Ethical and psychosocial considerations for hospital personnel in the Covid-19 crisis: Moral injury and resilience

This study aims at investigating the nature of resilience and stress experience of health care workers during the COVID-19 pandemic. Thirteen healthcare workers from Italian and Austrian hospitals specifically dealing with COVID-19 patients during the first phase of the pandemic were interviewed. Data was analysed using grounded theory methodology. Psychosocial effects on stress experience, stressors and resilience factors were identified. We generated three hypotheses. Hypothesis one is that moral distress and moral injury are main stressors experienced by healthcare workers. Hypothesis two states that organisational resilience plays an important part in how healthcare workers experience the crisis. Organisational justice and decentralized decision making are essential elements of staff wellbeing. Hypothesis three refers to effective psychosocial support: Basic on scene psychosocial support based on the Hobfoll principles given by trusted and well-known mental health professionals and peers in an integrated approach works best during the pandemic.

All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this this version posted November 18, 2020. ;https://doi.org/10.1101https://doi.org/10. /2020 NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.  In a global pandemic, a different set of rules has to be applied to healthcare delivery which can lead to significant distress [31]. The ethical bitterness of triage decisions is well-1 0 8 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this this version posted November 18, 2020. ;https://doi.org/10.1101https://doi.org/10. /2020 known and medical organisations and professionals have found different ways to deal with it 1 0 9 [36,37]. Some of the most common ethical challenges in the response to COVID-19 can be 1 1 0 conceptualized in triage, shortage of personal protective equipment and non-pharmaceutical 1 1 1 interventions. The following strategies can be identified to deal with the dilemmata [38][39][40]. found to allocate resources such as the criteria of medical need and efficacy of treatment. workers, a balance between duty to care and protection of health care workers has to be 1 1 9 found. Especially in a situation in which the availability of health care workers is crucial, they  Non-pharmaceutical interventions: All non-pharmaceutical interventions that are 1 2 2 critical (with regards to e.g. data protection), need to be appropriate with regard to necessity, proportionality and minimization (of e.g. data use). be extremely stressful [9,21,23,25]. Data showed, that altruistic risk-acceptance during the All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this this version posted November 18, 2020. ;https://doi.org/10.1101https://doi.org/10. /2020 negative impacts. Underlining the valuable altruistic attitude can reduce psychological 1 3 6 distress especially in those who are quarantined [10]. Altruism being a protective factor, 1 3 7 elements that reduce altruism and endanger one´s self view as a helper may do a lot of 1 3 8 damage for health care personnel. Receiving good guidance and not having to take decisions on their own as well as   Another stress-reducing factor is training and support in using PPE. The provision of 1 5 2 equipment, including masks and suits as well as infection control guidance by the hospital correlates of burnout and stress among nurses engaged in the SARS outbreak in Canada.

5 7
Raising awareness of the effects of disease prevention measures among staff with reduced  Furthermore, psychosocial interventions may be more helpful than support that is too 1 6 1 much focused on clinical psychological interventions [22]. Psychosocial services among 1 6 2 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.18.20232272 doi: medRxiv preprint 8 health care workers provided either by peers or by well-known and trusted mental health 1 6 3 professionals -such as psychologists, clergy or psychiatrists -have been shown to be 1 6 4 especially helpful [10]. Signalling in-group messages that all staff are in this together and 1 6 5 nobody has to carry decisions alone are considered as very stress relieving [9,10,21]. Opportunities for psychoeducation and psychosocial counselling are essential for the 1 6 7 protection of personnel. However, during the pandemic the provision of psychosocial support 1 6 8 has been reported to be most effective when including hospital management and leadership and focusing on pragmatic support based on the actual needs of the staff. This can be done  Pre-job-trainings on identification and responding to psychological problems in of psychological support might be higher when provided low-threshold access. Counsellors  In many cases, anonymous helplines and counselling have been established in order 1 7 9 to reach highly affected staff in the hospitals. However, considering the access to 1 8 0 psychological services, it should be taken into account, that staff with subthreshold and mild 1 8 1 levels of mental health disturbances might actually be more likely to take action and be 1 8 2 motivated to acquire skills to help others, than those with more severe disturbances who  If we look at the concept of resilience in the context of natural hazards we can identify 1 8 6 three core elements: resistance, recovery and adaptive capacity [41,43]. Resistance is the 1 8 7 ability of a system to withstand a threat, which is mostly a consequence of strength and preparedness. Recovery is the ability to bounce back and be able to come back to a state of 1 8 9 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.18.20232272 doi: medRxiv preprint 9 normal functioning after a disaster has hit and adaptive capacity refers to the ability to learn workers, evidence points in the way that resilience building on the organisational level is of 1 9 4 utmost relevance for pandemics. This goes way beyond the development of adequate 1 9 5 training and psychosocial interventions, although these are also very important to have. According to the authors, two constructs have been relevant for organizational 1 9 7 resilience in a pandemic: Organizational justice and "magnet hospitals". Organisational healthcare staff into account and take their concerns seriously. The first element of 2 0 0 organizational justice is relational justice, the ability of team leaders and managers to 2 0 1 suppress their own prejudice and treat their staff in an honest and just manner. The second 2 0 2 element is decisional justice, which mainly refers to fair decision making. Organisational  Another relevant construct for organisational resilience was found in so called "magnet 2 0 6 hospitals". These were characterized by decentralized decision-making and nurses being 2 0 7 amongst administrative staff and management, a flexible approach to shifts and continuous 2 0 8 effort in training and self-administration of units. These results confirm other findings from organizational psychology that have shown 2 1 0 that a high amount of demands mixed with low control (low decision making capacity and low 2 1 1 influence in staff) and an imbalance between effort and reward has negative effects on staff  All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.18.20232272 doi: medRxiv preprint 1 0 In this study, we are trying to identify the subjective experience, stressors and 2 1 5 collective coping strategies used in hospital care to relieve stress from a psychosocial 2 1 6 perspective and manage the challenging situation of the pandemic. The objective of the 2 1 7 study was to find out what the main stressors as well as the main stress reducing factors 2 1 8 were during the first phase of this pandemic for healthcare workers in the hospital sector. As 2 1 9 many of the psychosocial challenges during the pandemic have to be considered from an 2 2 0 ethical perspective, we were also interested in ethical questions in COVID-19. Amongst 2 2 1 others, these included resource distribution when there is shortage in supplies as well as the 2 2 2 duty to help while risking one's own health and health of relatives considered vulnerable. We 2 2 3 assumed to find indicators of moral injury in healthcare workers to be of importance during 2 2 4 this first phase of the pandemic. However, we were not only interested in the negative effects 2 2 5 of the pandemic on healthcare workers. We were especially interested in (organizational and 2 2 6 team) resilience, namely indicators for resistance, recovery and adaptive capacity of healthcare workers themselves as well as the systems and organisations they work in. The data presented in this article is part of a larger study done in the course of an EU Project (NO-FEAR) where we use a mixed methods approach alternating between 2 3 0 quantitative (online) surveys and qualitative interviews and focus groups. As a first step, we 2 3 1 chose to conduct expert interviews. The reason for this choice was the fact that we expected 2 3 2 an explorative approach to be the best way to gain more detailed insights in both, the 2 3 3 positive and negative subjective experiences of healthcare workers during this first and very 2 3 4 stressful phase of the disaster. As we did our study very early in the crisis we knew that 2 3 5 healthcare workers were still in the middle of a very stressful job and expected them not to 2 3 6 be willing to do long interviews or fill out questionnaires. We thus focused on a small number 2 3 7 of experts working in the field including mental health professionals as well as experts from contrast and thus tried to get interviews from rather different contexts and backgrounds. Our All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.18.20232272 doi: medRxiv preprint 1 1 field and in a leading position or a position that allowed to speak for their colleague 2 4 2 healthcare workers from an expert position.

4 3
Our aim was to develop some basic hypotheses regarding resilience and stress 2 4 4 experience of healthcare workers in the first phase of the COVID-19 pandemic. These 2 4 5 hypotheses shall be a basis for the interpretation of data from questionnaires and interviews 2 4 6 gained in later phases of the pandemic. We are well aware that these data are far from 2 4 7 generalizable, but expect our hypotheses to be a good basis for further research as well as  In total, 13 experts were interviewed by two teams. Participation of experts was 2 5 0 requested via telephone calls. Eligible experts included in the study sample had to be health interviewed were informed about the scope of the interview and their rights. They subsequently signed an informed consent that was previously reviewed by the NO-FEAR Project External Ethics Advisory Board. The Austrian team consisting of male and female scientifically experienced psychologists conducted two semi-structured interviews with mental health experts from  Additionally, we focused on strategies that health care workers experienced as helpful or 2 6 7 stress relieving to identify possibilities for psychosocial support measures. All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.18.20232272 doi: medRxiv preprint 1 2 One additional interview was conducted on challenges from the ethical and legal   Table 1 is giving an overview on basic demographic data of the study sample.  The interviews were transcribed and eventually analysed using Grounded Theory 2 7 7 methodology. Data was discussed with participants to allow for comments and correction.

7 8
We coded data into several concepts and subcategories to identify psychosocial and ethical  All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.18.20232272 doi: medRxiv preprint  Emotionally, fear of getting infected and subsequently infecting families and friends 2 9 2 are common feelings. Experts point out the prevalence of feelings of guilt and shame as 2 9 3 employees feel like being "plague spreaders". This resulted in many health care workers not returning home as they did not want to loss of trust in the system, as they do not feel sufficiently protected. appropriate compensation is central. Among doctors, more silent reactions seem common 3 0 5 which has been described by one interview partner as a "paralysis of thought". preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.18.20232272 doi: medRxiv preprint 1 4 One other interview partner points out "that fear of infection among health care 3 1 1 workers is higher after the shifts when resting at home" (Interview partner 1, Italy).

1 2
Socially, stress reactions included increased lack of trust in external or newly hired 3 1 3 workers, as more experienced staff are afraid that unexperienced workers might make 3 1 4 mistakes. But also a loss of trust in the hospital and in oneself as well as a need for a caring 3 1 5 leadership was observed. gratifications. One aspect that is frequently mentioned is the "sharing of moments", be they to the patient and the relatives. The relationship the personnel is looking for is more similar to 3 2 3 a child-parent scheme than subordinate-superior scheme." (Interview partner 3, Italy). Physically, exhaustion and fatigue are prevailing. This is also true for staff in the 3 2 6 laboratories due to high amounts of positive blood test results. Psychologists report about 3 2 7 insomnia as a common issue of health care workers according to the experts that were 3 2 8 interviewed.

2 9
Cognitive stress reactions include confusion and unrest as well as high levels of  All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.18.20232272 doi: medRxiv preprint 1 5 by many experts. This also led to a feeling of "us" versus "them" meaning "us frontline 3 3 7 workers" versus "them in the background". The COVID-19 pandemic is a rapidly evolving situation in which the protection of the the rapidity with which patients tested positive came into hospitals and/or the rapidity of when the first wave or pandemic as a whole will end, increased stress for healthcare Healthcare workers often had to work in newly formed teams that consisted of staff that new roles had to be carried out, many of which health care workers did not feel were another common stressor. Normal social exchange during breaks as well as common 3 6 2 routines could not be used during the shifts. Stress is also grounded in the unavailability of 3 6 3 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.18.20232272 doi: medRxiv preprint material for personal protection, which aggravated in further imbalance between demand and 3 8 0 supply of equipment. In other cases, lack of resources has not been a major issue. However, 3 8 1 even then the anticipated lack of resources created a lot of anxiety and stress. Triage  where trust in the system as well as the self-image as a good person were endangered. All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.18.20232272 doi: medRxiv preprint 1 7 The situation was enhanced by a loss of professional distance: Many workers were 3 9 0 mediating between families and patients, as relatives were not allowed to visit. These tasks 3 9 1 differed significantly from the daily routines that staff were used to and forced the healthcare 3 9 2 workers into an unusual intimacy towards patients and relatives that endangered their 3 9 3 professional distance. Many healthcare workers were providing psychosocial support to 3 9 4 patients who were dying alone in the hospitals. In these cases, part of their professional 3 9 5 distance got lost, as they were not able to use the protective strategies they normally use. Due to protective measures, the relationship to patients was experienced differently. Healthcare workers experienced PPE as limiting communication and trust building especially 3 9 8 in relationships with "strangers".  For many health care workers appropriate risk assessment was difficult, as staff was All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.18.20232272 doi: medRxiv preprint 1 8 with them at home. As a reaction, many healthcare workers did not go home to sleep and  Nevertheless, "external" support by leadership, peers and mental health professionals  to more stable routines and therefore was experienced as stress relieving. There was a need 4 3 7 for regularly updated guidelines for protection of health care workers as well as accurate communication worked and the more the messages were experienced as clear, the better 4 4 0 they were received by the personnel. One expert pointed out, that 4 4 1 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.18.20232272 doi: medRxiv preprint 1 9 "written information wasn't read and thus did not reach all health care workers, while 4 4 2 recorded videos by the management were much more efficient with hundreds of employees 4 4 3 watching (Interview partner 5, Austria).

4
Another strategy that was well received was holding regular meetings of team leaders  Enabling a dialogue between management and staff was very much appreciated. In 4 5 0 some hospitals, feedback mechanisms were established. For example, one person from the 4 5 1 management was made responsible for "listening to staff" and to take care that reported 4 5 2 needs and problems were addressed in a timely manner. These were often just "little things", and provide a sense of safety and appreciation. guidance and training on how to use equipment has been reported as one of the most 4 6 0 important strategies in order to relieve stress for health care workers. In some hospitals, on how to relate to severely ill patients and relatives were experienced as helpful. Additionally, possibilities to exchange among peers on clinical cases and decision making as well as sharing of special moments has been pointed out as helpful to promote peer support 4 6 7 and relief stress among staff. This often asked for an adaptation of areas where people can 4 6 8 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this this version posted November 18, 2020. ;https://doi.org/10.1101/2020 doi: medRxiv preprint 2 0 meet and exchange while keeping the necessary distance, like outside areas or larger 4 6 9 rooms.

7 0
Guidance and joint decision making regarding triage has reduced a lot of stress. One 4 7 1 example is a pro forma "validation" of each patient before a shortage appeared in a medical decision had a basis from which to start and did not feel completely left alone. Mental health support was described as most successful when medical leadership was directly requested from the healthcare workers mainly for patients and relatives. psychiatrists or psychologists, was very well received and reduced stress in healthcare staff. Group interventions for staff were often not possible as staff was too overworked to attend.

8 6
However, free and anonymous service for staff is essential in providing low threshold support in quarantine has also been reported as well received but only when done by someone  has been reported to be interventions that are more successful than classical clinical 4 9 3 approaches. For example meeting the need for social gatherings (e.g. meeting once a week 4 9 4 on the balcony or during trainings in a big room) as well as support in providing small islands 4 9 5 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

9 6
Another example was setting up a piano in the ward for certain hours in order to confront 4 9 7 silence.

9 8
Psychologists and/or clergy visiting the wards in an outreaching manner was 4 9 9 described as much more effective as remote forms of support. Direct contact by mental 5 0 0 health professionals and management has been highly appreciated by staff. One expert 5 0 1 even spoke of a "knight blow" for the psychosocial staff and several experts pointed out that 5 0 2 medical staff did not take it well if mental health professionals stayed in the background only. The output of this article is directly related to the first one and a half months of the probably lead to different conclusions that will be drawn during the "lesson learning process", 5 0 7 that will take place at different levels in all the healthcare domain. However, this output has to be viewed as a gathering of first level immediate 5 0 9 impressions and experiences from healthcare staff involved in this massive healthcare crisis.

1 0
Like a ship's logbook it contains the day to day evolution and perception of the healthcare 5 1 1 personnel through the lens of mental health and medical experts who themselves were 5 1 2 deeply involved into the mission of "fighting corona".

1 3
In knowledge management, lesson learning is a process that is usually done after the 5 1 4 crisis with a debriefing mechanism and having players of multiple roles sitting at the table to 5 1 5 draw common conclusions. Lesson learning is usually defective in the part that deals with 5 1 6 retrieving and analysing first-hand experiences. We do not want this to happen for the  The importance of this work can be summarized in the following three points: All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.18.20232272 doi: medRxiv preprint 2 2 2) Steering the steps of a future lesson learning process 5 2 1 3) Precociously identifying issues that due to the long timeline of the COVID-19 crisis 5 2 2 might be of help to countries whose response will be postponed in time as well as to 5 2 3 all those who are faced with further waves of the pandemic.

2 4
One of the main hypotheses that we derived from our data is that moral injury seems to 5 2 5 be a common and central outcome of this crisis in many healthcare workers, be it in areas where the virus has had an extreme negative impact on the healthcare system or in areas where the pandemic had not so much negative impact (yet). Even though extreme measures and family have been experiences shared by many healthcare workers during this pandemic.

3 3
Our second main hypothesis is based on the findings that organizational justice as well  view, the challenge that we are facing now after the first phase of the pandemic is to rebuild 5 3 9 trust in ourselves as helpers as well as in the systems that we are working in. As Maunder et during this crisis as highly centralized organisations with an administration that did not All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

4 8
Our third hypothesis is that psychosocial support of healthcare workers during this 5 4 9 pandemic only works if it is given by trusted mental health professionals or peers in a very 5 5 0 basic manner integrated into the overall support by the management (trainings, information, 5 5 1 frameworks provided). The best support given during the first phase of a disaster response is 5 5 2 on scene support guided by the motto 'to act with the people and not for the people' [47].

3
Thus, mental health professionals who acted with the healthcare personnel and not for them (from an external safe position) and who acknowledged and appreciated the resilience that 5 5 5 this group already has and did not treat them as potential patients were the only ones whose 5 5 6 support was accepted.

7
Williams et al. [33] recommend several strategies to deal with moral distress and moral   According to our data, safety refers to good, honest and timely information given most directly by trusted leaders. Sufficient PPE but also visible efforts to support and protect staff 5 6 6 by the management as well as interest in how they manage the situation leads to a feeling of 5 6 7 safety in a situation where many healthcare workers had the feeling that there is no safe than those in other units where no one knew when a patient could be tested positive. This was especially true during the very first phases when PPE was not available for everybody.

7 1
Connectedness refers to efforts of staff themselves and their leadership as well as mental 5 7 2 health professionals and peers to enhance group cohesion and allow for social exchange in 5 7 3 All rights reserved. No reuse allowed without permission. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this this version posted November 18, 2020. ; https://doi.org/10.1101/2020.11.18.20232272 doi: medRxiv preprint 2 4 spite of difficult circumstances. Examples were meetings on the balcony after the shift for a 5 7 4 quick exchange. Connectedness also was expressed by proactive contact to those who had 5 7 5 to stay at home because of infection or quarantine. Calm was reached by providing enough 5 7 6 space for rest and recovery (e.g. accommodation when staff does not want to go home).

7 7
Calm was also reached by re-establishing normalcy including common rituals during holidays 5 7 8 or other efforts to establish a normal working environment. Self and collective efficacy has 5 7 9 been reached by preparing staff for their new tasks, taking joint decisions for example by or lost their lives due to the pandemic. Faced with extremely distressing ethical dilemmas,