Psychological Impact of a Pandemic Widespread in Healthcare Workers: The Italian and Swiss Perspective Early After of CoVid-19 Outbreak

Background. We investigated the COVID19-related psychological impact in healthcare workers three weeks after its onset in Italy and in Italian-speaking regions of Switzerland. All professional groups of public hospitals in Italy and Switzerland were asked to complete a 38 questions online survey investigating demographic, marital and working status, presence of stress symptoms and need for psychological support. Results. Within 38 hours a total of 3,038 responses were collected. The subgroup analysis identied specic categories at risk according to age, type of work and region of origin. Critical care workers, in particular females, reported an increased number of working hours, decline in condence in the future, presence of stress symptoms and need for psychological support. People reporting stress symptoms and those with children declared a higher need for psychological support. Conclusions. The large number of participants in such a short time advocates for a high interest on topic among hospital workers. The COVID19 outbreak could have been and still be a repeated trauma for many health professionals, with risk of future psychiatric sequelae. It is of outstanding importance to implement short and long-term measures to mitigate impact of the emotional burden of this pandemic while at the same time dealing with its clinical challenges.


Background
The world has been (and is still) experiencing a mass casualty event [1], being stressful and intrusive by affecting daily life and putting adaptive skills to the test. Psychologically, during the early stages of the outbreak we have experienced the so-called alarm stage [2] because of a virus that con ned all of us at home (or at work, in the case of health personnels), putting social relationships, freedom and economy at strain. Feelings of inadequacy or insu ciency are common and one can experience a sense of being lost and confused [2]. Moreover, we are still subjected to constantly heavy load of information coming from the mass media and which is not at all times consistent or comprehensive [3].
After the initial stun, anxiety, irritability and restlessness come forth. There are some reactions that can be considered normal: physical reactions like palpitations or di culty breathing (expression of a sympathetic system activation, not of Coronavirus); cognitive reactions like disorientation, di culty giving a meaning to the pieces of information received and understanding the real severity of the event; emotional reactions like anguish, stunning, shock, anger, sadness [4][5][6][7] and fear of people one can comes across while grocery shopping, walking into a pharmacy or into a hospital (we could even feel stuck or deprived of the desire to carry out daily activities); behavioral reactions (di culty communicating, hyperactivity, ease to either verbal or physical confrontation, lack of functionality in the work place). One could also experience the inability either to unwind and relax, or to fall asleep or to sleep.
It is likely that everyone would have experienced marked stress reactions and that some will develop posttraumatic stress disorder (PTSD) [8]. The response to a traumatic event is subjective and dependent on numerous factors such as social relationships that can offer emotional support (friends, family, colleagues) and the kind of working activity [8]. Facing this largescale infectious public health crisis, medical staff are under both physical and psychological pressure or, at least, it can be hypothesized [2,9]. In this scenario, the mental status of healthcare professionals is dangerously jeopardized, because of both the key and front-line role they play and the risk of becoming infected [5,8].
The aim of this study was to investigate the psychological impact of CoVid-19 outbreak in hospital workers in the rst three weeks of outbreak in Italian speaking regions, such as northern Italy, where the European outbreak begun, and southern Switzerland, that followed few days later.

Demographics, social and working characteristics
The majority of our respondents were physicians, female and people aged 26-36. Details on age, gender, regions of origin and working characteristics are reported in Table 1 and Fig. 1.

Stress symptoms
Details are reported in Table 2. The decision tree analysis showed a higher probability of sleeping less than normal (accuracy 62.2%) in people aged > 30 from Italy and having one or more children in their family. In particular, nurses of the critical care with history of anxiety problems were the category at higher risk to suffer from sleep disturbances (accuracy 82.9%). Need for psychological support and con dence in the future In the last three weeks 706 people perceived to need a specialized psychological support (23.3%). Con dence in the future was diminished for 1404 persons (46.3%), increased for 246 subjects (8.1%) and did not change for 1384 people (45.6%). People feeling the need of professional psychological support were more often female (C.I. 24.8-28.3% vs. 13.0-17.1% for females and males respectively), in a couple (C.I. 23.7-27.4%) or subjects reporting stress symptoms (Fig. 2). Long-career workers (C.I. 7.1-15.0%) and ambulance services (C.I. 7.5-20.8%) expressed less often the need for psychological support. As reported in Fig. 3, different professional categories expressed different psychological support need. People stating an increased con dence in the future were more often males (C.I 5.4-11.6% vs. 9.2-30.2% for females). As expected, among people with diminished con dence in the future there was a higher prevalence of females (C.I 34.2-60.6% vs. 32.6%-45.7% for males).
Plotting the data by country, the con dence in the future in the last three weeks was more lowered in people from Italy as compared to Switzerland (C.I. 46.0-52.3% vs. 34.2-41.1%). This difference did not change by plotting data either by work category or by age or length of career. Surprisingly, by plotting the decreased con dence in the future for the type of hospital work, all categories were equally affected except of ambulance services (C.I. 16.1-33.6% vs. 40.0-51.2% for other categories).
Finally, history of anxiety resulted to be an independent factor associated to a reduced con dence in the future (C.I. 47.7-60.2% vs. 37.6-42.5% with and without history of anxiety respectively).

Discussion
The present study highlights elevated stress levels in a cohort of Italian-speaking hospital workers during the early phases of the outbreak of Sars-Cov2 infection due to the novel CoVid-19. To best of our knowledge, our study was the earliest to explore the psychological impact of this novel pandemic on healthcare workers among Italian speaking countries (Italy and Italian-speaking regions of Swizerland).
Hospital workers have engaged with catastrophic events caused by the massive CoVid-19 outbreak [11].
Beyond demanding clinical and logistic issues, hospital employees have to deal with their own physical and mental health [11][12][13][14][15]. Work-related stress has shown to impact physician's mental health, patients' care quality and the e ciency of the healthcare system [16]. Consequences in terms of mental health in the context of maxi-emergency situations may be even more pronounced than normal. Feelings of inadequacy, insu ciency, lostness or confusion are common and hospital workers may even experience negative behavioral reactions, depression and illness, possibly leading to lack of effectiveness and e ciency at work [4][5][6].
The large number of responses and the small percentage of uncompleted surveys (2,7%) in less than 40 hours advocates for a high interest in the subject among healthcare personnel. This may also express the need among healthcare workers to communicate, to be heard and understood that encompasses professional attitude and rules of conduct and touches the psyche and emotions.
More than 50% of survey participants were physicians, probably because the survey promoters were physicians both in southern Switzerland and northern Italy and invitations to participate were sent through personal contacts and social media. However, the high turnout rate of several other hospital categories could mirror the concept that the present is a common burden among all health workers.

Demographics, social and working distribution
The population aged 26-45 represents the majority of the sample, probably because it is are more likely to be reached by social media. The small number of people aged between 18-25 is likely to re ect the relatively low number of very young people working in hospitals as con rmed by the fact that medical trainees under 26-years of age are not yet medical doctors in Italy and Switzerland. The distribution of worked years in hospital roughly re ects the age distribution.
In our sample, two thirds of survey participants were female. This could be interpreted according to the fact that a growing percentage of medical professionals is represented, nowadays, by women. It could be also speculated that gender differences may play a role in stress perception and management. Hormonal differences in hypothalamic-pituitary-adrenal axis and autonomic nervous system response are wellknown. It has been hypothesized that females tend to experience more frequently anxious and depressive states (also with physical symptoms), being more aware of potential stressors rather than men [17] In our sample, percentages of married, single, divorced etc. and other demographic characteristics were grossly comparable to the Swiss and Italian population [18,19].
Nearly one third of hospital workers in our study had to increase their working activity. It represents the rst, well-recognized stress factor, proportional to the increase of worked hours [20]. As expected, health care professionals having to increase their working hours, are those employed either in critical care settings or general practitioners, the rst categories having to be confronted with the CoVid-19 emergency. On the other hand, in several hospitals in northern Italy and southern Switzerland a dramatic reduction of surgical elective cases and outpatient clinic activity has been deemed necessary to contain virus spread and, as a consequence, surgeons as a category experienced a reduction in working hours. Notwithstanding, a reduction of working hours should not be considered less insidious, as it could nonetheless jeopardize the mental status and could lead to depressive symptoms, hopelessness and uselessness [21].

Stress symptoms
Sleep disturbances can be caused by stress and be related to post-traumatic stress disorder and the rst response is generally considered a period of arousal and wakefulness [22,23]. A great proportion of hospital workers in our study declared to sleep less than normal and to feel less restored by sleep in general, presumably as a reaction to the stressful circumstances and it is likely to re ect the high prevalence of sleep arousal and anxiety symptoms in healthcare professionals.
The correlation analysis showed a relatively higher prevalence of stress symptoms among young professionals with at least one child to care of, as response of having to face the CoVid-19 outburst emergency.
Similarly, an increase or a decrease in in food intake may mirror a reaction to a stressful situation and, actually, in our sample, only 40% of the interviewed reported no change [24]. Moreover, among smokers, there was a high proportion of participants that declared to have increased the number of cigarettes per day in the rst three weeks since the outbreak onset. A smaller, but signi cant proportion of people reported an increase in alcohol consumption. All these behavioral changes can be considered stress symptoms [25,26]. In our study, a strong correlation was found between the above-mentioned symptoms and the need for psychological support. In fact, people reporting such symptoms were more prone to report the need for psychological support. As far as we know, this is the rst time that such a need is quanti ed within the emergency of CoVid-19.
Need for psychological support Interestingly, people with children reported the most stressful answers and felt the need for professional psychological support more often than their not-parenting peers. Despite some might consider this obvious, such data are nonetheless of outstanding importance, as hospital employees are often in the fertile age range. In the setting of a massive viral outbreak, hospital caregivers experience important issues in caring for their children because of the closure of schools and of other facilities. Parents probably do not experience only the stress related to future uncertainty but also the fear of getting infected and possibly transmit such an infectious disease to their offspring. Speci cally, CoVid-19 has shown to be particularly infective also for hospital workers and casualties have been reported among hospital staff [12]. Coherently, many hospital workers with children (nearly 95%) declared to have reduced their physical contacts with other family components.
Almost all subcategories of hospital workers perceive the stress related to the outbreak equally. Only longcareer workers and ambulance service personnel reported to need psychological support less as compared to other categories. This may be explained by the high level of experience and long-standing training in stress management in long-career employees. On the other hand, emergency services personnel may be used to address stressful situations, as they are part of a coordinated and ordered emergency response and have to constantly handle very high levels of stress [27,28].
The actual outbreak could be a repeated trauma for many healthcare categories, putting them at risk of psychiatric sequelae such as PTSD [8,29]. The present outbreak could be a repeated trauma for many health professionals and could put them at risk of developing psychiatric sequelae such as posttraumatic stress disorder. It is of outstanding importance in such a critical situation to promptly implement measures to mitigate the impact of the emotional burden of the present COVID-19 pandemic while at the same time dealing with its clinical challenges [2,4].
In particular, positive behaviors such as healthy eating, sport practicing and sleeping an adequate number of hours have shown to impact and reduce the impact of stress [30]. Many other strategies have also shown to be effective, such as the implementation of debrie ng sessions and group therapies to share experiences and relieve the sorrow related to challenging and stressful situations. Programs of deescalation of tension through mindfulness techniques could be also cost-effective and easily implemented in routine practice to prevent future development of acute and chronic PTSD, major depression or suicidal behaviors [31,32].
Our study has some limitations. Firstly, our data were collected in a completely anonymized and we have not certainty on their truthfulness. Nonetheless, our data denote a certain internal coherence, that can be interpreted as trustworthiness. Secondly, it can be considered an initial evaluation and further studies in different time points of the outbreak are needed to understand the impact of CoVid-19 emergency on healthcare workers' psychology and mental health. Thirdly, our sample is mainly (but not exclusively) composed of physicians and that could lead to a selection bias among the different hospital working categories and may not entirely re ect the actual global status.

Conclusions
Perceived psychological burden in hospital workers has to be considered relevant since the early phases of an infectious disease widespread. CoVid-19 outbreak has, therefore, to be considered as a major stress event, able to threaten the physiological and psychological integrity of individuals, in particular healthcare professionals. Counseling, psychological support and awareness on this subtle threat to hospital workers should not be overshadowed, as even hospital e ciency may be jeopardized.

Methods
Aim of the present study was to explore the psychological effect of CoVid-19 outbreak among healthcare workers in Italian speaking regions three weeks after the outbreak.

Study design and participants
Between 14th and 16th March 2020, hospital workers of different public hospital in Switzerland and Italy were asked to participate a survey regarding the psychological impact of CoVid-19 outbreak on the daily life, at the time point when the virus mainly widespread in northern Italy, Iran and China, but hadn't assumed an extensive dissemination in Switzerland yet. Participants were personally invited by the Authors to answer the questionnaire via social media (either Facebook or WhatsApp) and agreed voluntarily to the questionnaire being ful lled. All data were anonymized, so that, according to the actual Swiss legislation, no informed consent or institutional review board approval were needed. All staffed hospital teams were included, such as nurses, physicians, obstetricians, pharmacists, technical service, hospital canteen, laundry and others.

Questionnaire
The questionnaire (see Suppl. material available on line) was built on Google Forms (https://www.google.com/forms/about/) and consisted in 38 questions investigating demographic, social and working status (gender, age, civil status, total number of people and people > 65-year-old in the family, worked years in hospital, profession, department, working time changes since the beginning of the outbreak, chronic diseases worsening, alcohol drinking, drugs consumption, social media use, hobbies, history of anxiety and medical therapy), stress symptoms (alimentary behavior changes, new symptoms occurring in the last three weeks, smoking status and number of cigarettes/day, number of hours spent to get information and to speak about the CoVid-19 widespread, detachment from family due to the quarantine, stress behaviors in children, changes in physical activities, change in physical contacts) and need for psychological support (actual perception of the need of quali ed psychological support and overall trust in the future). We decided to not use validated questionnaires because they did not guarantee exibility and rapidity of use; furthermore explorative surveys are a method already adopted for other researches about pandemics [8]. All data were collected on an Excel database (Microsoft Excel® 2019) and analyzed.

Statistical methods
Statistical analysis was performed using the open source packages "Pandas," "NumPy," "SciPy," "Seaborn," and "PyMC" for Mac Os X versions 0.23.0, 11.1.3, 1.1.0, 0.8.1, and 2.3.6, respectively. Our analysis was based on highly credible intervals (C.I.) of parameter estimates (*); statistical signi cance was considered achieved whenever credible intervals would be non-overlapping. In order to compare proportions in different sub-populations, we performed a Bayesian estimate of the parameter distribution for a Bernoulli stochastic variable using a non-informative uniform prior. The posterior was then plotted in order to have a graphical overview, and credible intervals were computed by assessing the highest density interval (HDI) at 95%. Only as a reference, whenever two proportions had to be compared, we used also a two-tail P test computation based on classic proportion comparison using chi-square. Statistical signi cance was considered p < 0.05. All MCMC runs were checked for adequacy on the basis of Raftery-Lewis diagnostics and by visually inspecting Z score plots. MCMC runs were for 40,000 iterations with a 5,000-iteration burn-in [10].

Declarations
Ethics approval and consent to participate The study was conducted in full agreement with the national and international regulations, and the Declaration of Helsinki (2000). Participants' personal information, including names, were kept anonym to maintain and protect privacy and con dentiality of data. All participants were fully informed about the study requirements and were required to accept the data sharing and privacy policy before participating in the study. The anonymous nature of our survey did not allow tracing in any way sensitive personal data. Therefore, the present study did not require approval by Ethics Committee.

Consent for publication
Not applicable.

Data availability
Raw data supporting the conclusions of this manuscript will be made available by the authors, without undue reservation, to any quali ed researcher.

Competing interest
The authors have no con ict of interest to declare.

Funding
The authors received no funding to carry the present research.  Need for psychological support in participants with and without stress symptoms