emotional experience during the 1st wave of the COVID-19 pandemic (personal experience of anxiety during the 1st wave due to fear of infecting loved ones or of being infected by SARS-CoV-2 and job satisfaction).

*Nagelkerke Pseudo-R2: 0.13.

**Nagelkerke Pseudo-R2: 0.28.

In the group with symptoms of anxiety, more hospital workers declared that they were anxious during the first wave of the pandemic than in the group without symptoms of anxiety (86% vs 46%, p = 0.001); the same was true for symptoms of depression (76% vs 59%, p < 0.001) and PTSD (90% vs 58%, p < 0.001). In addition more employees reported a history of burnout or depression in the group with symptoms of anxiety than in the group without (34% vs 15%, p < 0.001); the same was true for symptoms of depression, (39% vs 18%, p < 0.001) and PTSD (34% vs 21%, p = 0.001). Finally, there were fewer employees who had high job satisfaction in the group with symptoms of anxiety than in the group without (57% vs 87%, p < 0.001) and the same results were found for symptoms of depression (46% vs 82%, p < 0.001) and PTSD (48% vs 79%, p < 0.001).

Multivariable analysis

The relationship between the female gender and anxiety remained significant, although smaller in magnitude, after adjusting for demographic characteristics, working conditions, and a history of professional burnout or depression (OR, 1.70; 95% CI: 1.11–2.64) (Table 2, model 1). However, the association between gender and anxiety (Table 2, model 2) was no longer significant after adjusting for the emotional experience of the first wave (OR, 1.48; 95% CI: 0.90–2.45), suggesting that this variable may substantially explain the association.

We observed a comparable pattern for the relationship between the medical profession and anxiety, which remained significant (Table 2, model 1) after adjusting for demographic characteristics, working conditions, and a history of professional burnout or depression (OR, 0.56; 95% CI: 0.32–0.96). However, adjusting for the emotional experience of the first wave made this relationship non-significant (OR, 0.54; 95% CI: 0.29–1.00).

Finally, the relationship between a history of professional burnout or depression and PTSD remained significant after adjusting for demographic characteristics and the working condition (OR, 1.80; 95% CI: 1.13–2.83) (Table 4, model 1) but not after adjusting for the emotional experience of the first wave (OR, 1.47; 95% CI: 0.89–2.39) (Table 4, model 2).

In multivariable analysis (model 2), healthcare workers who retrospectively reported that they felt anxious during the first wave showed more symptoms of anxiety (OR, 5.83; 95% CI: 3.89–8.90), depression (OR, 1.70; 95% CI: 1.11–2.65), and PTSD (OR, 4.47; 95% CI: 2.35–9.27) at the moment of the survey. Hospital workers with a history of burnout or depression had an increased risk of anxiety (OR, 2.63; 95% CI: 1.75–3.97) and depression (OR, 2.25; 95% CI: 1.47–3.43).

The single variable associated with a lower risk of all three studied mental health symptoms (model 2) was high job satisfaction (OR, 0.19; 95% CI: 0.13–0.29 for anxiety; OR, 0.20; 95% CI: 0.13–0.29 for depression; and OR, 0.28; 95% CI, 0.18–0.44 for PTSD).

Being married or partnered was independently associated with a lower risk of PTSD symptoms (Table 4, model 2) (OR, 0.50; 95% CI: 0.32–0.80).

Discussion

Our cross-sectional survey, which included 780 hospital workers, identified a prevalence of probable anxiety, depression, and PTSD after the first wave of the COVID-19 pandemic of 41%, 21%, and 14%, respectively. The following three factors were independently associated with the presence of the three types of psychological symptoms: those who reported to have been anxious during the first wave, those with a previous history of burnout or depression, and those who had themselves been infected or who had colleagues or relatives who were infected. This suggests that at the beginning of any epidemic emergency, screening for ongoing psychological problems among healthcare workers should be carefully conducted by hospital management to protect those who are the most vulnerable as a primary preventive measure.

Our survey also showed that high job satisfaction had a positive influence, resulting in better mental health scores.

Other studies in the literature concerning healthcare workers during the COVID-19 pandemic reported a prevalence of anxiety from 23% to 51% and depression from 22% to 50% [15-17] based on assessment measurements. A systematic review of 29 studies reported a median prevalence of 24% for anxiety and 21% for depression among healthcare workers [18].

Our hospital workers had a higher prevalence of anxiety symptoms than the general French population over the same period but a lower frequency of depressive symptoms based on results published on the “Santé Publique France” website [19]. In an August 2020 survey, approximately 18% of the French population interviewed reported anxiety (21% for our hospital workers for the same HAD scale cut-off as in the survey) and 11% depression (9% for our hospital workers).

The prevalence of COVID-19 related-PTSD symptoms among healthcare workers has been reported to be 71% [20], 21% [21], and 15% [22] versus our rate of 14%. This difference in the prevalence is likely linked to the instruments we used to assess symptoms and the moment when the survey was carried out: during the peak of the pandemic versus after the end of the first wave. Indeed, several studies carried out after the SARS and MERs epidemics found significant rates of PTSD several months after the epidemic [2, 23]. Our survey was conducted from 2 to 6 months after the first wave.

We found no difference in the prevalence of psychological symptoms between administrative staff and healthcare workers or between staff working in zones with COVID-19 patients and those without. Employees working in the two hospitals showed the same levels of emotional suffering after the first wave, whether they were directly exposed to patients or not. Many previous studies have reported a higher prevalence of emotional symptoms among front-line healthcare workers [24, 25]. However, the results of several recent studies are in accordance with ours, with an identical emotional impact on healthcare workers directly exposed to COVID-19 patients land those who are not [23, 26, 27]. Liang et al. [28] even showed that frontline nurses working with COVID-19 patients had significantly lower traumatization scores than non-frontline nurses, the general public, or medical teams aiding COVID-19 control efforts. Indeed, worries related to the fear of being infected or higher-than-usual exposure to death among COVID-19 units might have been balanced by the higher availability of protection equipment and greater feeling of usefulness. As the impact of constraints on mental health can be offset by higher rewards [28], the societal praise of healthcare professionals working in COVID-19 units may have acted as a protective factor, much like the level of job satisfaction. Another explanation may be that the anxiety caused by the sudden population lock-down was more dominant than factors intrinsic to their hospital workplace, as suggested by Milgrom et al [26].

One factor that was independently associated with higher symptoms of anxiety, depression, and PTSD was to have been anxious during the first wave. We certainly cannot draw conclusions about the relationship between anxiety during the event and the occurrence of distant emotional symptoms based on the results of a single study using such a retrospective measure by a dichotomous question. In addition, such a relationship could have resulted from the simple fact that people with a higher level of general anxiety are prone to have been more anxious in the past. However, this observation could lead to a prospective study to evaluate the impact of secondary prevention strategies targeting workers who feel the most anxious during a crisis.

Another factor associated with a significantly increased risk of showing adverse psychological outcomes was having been infected by the virus, or having colleagues or relatives who had been infected. This factor has already been found in other studies [25, 29] and suggests that this population would be more vulnerable to emotional distress and should receive specific support. A high risk of infection may leave workers feeling vulnerable because COVID-19 is highly infectious, has a high morbidity rate, and is potentially fatal [28].

We observed that employees who had already had depression or a history of burnout had more symptoms of anxiety and depression. This result is not surprising. It has been shown that people with a pre-existing mental health disease were among the groups at highest risk for a range of psychiatric distress symptoms during the COVID-19 pandemic [30]. Moreover, those with anxiety-related disorders reported greater fears about danger and contamination and traumatic stress symptoms [31].

On the contrary, we observed a lower prevalence for the three studied mental health disorders among respondents who reported high job satisfaction. Accordingly, Wang et al. showed that job satisfaction was a factor related to PTSD of nurses exposed to COVID-19 in China [32]. Mental health is highly relevant to work satisfaction. Healthcare workers who are dissatisfied with their job often feel that they are working in a dysfunctional system that affects the quality of their tasks and their self-esteem, factors associated with a higher risk of developing symptoms of anxiety, depression, and stress [20].

We used a hierarchical strategy in the multivariable analysis to observe the impact of the emotional experience during the crisis on the studied variables: the risk of symptoms of anxiety was higher for women in model 1, but by adding the emotional experience as a variable in model 2, anxiety and female gender were no longer significantly associated. Thus, although female gender was strongly associated with symptoms of anxiety, the causal pathway for this relationship was likely to have been through the emotional experience during the crisis.

We found similar results for participants with a history of burnout or depression, who had more PTSD symptoms. When we adjusted for the variable of emotional experience during the crisis, a history of burnout or depression was no longer associated with PSTD symptoms. The relationship between a history of burnout or depression and PTSD also appears to have been mediated by the emotional experience of the crisis.

The impact of the personal experience of anxiety during the crisis was particularly strong when the respondents reported that their anxiety was focused on their job. There appears to be a strong relationship between such anxiety just at the time of the first wave (reported retrospectively by the respondents) and the prevalence of significant symptoms of anxiety, depression, and PTSD afterwards. Thus, by identifying those employees who feel the most anxious during a health crisis it may be possible to implement secondary prevention strategies targeted to these staff members in the hope of reducing the prevalence of post-crisis mental health disorders.

In our study, medical professionals reported less anxiety. Resilience, which is defined as the capacity to cope with and positively adapt to adversity, is an important protective factor and is of particular concern to researchers in the field of adversity [28]. It has been demonstrated during the COVID-19 epidemic that resilience can help to reduce worry, anxiety, and depression [33, 34]. Medical staff members have a higher level of education, which is positively related to resilience [35]. They also generally have greater decisional latitude at work, which is a protective factor against professional stress.

Marital status appeared to be predictive of high levels of PTSD symptoms, as shown in previous studies that examined the mental health effects of the SARS outbreak [24, 36, 37].

To improve the mental bell-being of hospital workers during health crises, many authors have suggested that special care should be taken to address the level of anxiety, depression, and PTSD symptoms among both health personnel and public service providers, such as by ensuring clear communication, adequate supplies of protective equipment, and access to psychological intervention [33]. They also propose that during breaks, staff should be provided with food and other daily living needs [16]. It has been noted that concrete measures to develop rest areas/break rooms and the possibility of having leisure and moments of relaxation are more appropriate to addressing the needs of caregivers than formal psychological support [25, 38]. Thus, informal mechanisms may be more successful [39] in which, for example, counsellors or retired nurses visit healthcare workers in rest areas [28].

Future longitudinal research is needed to evaluate the medium- and long-term psychological impact of the pandemic on hospital workers and to identify patterns and the co-occurrence of risk factors for adverse mental health outcomes. Intervention studies in real-world settings should be additionally conducted to investigate under which interventions and specific circumstances resilience may be best fostered and the mental health of frontline professionals supported during and after a disease outbreak.

Our study had several limitations. First, it was based on a single hospital group in Greater Paris, limiting the generalizability of our findings to locations less affected by the pandemic in our country. Second, our survey was conducted after the first wave before knowing that there would be subsequent waves afterwards, which may affect the interpretation of our results. Third, as is common for such surveys, there was probably a selection bias for responders due to the sampling methods. Fourth, the scales we used allowed us to evaluate the level of anxiety, depression, and PTSD symptoms, but cannot alone be used to diagnose these disorders. Finally, the cross-sectional nature of the data also limited our ability to assess whether there could be a causal relationship between the health crisis and mental health outcomes, as no pre-COVID data were available. Indeed, French studies assessing the prevalence of anxiety and depressive symptoms among healthcare workers before the pandemic found high levels, similar to our results, and therefore raise the question of the aggravation of these disorders by the COVID-19 health crisis [40].

Notwithstanding, our study also had several strengths: 1/ the size of our sample, 780 hospital workers, assured that the study was sufficiently powered; 2/ we had a 29% response rate to our survey, which is close to published rates [41-43], especially those published during the COVID-19 pandemic [44-46]; 3/ the homogeneity of the sample, as all respondents were working in the same university hospital in Greater Paris, which was severely affected by the first wave of the pandemic; and 4/ the distribution of our sample is representative of the hospital worker population and included every type of profession, including administrative staff.

Conclusions

We conducted a cross-sectional survey study to assess socio-emotional factors associated with probable anxiety, depression, and PTSD among 780 hospital workers a few months after the first wave of the COVID-19 pandemic in France.

The level of mental health symptoms found in our study does not appear to be much higher than during the pre-pandemic period. In addition, by including all employees in the survey, it is apparent that the risk of mental suffering for hospital workers was similar, regardless of the professional category (frontline healthcare or non-clinical workers) and whether or not there was direct exposure to ill patients.

However, specific interventions to promote mental well-being in hospital employees exposed to a health crisis such as the COVID-19 pandemic should be implemented as primary prevention. Particular attention should be given to those who are infected or have colleagues or relatives who are infected and those with a history of burnout or depression. Moreover, as secondary prevention, it may be useful to identify hospital workers who feel especially anxious during the crisis, because they appear to be more vulnerable to mental distress afterwards.

Finally, despite the heavy workload and demands that it entails, hospital workers in general are satisfied with their profession, which is a protective factor against psychological pathologies.

Submitted questionnaire for the survey.

(DOCX)

Click here for additional data file.

Univariable analysis of factors associated with the presence of emotional symptoms.

(DOCX)

Click here for additional data file.

Full database.

(XLSX)

Click here for additional data file.

14 Jul 2021

PONE-D-21-17297

Factors associated with psychological symptoms in hospital workers of a French hospital during the COVID-19 pandemic: lessons from the first wave.

PLOS ONE

Dear Dr. d'Ussel,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

My comments are the following:

With so many predictors an issue of overfitting is a real danger. Can you, please, report on the percentage of variance explained in the DVs (any pseudo R2, e.g., Nagelkerke).

What about multicollinearity among the predictors? Was this analysis performed (e.g., Tabachnik and Fidell, 2013)?

Some decisions appear arbitrary or not elaborated properly. For example, why age variable was categorized instead of used as it is? Since you categorized it from the beginning - even in the questionnaire (i.e., it was not done post-hoc) - there must be some reasons for it, that were not elaborated enough. Furthermore, you decided to use the first category (age 18-25) as a referential value. I wonder why, why not the last one? Anyway, you should elaborate on it and justify other decisions in binarizing/categorizing your variables (prof. experience, job satisfaction etc.)

Generally, why did you decide to use cut off criteria for anxiety, depression, and PTSD and binarize them instead of exploit the full amount of information in these variables as the continual ones? You had rightly mentioned as one of the limitations of your work that these were not proper diagnoses, so why did you choose to behave as they were? Please, elaborate.

In order to make tables more readable I suggest at least to skip unadjusted model or model 1.

Please submit your revised manuscript by Aug 28 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

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:  http://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,

Goran Knezevic

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 including your ethics statement:

"The protocol has been approved by the institutional ethics committee IRB number IRB00012157 (n° initial agreement 436 and registered on national institute of health data platform).

No written informed consent was required. The authors guarantee the anonymization of all data collected. "

Please amend your current ethics statement to include the full name of the ethics committee/institutional review board(s) that approved your specific study.

Once you have amended this/these statement(s) in the Methods section of the manuscript, please add the same text to the “Ethics Statement” field of the submission form (via “Edit Submission”).

For additional information about PLOS ONE ethical requirements for human subjects research, please refer to http://journals.plos.org/plosone/s/submission-guidelines#loc-human-subjects-research.

3. You indicated that informed consent was not necessary for your study. Could you please provide further details on why your study is exempt from the need for consent and confirmation from your institutional review board or research ethics committee (e.g., in the form of a letter or email correspondence) that ethics review was not necessary for this study? Please include a copy of the correspondence as an "Other" file.

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. Please amend either the title on the online submission form (via Edit Submission) or the title in the manuscript so that they are identical.

6. Please include a copy of Table 2 which you refer to in your text on page 10.

[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: Yes

Reviewer #2: Partly

**********

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

Reviewer #1: Yes

Reviewer #2: 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

**********

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

**********

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 topisc of the paper is relevant and actual aiming to analyze the mental health problems and emotional status of hospital workers during pandemic of COVID-19. Methods are appropriate, the results and discussion are well done, and limitations of the study is properly recognized.

The Table 1. and 2. are followed by the description which is redundant, so the tables could be deleted or the text should point out only important findings.

Reviewer #2: The manuscript “Factors associated with psychological symptoms in hospital workers of a French hospital during the COVID-19 pandemic: lessons from the first wave” presents data collected during the first wave of the COVID19 pandemic in French hospitals. Findings suggest that healthcare workers during the COVID19 pandemic are at heightened risk for anxiety, depression, and PTSD.

I find the paper timely due to the very heavy burden under which health care professionals are. However, the presented study has several shortcomings, and the majority of my concerns are related to the methodological aspects of the study. In addition, the study is designed to be purely exploratory, and the impact and the relevance of the study is not highlighted at all.

The major problem is the questionnaire used to collect data on socio-demographics. Its structure, type of questions (majority of items are binary and some are multiple responses), and information collected prevent us from making solid judgments and conclusions – this is extremely important as these items are major predictors in the study. Some items (especially those about COVID19 infection) are ambiguous and nothing can be deduced based on the respondents’ answers – were respondents infected or whether they infected someone else? These should have been kept separate to allow for making conclusions. I am missing the rationale why the majority of items are categorical. For example, respondents were grouped by age in several groups, and this also applies to other variables like working experience, patient management, etc. – my advice is to use the variables as continuous instead of categorical variables. Do two hospitals from which participants were recruited differ in some aspects of work or they were selected at random? Is there some reason to believe that healthcare workers would perceive and experience the pandemic differently in these two hospitals?

Importantly, items investigating mental health status before the COVID19 pandemic are not precise, some of them are ambiguous, and we cannot make any solid conclusion about mental health during the pandemic or how prior experience affected current mental health status. Thus, the main aim of the study worded as assessing the psychological impact of the COVID19 pandemic is not feasible at all – the study is cross-sectional and even more, items assessing mental health before the pandemic do not allow us to make some conclusions about it.

I strongly advise changing the analytical strategy. Running so many analyses ended up in having findings compartmentalized, fragmented, and we are missing the big picture about the mental health outcomes in healthcare workers. I suggest running a model in which variables would be included simultaneously to allow for a comprehensible overview of the study variables and their relationships.

Tables are huge and not very user-friendly – try to present data in a more intelligible fashion. Text describing tables is very detailed and redundant.

There are some paragraphs in the discussion section that appear to be disconnected from the rest of the text. For example, the third paragraph on p. 23 “So it’s suggested that concrete measures to develop rest rooms and the possibility of having leisure and moments of relaxation are more appropriate to the needs of caregivers than formal psychological support(22,36), and informal mechanisms might be more successful(37) where, for example, counsellors or retired nurses visit healthcare workers in rest areas(25).” How is this paragraph related to the rest of the discussion?

Language needs proofreading, please check the manuscript carefully – some grammar errors are present in the manuscript.

Overall, due to a large number of issues in the questionnaire used to collect data that I listed above, I cannot recommend this paper for publishing

**********

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: Yes: Ljiljana Lazarevic

[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.

7 Sep 2021

Response to academic editor:

We agree with the reviewer. We added the percentage of variance explained (Nagelkerke pseudo R²).

As you will see on the following tables 1, 2 and 3 there is no over fitting. Moreover we use 2 by 2 tables to test relationship between variables. This lead us to suppress some variables: Professional experience linked with Professional experience in the same hospital unit, and Usual job satisfaction linked with Current job satisfaction.

To elaborate the questionnaire and to decide which variable to include in it, and furthermore how to use these variables (in binarizing/categorizing them), we have studied the previous literature with attention. We have reviewed the literature about the psychological impact of previous epidemics on healthcare workers, and their associated factors. Thus, we have used the different items as they were used in this literature: age (Chong et al. 2004; Wu et al. 2009), work experience (Chong et al. 2004), job title (Chong et al. 2004), marital status(Chong et al. 2004; Lin et al. 2007; Su et al. 2007; 2007; 2007; Wu et al. 2009), past history of psychological disorder (Su et al. 2007), working units (Chong et al. 2004; Su et al. 2007)…

This is always an arbitrary choice. Usually, when a variable is ordered we choose the highest or the lowest category. As regard age we finally binarized the variable, according to the limit corresponding roughly to the median. We also considered only 2 sites (Hospital Paris Saint-Joseph and Hospital Marie-Lannelongue) because most employees working at both hospitals worked mainly in Paris Saint-Joseph. As regard profession, we chose arbitrary as baseline administrative healthcare because those persons belonging to these profession were the less expose to Covid-19.

After discussion with psychiatric experts we have chosen the HAD scale and PCL to determine anxiety, depression and PTSD symptoms even if these scales were not considered as diagnostic scales but as screening tests: the HADS is quick to fill, has a good validity in its French translation, and can be used among any population. According to this review (Bjelland et al. 2002) it must be used by binarization with a cut off. The best advantage with the PCL is that it summarized the characteristics of latest DSM classification for PTSD. This scale is also used with a published binarized cut off (Yao et al. 2003)

In clinical epidemiology it is usual practice to present both adjusted and unadjusted OR. We therefore maintain the unadjusted model in the table.

When using a hierarchical model it is also the way to present results (Gazmararian et al. 2000)

Table 1 : VIF for each regression logistic for symptoms of anxiety

Model 1 Model 2

Age 1.07 1.07

Gender 1.03 1.04

Profession 1.11 1.20

Place of professional practice 1.03 1.04

Respondents who were infected or having infected colleagues or relatives 1.02 1.04

Professional burn-out or depression history 1.01 1.04

Anxiety during 1st Wave 1.11

Fear of contaminating relatives during the 1st wave 1.13

Current job satisfaction 1.06

Table 2 : VIF for each regression logistic for symptoms of depression

Model 1 Model 2

Age 1.06 1.07

Gender 1.02 1.04

Place of professional practice 1.04 1.04

Remote work 1.09 1.09

COVID-Unit assignment 1.13 1.13

Marital situation 1.03 1.03

Respondents who were infected or having infected colleagues or relatives 1.04 1.03

Professional burn-out or depression history 1.02 1.03

Anxiety during 1st Wave 1.03

Current job satisfaction 1.03

Table 3 : VIF for each regression logistic for symptoms of post-traumatic stress

Model 1 Model 2

Age 1.04 1.05

Gender 1.02 1.03

Place of professional practice 1.05 1.05

Non-clinical professional activity 1.02 1.04

Marital situation 1.03 1.04

Respondents who were infected or having infected colleagues or relatives 1.02 1.03

Professional burn-out or depression history 1.02 1.04

Anxiety during 1st Wave 1.05

Fear of contaminating relatives during the 1st wave 1.05

Current job satisfaction 1.01

Response for Journal Requirements:

The study has been done in accordance with the Declaration of Helsinki and approved by the GERM (Groupe Ethique et Recherche Médicale/ Ethics and Medical Research Group) from the Hospital Paris Saint-Joseph (IRB number 00012157).

This research is part of the institutional care for its employees. In this context, the French regulations (JORF n°0160 of 13 July 2018 text n°110, MR-004) do not require consent but require the transmission of an information note to the employees setting out the purpose of the research. The employees’ non-opposition to the use their data for research purposes is also collected in accordance with the European General Data Protection Regulation (GDPR).

Certification of ethical opinion from the institutional ethics committee is provided as a complementary document.

Data are contained within the Supporting Information files, and available in Excel format on direct request to the first author : mdussel@ghpsj.fr

Response to Reviewer #1:The manuscript has been modified to avoid redundancies, and table 2 has been deleted

Responses to Reviewer#2: To elaborate the questionnaire and choose the variables, we have studied the literature about the psychological impact of previous epidemics on healthcare workers (Chong et al. 2004; Wu et al. 2009; Lin et al. 2007; Su et al. 2007). Especially for the item about COVID-19 infection, we have based on this article (Wu et al. 2009) where it was asked if the healthcare workers had been infected or had friends or relatives who had been infected

To elaborate the questionnaire and to decide which variable to include in it, and furthermore how to use these variables (in binarizing/categorizing them), we have studied the previous literature with attention. We have reviewed the literature about the psychological impact of previous epidemics on healthcare workers, and their associated factors. Thus, we have used the different items as they were used in this literature: age (Chong et al. 2004; Wu et al. 2009), work experience (Chong et al. 2004), job title (Chong et al. 2004), marital status(Chong et al. 2004; Lin et al. 2007; Su et al. 2007; 2007; 2007; Wu et al. 2009), past history of psychological disorder (Su et al. 2007), working units (Chong et al. 2004; Su et al. 2007)…

Although belonging to the same group, the two hospitals have different specificities, and the recruitment of patients is therefore different. For example, Hospital Marie Lannelongue does not have an emergency department. It was not impacted in the same way during the 1st wave, since 635 patients were hospitalized at Hospital Paris Saint Joseph compared to 199 at Hospital Marie Lannelongue. Therefore, one would have thought that the prevalence of anxiety, depression and PTSD symptoms among employees would be different in the two hospitals.

We agree with this comment, and we notified it in the discussion, line 341: Last, this cross-sectional study cannot demonstrate that COVID-19 is responsible for additional psychological burden in frontline healthcare professionals, as no pre- COVID data are available

As the title of the article indicates, the main objective of the study was to identify factors associated with the mental status of hospital workers. We have modified the last sentence of the introduction to make clear that this objective is the main one. The short title has been also modified

We agree with the reviewer that there are many variables. Thus, as suggested for statistical problem we suppressed some of them and we grouped categories (e.g. age is now binarized as compared with 6 categories in the previous version of the paper). Our strategy was based on the strategy published by (Gazmararian et al. 2000) using a classification variables: demographical, professional and emotional.

We simplified the tables by regrouping categories for some variables and suppressing redundant ones. We also simplified text to avoid any redundancy between the results chapter and tables.

Finally, we suppressed table 2.

We agree partly with the reviewer. The text is not directly connected to our results. It is more adapted to a perspective section, where we moved it.

The manuscript has been carefully checked to correct grammar errors

We respect the opinion of the reviewer. However, our goal was to identify factors predictive of mental health impairment at the end of the first wave. We do think that our results obtain with many variables offer perspectives for prevention choices during subsequent waves.

Submitted filename: response to reviewers.docx

Click here for additional data file.

1 Nov 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,

Goran Knezevic

Academic Editor

PLOS ONE

[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: All comments have been addressed

Reviewer #2: (No Response)

**********

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 #2: Yes

**********

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

Reviewer #1: Yes

Reviewer #2: 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

**********

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

**********

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 topic of the manuscript is relevant and actual. The authors have adequately addressed r comments of reviewers, so I believe that the manuscript is now acceptable for publication. There still are some minor suggestions:

In the abstract, in 32 line, the authors have written Post-traumatic Stress Distress (PTSD) instead Post-traumatic Stress Disorder

The subtitle Materials and Methods, line 94- It would be recommended to stay only Methods (without Materials)

Reviewer #2: I have reviewed the previous version of the paper “Factors associated with psychological symptoms in hospital workers of a French hospital during the COVID-19 pandemic: lessons from the first wave”. The authors have improved the manuscript, and I would like to thank them for that. However, in their letter with responses they listed responses to the comments without including original comments of reviewers, making it very difficult to say which response correspond to which comment. Therefore, I suggest authors to include both comments of the reviewers and responses in the same file.

Despite manuscript being improved, there are still numerous issues that have to be solved. The most important thing relates to the writing itself – numerous sentences are ambiguous and not clear, several are imprecise, and this must be resolved. This makes manuscript very difficult to read, and impedes making judgements about the results. I’ll state some of the examples:

Abstract – “Significant symptoms of anxiety (41%), depression (21%) and PTSD were reported by 41%, 21% and 14% of the participants, respectively” – not clear why these percentages repeat?

Abstract – “There was no difference regarding type of occupation or assignment in a COVID-19 unit.” – there was no difference in what?

p. 10 – “Compared with female gender, male gender was associated with a significantly lower prevalence of symptoms of anxiety (87% vs. 13%; P=0.001) and PTSD (89% vs. 11%; 178 P=0.029), but a comparable prevalence of depression (83% vs. 17%; P=0.73). “ – check percentages for depression.

p. 10 – “Medical professionals had lower prevalence of anxiety symptoms than other workers (11%, P=0.033).” – lower prevalence than other workers, how are these other workers? Sentence is not clear.

p. 10 – “Healthcare workers not working in a COVID-19 Unit were more prone to depression (60% vs. 40%; P=0.023)” – more prone to depression compared to whom?

p. 16 – “Regarding emotional experience at the time of the first wave, the hospital employees who reported to have been anxious during the first wave had more anxiety (86% vs. 14%; p<0.001)more depression (76% vs. 24%; P<0.001) and more PTSD symptoms (90% vs. 10%; P<0.001) in post crisis than those who have not reported anxiety.” – what is post crisis , sentence is not clear.

p. 16 – “Compared to patients without history 211 of burnout or depression those having such a history had greater anxiety (34% vs. 66%; P<0.001) , depression (61% vs. 39%; P<0.001), and PTSD symptoms (34% vs. 66%; P=0.001).” – please be consistent, either write bigger vs smaller numbers, or smaller vs bigger.

p. 17 – “In multivariable analysis (tables 2, 3 and 4), healthcare workers who reported to have been anxious during the first wave reported more anxiety (OR, 5.83; 95% CI: 3.89-8.90), depression (OR, 1.70; 95% CI: 1.11–2.65) and PTSD symptoms (OR, 4.47; 95% CI: 2.35–9.27)” – it seems that something is wrong with this statement – those who said that are anxious reported they were more anxious – it’s circular, check this.

p. 21 –“ It appears to be a strong relationship 318 between this emotional experience at the time of the first wave and the prevalence of significant symptoms of anxiety, depression and PTSD afterwards.” Perhaps I’m missing something, but this sentence is not clear to me – the paper investigates mental health in the first wave of the pandemic, so I don’t understand to what “significant symptoms of anxiety, depression and PTSD afterwards” refer?

p. 21 – “The fact that introducing this variable in the model resulted in rendering some predictive variables (e.g. gender) non-significant strongly suggests that anxiety at the time of the first wave explains a substantial part of the association between vulnerability factors (e.g. history of depression) and subsequent symptoms of anxiety, depression and PTSD” – again, I don’t understand this sentence. What is the variable introduced in the model? ‘what is the connection between gender, history of depression, and anxiety during the first wave of the pandemics? And how this information informs us about secondary prevention strategies?

The study explores mental health issues after the first wave, but throughout the text authors compare mental health before the first wave, during the first wave and after the first wave. This creates confusion as there are no data to compare mental health during and after the first wave.

Table 1 – number of participants responding to specific items differs, and it would be good to include this information consistently for each item – at the moment, for some items information is given, for some it is omitted.

Table 2 and Table 3 – explain acronyms in the title, HADa and HADd in the notes.

Table 3 – check cell Place of professional practice, duplicated text

Please don’t start sentences with numbers (percentages).

To conclude, the major issue preventing me from recommending the manuscript for publication is the writing which has to be substantially improved.

**********

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: Prof. Tamara Dzamonja Ignjatovic, PhD

Reviewer #2: Yes: Ljiljana Lazarevic

[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.

2 Dec 2021

-Scientific reports should be simple and easy to understand, sentences more precise, without ambiguities.

+We have revised the entire text to make it more clear, readable and unambiguous

-After removing adjusted coefficients you still preserved the same table descriptions (did not remove word "adjusted"). Please, be careful with such things.

+We try to make more explicit the statistical methods paragraph.

We also modified legend of tables 2, 3 and 4.

-However, the central issue related to your overall findings is the following: you neither obtained higher prevalence of the studied symptoms during Covid19 pandemics (when compared to pre-pandemic conditions, e.g. Hardy et al., 2019), nor you obtained the difference in these symptoms between those working in Covid-19 units and other units. So, in the light of what you have reported, it is closer to the truth that none of the professional categories were affected by the pandemic, than that everyone was affected "not just front-line or even clinical staff" (p. 26). +That’s true; our study doesn’t allow us to know if the first wave has impacted the hospital workers emotionally.

But we can observe that the level of mental suffering is the same whatever the professional category and the proximity of involvement with patients and their families during the 1st wave.

So we have changed the conclusion in order to be more precise.

-In other words, the relationships that you have obtain between your predictors and symptoms, mostly reveal the usual pattern of symptom correlates independent of the pandemic.

+Exact. And that is interesting according to us! One could have imagined that the frontline workers had higher levels of mental suffering after the 1st wave; our study doesn’t show that.

We have modified the discussion to emphasize that notion

-Even what appears as something uniquely related to the pandemics - the level of anxiety in the first wave - can reflect nothing more than individual differences in trait anxiety of the participants (especially having in mind that is was measured retrospectively and very roughly with one dichotomous question). You mentioned it sporadically in limitations, but you should make it more salient to a reader. I would recommend to devote full attention to this issue and to elaborate on it properly.

+We agree with your remark and we added a section about this notion in the discussion:

One factor which was independently associated with higher symptoms of anxiety, depression and PTSD was having had a personal experience of anxiety during the first wave. But it’s difficult to conclude with this observation; the statistical result is interesting and it could be a track of further prospective work to implement strategies of secondary prevention among the most anxious workers during the crisis. On the other hand our retrospectively measure, with a dichotomy question, is not sufficient to conclude about the relation between anxiety during the event and occurrence of distant emotional symptoms.

Comments Reviewer 1

- The topic of the manuscript is relevant and actual. The authors have adequately addressed r comments of reviewers, so I believe that the manuscript is now acceptable for publication. There still are some minor suggestions:

In the abstract, in 32 line, the authors have written Post-traumatic Stress Distress (PTSD) instead Post-traumatic Stress Disorder

+Done

-The subtitle Materials and Methods, line 94- It would be recommended to stay only Methods (without Materials) +Done

Comments Reviewer 2

-I have reviewed the previous version of the paper “Factors associated with psychological symptoms in hospital workers of a French hospital during the COVID-19 pandemic: lessons from the first wave”. The authors have improved the manuscript, and I would like to thank them for that. However, in their letter with responses they listed responses to the comments without including original comments of reviewers, making it very difficult to say which response correspond to which comment. Therefore, I suggest authors to include both comments of the reviewers and responses in the same file.

+We are sorry that you didn’t access to the adequate version of our responses

The file called “point-by-point response to reviewer” included a table with the comments in the colon 1 and our responses in the colon 2.

We will be attentive that you can access to this form for the second rewieving.

-Despite manuscript being improved, there are still numerous issues that have to be solved

The most important thing relates to the writing itself – numerous sentences are ambiguous and not clear, several are imprecise, and this must be resolved

This makes manuscript very difficult to read, and impedes making judgements about the results

+We edited the manuscript a second time to make it more precise and to avoid any ambiguity.

-Abstract – “Significant symptoms of anxiety (41%), depression (21%) and PTSD were reported by 41%, 21% and 14% of the participants, respectively” – not clear why these percentages repeat?

+Correction:

Significant symptoms of anxiety, depression and PTSD were reported by 41%, 21% and 14% of the participants, respectively.

-Abstract – “There was no difference regarding type of occupation or assignment in a COVID-19 unit.” – there was no difference in what?

+There was no difference among workers in anxiety, depression and PTSD symptoms regarding type of occupation or assignment in a COVID-19 unit.

-p. 10 – “Compared with female gender, male gender was associated with a significantly lower prevalence of symptoms of anxiety (87% vs. 13%; P=0.001) and PTSD (89% vs. 11%; 178 P=0.029), but a comparable prevalence of depression (83% vs. 17%; P=0.73). “– check percentages for depression We have verified the figures. There is no mistake in the tables. But inappropriate choice of figures was done in the text of results section. This choice does not change the interpretation of the p-value.

+We corrected it.

-p. 10 – “Medical professionals had lower prevalence of anxiety symptoms than other workers (11%, P=0.033).” – lower prevalence than other workers, how are these other workers? Sentence is not clear. We compared the different professional categorizations by regrouping different similar professions in a same group, to limit the number of variables: medical professionals (physician, pharmacologist, biologist), administrative healthcare workers, caregivers (nurse, assistant nurse, nurse manager), others caregivers (Physiotherapist, stretcher-bearer, Radiologic Technologist, Psychologist), midwifes and others workers.

+The reviewer is right, we clarified the sentence. (line 150, table 1)

-p. 10 – “Healthcare workers not working in a COVID-19 Unit were more prone to depression (60% vs. 40%; P=0.023)” – more prone to depression compared to whom? Tables are correct. Inappropriate choice of figures was done in the text of results section. This choice does not change the interpretation of the p-value.

+We corrected the results section.

-p. 16 – “Regarding emotional experience at the time of the first wave, the hospital employees who reported to have been anxious during the first wave had more anxiety (86% vs. 14%; p<0.001)more depression (76% vs. 24%; P<0.001) and more PTSD symptoms (90% vs. 10%; P<0.001) in post crisis than those who have not reported anxiety.” – what is post crisis , sentence is not clear.

+Tables are correct. Inappropriate choice of figures was done in the text of results section. This choice does not change the interpretation of the p-value. We corrected the results section.

-p. 16 – “Compared to patients without history 211 of burnout or depression those having such a history had greater anxiety (34% vs. 66%; P<0.001) , depression (61% vs. 39%; P<0.001), and PTSD symptoms (34% vs. 66%; P=0.001).” – please be consistent, either write bigger vs smaller numbers, or smaller vs bigger

+Tables are correct. Inappropriate choice of figures was done in the text of results section. This choice does not change the interpretation of the p-value. We corrected the results section.

-p. 17 – “In multivariable analysis (tables 2, 3 and 4), healthcare workers who reported to have been anxious during the first wave reported more anxiety (OR, 5.83; 95% CI: 3.89-8.90), depression (OR, 1.70; 95% CI: 1.11–2.65) and PTSD symptoms (OR, 4.47; 95% CI: 2.35–9.27)” – it seems that something is wrong with this statement – those who said that are anxious reported they were more anxious – it’s circular, check this.

+We note that this notion of anxiety during the 1st wave is not clear and we have specified it in the entire document:

In multivariable analysis (tables 2, 3 and 4), healthcare workers who retrospectively reported that they felt anxious during the 1st wave had more symptoms of anxiety (OR, 5.83; 95% CI: 3.89-8.90), depression (OR, 1.70; 95% CI: 1.11–2.65) and PTSD symptoms (OR, 4.47; 95% CI: 2.35–9.27) at the moment of the survey.

-p. 21 – “It appears to be a strong relationship 318 between this emotional experience at the time of the first wave and the prevalence of significant symptoms of anxiety, depression and PTSD afterwards.” Perhaps I’m missing something, but this sentence is not clear to me – the paper investigates mental health in the first wave of the pandemic, so I don’t understand to what “significant symptoms of anxiety, depression and PTSD afterwards” refer?

+Our survey had been diffused around 3 months after the end of the 1st wave in our hospital located in Paris, without knowing that other waves would come later.

We wanted to investigate the emotional status of the hospital workers after this event that was the 1st wave.

Our study shows there is a relationship between those who retrospectively felt anxious during the 1st wave, and the prevalence of symptoms of anxiety, depression and PTSD some weeks later:

It appears to be a strong relationship between this anxious experience just at the time of the first wave (reported retrospectively by the responders) and the prevalence of significant symptoms of anxiety, depression and PTSD afterwards.

-p. 21 – “The fact that introducing this variable in the model resulted in rendering some predictive variables (e.g. gender) non-significant strongly suggests that anxiety at the time of the first wave explains a substantial part of the association between vulnerability factors (e.g. history of depression) and subsequent symptoms of anxiety, depression and PTSD” – again, I don’t understand this sentence. What is the variable introduced in the model? ‘what is the connection between gender, history of depression, and anxiety during the first wave of the pandemics? And how this information informs us about secondary prevention strategies?

The study explores mental health issues after the first wave, but throughout the text authors compare mental health before the first wave, during the first wave and after the first wave. This creates confusion as there are no data to compare mental health during and after the first wave.

+We have totally re-written this section in order to let it more readable:

By using a hierarchical strategy in the multivariable analysis, we observed the role played by the anxious experience on the studied variables: the risk of symptoms of anxiety was higher among women in the model 1, but if we enter the emotional experience as variable in the model 2, symptoms of anxiety and gender were no longer associated. Thus, although woman gender is associated with symptoms of anxiety, the causal pathway for these relation is likely to be through the emotional experience during the crisis.

We found similar results for participants with a history of burnout or depression who had more PTSD symptoms: when we adjusted with the variables of emotional experience during the crisis, history of burnout or depression was no longer related to PTSD symptoms. The relationship between history of burnout or depression and PTSD appears to be mediated by the emotional experience of the crisis.

The impact of the personal experience of anxiety during the crisis was particularly strong, especially when the respondents said that this anxiety was focused on their job. There is a strong relationship between this anxious experience just at the time of the first wave (reported retrospectively by the responders) and the risk of symptoms of anxiety, symptoms of depression and PTSD afterwards. So we can consider that identifying that staffs who feel most anxious during a health crisis and implementing secondary prevention strategies targeted at these staff, we could reduce the prevalence of post-crisis mental health disorders.

-Table 1 – number of participants responding to specific items differs, and it would be good to include this information consistently for each item – at the moment, for some items information is given, for some it is omitted.

+When the number of participants was 780 (total), we didn’t notified it; but if you think is clearer we notified for each item

-Table 2 and Table 3 – explain acronyms in the title, HADa and HADd in the notes.

+done

-Table 3 – check cell Place of professional practice, duplicated text

+done

-Please don’t start sentences with numbers (percentages).

+done

Submitted filename: Response to reviewers.docx

Click here for additional data file.

30 Dec 2021

1. I would like you to take suggestion on improvement of the language, by r#2, seriously. It means that you have to have proofread done by the professional translator. Without the substantial improvement of the readability of the text, we will not be able to accept the manuscript. The reason is that, generally speaking, your sentences are ambiguous and, frequently, difficult to understand.

2. It also seems to me that you did not give due consideration to my comments regarding the way you generally interpreted your results. To reiterate, my understanding is that your overall findings are in accordance with the following scenario: the crisis did not affect the level of symptoms you assessed in health care workers in any substantial manner (namely the prevalence of the symptoms among health care workers did not seem to be higher compared to pre-pandemic conditions); as a consequence, the differences between subgroups of health care workers were not found; relationships between anxiety during the first wave (retrospectively measured) and depression and anxiety could result from the simple fact that people with the higher level of trait anxiety are prone to be more anxious in the past, as well as to the tendency of the people with higher trait anxiety to retrospectively report higher level of anxiety (even if they did not have the level of anxiety they tend to report, which is the limitation of the cross-sectional nature of the study). Obviously, the crucial feature of such a scenario is that it is unrelated to pandemics. You have address this issue in the limitations paragraph (rows 425-430), but I urge you to mention it in conclusions. For example, you can add the first sentence in Conclusions something like...but it does not appear to be elevated significantly compared to the the pre-pandemic conditions.

Please submit your revised manuscript by Feb 13 2022 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

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,

Goran Knezevic

Academic Editor

PLOS ONE

[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: All comments have been addressed

Reviewer #2: (No Response)

**********

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 #2: Yes

**********

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

Reviewer #1: Yes

Reviewer #2: 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

**********

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

**********

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: All comments of the reviewers were addressed adequately and carefully considered by the authors. The manuscript have been significantly improved- more clearly written and conclusions are more precisely conducted from the results.

Reviewer #2: I appreciate that the authors tried to improve the manuscript and revise the text. The major drawback of this manuscript is the quality of the language. I understand that the authors tried to improve the manuscript, but it is of absolute importance that the paper be rewritten by a professional (native speaker of the English language). It is not possible to list all problems I noticed, and I flagged only some. If the text is not edited and proofread, I cannot recommend its publication. I understand this feedback is disappointing but my intention is only to make the manuscript better, and I hope that authors will interpret my feedback positively.

Below I list some of the language issues I detected.

Lines 106-107: The entire hospital workers were invited to complete a survey online, with a link sent to the mailing list.

Instead of ENTIRE it should be ALL.

Lines 107-108: This was a questionnaire in "Microsoft forms" electronic format that allows the secure sending and storage of responses.

“THIS” is not clear. I understand what you wanted to say, but it is not grammatically correct. Rewrite the sentence.

Lines 108-109: Inclusion criteria were for the responders to have worked during the period from March 15 to May 15 and to be a volunteer to complete the survey.

This sentence has to be rephrased, wording is odd.

Lines 114-115: They were: (the whole of the submitted questionnaire is accessible as supporting information 115 S1 Appendix).

Again, language is not okay.

Lines 130-131: It was made clear in the questionnaire that the stressful event to which respondents were referring was the health crisis (considered to have been over a month old at the time of the survey).

“over a month old” – not clear to what you refer. This has to be revised.

Lines 170-171: Half of the participants (46%) reported having taken care of patients with COVID-19 frequently (every worked day) or regularly (at least one day a week).

Every working day, not worked day. Also, one day in a week, or once a week.

Lines 171-172: 60% reported that a colleague, a 172 friend, or close relative had been infected by SARS-CoV-2.

The sentence should not start with a number.

Lines 174-176: Among the hospital employees, 62% reported that they were anxious during the period of 1st wave: 86% for their 176 family, 57% for themselves, 51% at work and 42% for the others.

The sentence is not clear. The colon makes it very unclear, how were these percentages obtained. Revise the sentence.

Lines 189-190: This proportion was comparable in the presence or absence of symptoms of depression (17% vs 19%; p=0.73).

I don’t understand this sentence.

Lines 248-254: There were more hospital employees who retrospectively reported at the moment of the survey that the first wave made them anxious in the group with symptoms of anxiety than in the group without symptoms of anxiety (86% vs 46%; p = 0.001), the same was true for symptoms of depression (76% vs 59%; p<0.001). There were more employees who reported a history of burnout or depression in the group with symptoms of anxiety than in the group without symptoms of anxiety (34% vs 15%; p < 0.001), the same was true for symptoms of depression (39% vs 18%; p < 0.001), and PTSD (34% vs 21%; p=0.001).

These sentences are not clear. ? Also, they appear to be circular – more people with anxiety symptoms is in the group with symptom of anxiety. Where there more people with symptoms of depression in the group of people with anxiety or in the group with depression? Second sentence is also not clear to me – what was true for the symptoms of depression and PTSD?

Lines 357-359: But it’s difficult to conclude with this observation; the statistical result is interesting and it could be a track of further prospective work to implement strategies of secondary prevention among the most anxious workers during the crisis.

OBSERVATION is not the appropriate word. Perhaps you can say: We cannot made final conclusion based on results of one study, …

To conclude, it is very difficult to get a complete picture on the quality of the results and discussion when a lot of text is not completely understandable.

Please do your best to improve the language of the text.

**********

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: No

Reviewer #2: Yes: Ljiljana Lazarevic

[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.

4 Feb 2022

1. I would like you to take suggestion on improvement of the language, by r#2, seriously. It means that you have to have proofread done by the professional translator. Without the substantial improvement of the readability of the text, we will not be able to accept the manuscript. The reason is that, generally speaking, your sentences are ambiguous and, frequently, difficult to understand.

-The manuscript has been entirely corrected by a professional translator.

2. It also seems to me that you did not give due consideration to my comments regarding the way you generally interpreted your results. To reiterate, my understanding is that your overall findings are in accordance with the following scenario: the crisis did not affect the level of symptoms you assessed in health care workers in any substantial manner (namely the prevalence of the symptoms among health care workers did not seem to be higher compared to pre-pandemic conditions); as a consequence, the differences between subgroups of health care workers were not found; Obviously, the crucial feature of such a scenario is that it is unrelated to pandemics. You have address this issue in the limitations paragraph (rows 425-430), but I urge you to mention it in conclusions. For example, you can add the first sentence in Conclusions something like...but it does not appear to be elevated significantly compared to the the pre-pandemic conditions.

-We modified every sentence in the manuscript which could still suggest that there was a causal relationship between the pandemic and mental health outcomes.

We added it clearly in conclusion.

Relationships between anxiety during the first wave (retrospectively measured) and depression and anxiety could result from the simple fact that people with the higher level of trait anxiety are prone to be more anxious in the past, as well as to the tendency of the people with higher trait anxiety to retrospectively report higher level of anxiety (even if they did not have the level of anxiety they tend to report, which is the limitation of the cross-sectional nature of the study).

-That’s true; and we wrote it clearly in discussion (line 312)

The major drawback of this manuscript is the quality of the language. I understand that the authors tried to improve the manuscript, but it is of absolute importance that the paper be rewritten by a professional (native speaker of the English language). It is not possible to list all problems I noticed, and I flagged only some. If the text is not edited and proofread, I cannot recommend its publication. I understand this feedback is disappointing but my intention is only to make the manuscript better, and I hope that authors will interpret my feedback positively.

Below I list some of the language issues I detected.

-We understood that major obstacle and a professional translator has proofread the entire manuscript. We thank the reviewer for having listed the major issues and suggested corrections

Submitted filename: Response to reviewers.docx

Click here for additional data file.

1 Apr 2022

Factors associated with psychological symptoms in hospital workers of a French hospital during the COVID-19 pandemic: lessons from the first wave.

PONE-D-21-17297R3

Dear Dr. d'Ussel,

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,

Goran Knezevic

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

20 Apr 2022

PONE-D-21-17297R3

Factors associated with psychological symptoms in hospital workers of a French hospital during the COVID-19 pandemic: lessons from the first wave

Dear Dr. d'Ussel:

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 Goran Knežević

Academic Editor

PLOS ONE

Table 3

Unadjusted and adjusted logistic regression for symptoms of depression (HADd > 7), n = 780.

	Unadjusted		Model 1*		Model 2**
	OR [95%CI]	p	OR [95%CI]	P	OR [95%CI]	p
Gender: Women (vs Men)	1.08 [0.69;1.71]	0.73	1.04 [0.65;1.71]	0.86	0.88 [0.53;1.50]	0.64
Age (years) ≤ 41 (vs > 41)	1.14 [0.80;1.61]	0.47	1.09 [0.75;1.59]	0.65	1.24 [0.83;1.85]	0.30
Marital situation: Living as a couple (vs single)	0.70 [0.48;1.01]	0.05	0.72 [0.49;1.06]	0.09	0.75 [0.50;1.13]	0.16
Familial situation: One or several children vs no children	1.21 [0.63;2.31]	0.56
Profession (vs healthcare administrator)		0.17
• Medical professional	0.64 [0.36;1.15]
• Caregiver	0.61 [0.38;0.96]
• Other Caregiver	0.97 [0.51;1.85]
• Midwife	0.29 [0.06;1.28]
• Other	0.77 [0.37;1.63]
Location of professional practice: Hospital Marie-Lannelongue (vs Hospital Paris Saint-Joseph)	1.38 [0.86;2.23]	0.20	1.40 [0.82;2.32]	0.20	1.22 [0.69;2.12]	0.48
Professional experience in the same hospital area: > 5 years vs < 5 years	1.16 [0.82;1.64]	0.40
Duty station during COVID-19 crisis
• COVID-area assignment: yes vs no	0.67 [0.47;0.95]	0.023	0.72 [0.49;1.07]	0.10	0.71 [0.47;1.07]	0.10
• Non COVID-area assignment: yes vs no	1.24 [0.88;1.76]	0.22
• Remote work: yes vs no	1.65 [1.03;2.62]	0.034	1.54 [0.92;2.54]	0.09	1.41 [0.81;2.42]	0.22
• Non-clinical professional activity: yes vs no	0.91 [0.58;1.44]	0.70
COVID-19 patient management: regularly / frequently vs never/rarely	0.88 [0.62;1.25]	0.48
Respondents who were infected or with colleagues or relatives who were infected (yes vs no)	1.31 [0.91;1.88]	0.14	1.33 [0.91;1.96]	0.14	1.22 [0.81;1.83]	0.35
Anxiety during 1st Wave: yes vs no	2.22 [1.49;3.29]	<0.001			1.70 [1.11;2.65]	0.017
• Anxiety for oneself: yes vs no	1.11 [0.73;1.69]	0.61
• Anxiety for family: yes vs no	1.04 [0.57;1.91]	0.89
• Anxiety for others: yes vs no	0.95 [0.63;1.44]	0.81
• Anxiety at work: yes vs no	0.98 [0.65;1.47]	0.91
History of professional burnout or depression: yes vs no	2.88 [1.97;4.19]	< 0.001	2.53 [1.70;3.74]	<0.001	2.25 [1.47;3.43]	< 0.001
Current job satisfaction: yes vs no	0.18 [0.12;0.26]	< 0.001			0.20 [0.13;0.29]	< 0.001

Model 1 included socio-demographic variables and model 2 the same variables plus variables related to the emotional experience during the 1st wave of the COVID-19 pandemic (personal experience of anxiety during the 1st wave due to fear of infecting loved ones or of being infected by SARS-CoV-2 and job satisfaction).

HADd = HAD-Depression.

*Nagelkerke Pseudo-R2: 0.14.

**Nagelkerke Pseudo-R2: 0.27.