This study aimed to examine the experience and impact of PMIEs in a nationally representative sample of NHS healthcare staff during the COVID-19 pandemic. Four key findings were observed. First, nearly a third of healthcare staff reported experiences of PMIEs, with acts of betrayal most frequently reported. Second, PMIE exposure was significantly associated with adverse mental health symptoms across clinical and non-clinical healthcare staff. Third, specific work-factors were significantly associated with expressions of moral injury, including being redeployed, a lack of PPE, and experiencing a lack of support. Finally, in all staff groups, those reporting symptoms of mental disorders were significantly more likely to experience PMIEs, with those in certain roles more likely to endorse specific PMIEs. Nurses reporting symptoms of mental disorders were significantly more likely to report all types of PMIEs than those without symptoms. Whereas doctors who reported mental disorder symptoms were only more likely to report betrayal events (not omission or commission). In addition to reporting betrayal events, other clinical and non-clinical staff reporting symptoms of mental health disorders were more likely than those without symptoms to report acts of omission.

That a substantial proportion (68%; 95% CI 67, 69) of the present sample met case criteria for one or more probable mental disorders, including PTSD, CMD and alcohol misuse, is consistent with previous studies [16,18,31], highlighting the current state of wellbeing in healthcare staff providing care to patients during the COVID-19 Pandemic. This research is novel in that it extends previous findings on moral injury to the UK healthcare workforce during the time of the COVID-19 Pandemic. Previous studies have found relationships between moral injury and mental disorders in other occupational groups, particularly military samples [4,5,32]. The significant relationship between experiences of moral injury and adverse mental health in UK healthcare staff found in the present study adds to the literature by illustrating that moral injury can be experienced by staff in non-military settings, and also have negative associations with mental ill-health. However, due to the cross-sectional design of the current study, we cannot infer causality. Nonetheless, the present results highlight that moral injury is associated with poor wellbeing in both clinical and non-clinical staff – particularly following acts of omission and betrayal. This highlights that moral injury is not solely applicable to clinical staff and does occur to staff who may not directly provide patient care. As safeguarding the wellbeing of staff is a priority for the NHS, these findings suggest that all staff, including those who do not work clinically, may experience moral injury and this should be considered when making provisions for staff wellbeing support.

A number of occupational factors were found to be associated with moral injury, including a lack of support from management and colleagues, perceived lack of PPE and having a colleague die of COVID-19. These findings are broadly consistent with a recent qualitative study which observed that military veterans may be more vulnerable to moral injury if they felt had a lack of support from chain of command [3]. This highlights that healthcare staff may benefit from targeted support from those in management roles, with managers prepared to have psychologically sensitive conversations with staff, offering informal support and signposting to services where necessary [10]. A lack of adequate PPE to protect staff was a problem for the NHS at the start of the pandemic in March 2020 [33]. Frankly preparing staff for the difficult tasks they will be asked to carry out in less than ideal and perhaps unsafe circumstances may potentially be protective and normalise distress [11]. Transparent discussion and acknowledgement when staff/organisational failings have occurred, as well as when staff have been placed in impossible situations, may also be experienced as validating and helpful. Those staff who re-deploy may also be most likely to benefit from (in)formal support and targeted advice. For example, staff who re-deploy should be adequately prepared for their role or, where time is lacking, be closely mentored as they begin their role. Staff should also be checked for redeployment suitability – for instance, if they are in a vulnerable group [18]– and be provided with information about the role and the risks associated so they are able to make an informed choice about re-deployment. Ideally, having redployment be explicitly voluntary may be beneficial. Where a colleague has died, it may be beneficial for managers to ensure that this is recognised and acknowledged as a impactful event and ensure appropriate support is in place. Previous studies have found that feeling satisfied by the notification of death process, where sufficient information was given in order to make sense of what happened, is related to less long-term mental distress [34]. Nonetheless, future qualitative studies to provide an in-depth understanding of the impact of PMIEs on healthcare staff wellbeing and their perceived support needs would be worthwhile in informing this process. Equally, during adversity it is possible to experience -traumatic growth [35] and this should also be examined in future healthcare staff wellbeing research.

Those with mental health problems across the healthcare sample were significantly more likely to report betrayal events. Additional research is needed to better understand how staff perceive the betrayal events to have occurred and by whom (e.g. colleague, the public, the government) to inform targeted interventions for repair. Where a betrayal event has occurred ‘in house’, reflective practice sessions, such as Schwarz rounds [36], that allow views to be expressed and solutions to be co-produced may be effective to mitigate the development of moral injury. If a betrayal is perceived to be caused by someone outside the health service, then a governmental or community approach may be more appropriate. As this study examined moral injury using the MIES, which includes items which could be interpreted in more than one way (e.g. ‘I feel betrayed by others outside the health service who I once trusted’), responses may not always reflect moral injury as it relates to staff’s occupational role. Nonetheless, recommendations of measures to manage moral injury arising outside the NHS (e.g. public enquiries) are beyond the scope of this paper

Nurses with mental disorders were more likely to report all types of PMIEs and may be more vulnerable to a broader range of PMIEs compared to other staff. It is possible that nurses may be particularly vulnerable to moral injury as they have considerable responsibilities and demands on their time, yet they also may have less ‘control’ over the care they are asked to provide despite often spending considerable time with patients and their relatives. In the general population, those most likely to experience mental disorders related to COVID-19 are often younger, female and those with children [37,38] - criteria which many nurses fit. However, given the cross-sectional nature of the study, it is unclear whether exposure to PMIEs causes mental health problems, or whether having mental health problems makes one more likely to experience PMIEs. No prospective studies of moral injury have been conducted to date and this leaves a considerable gap in our understanding of the development and impact of moral injury over time.

This study has several strengths and limitations. Amongst the strengths was the inclusion of a large and representative, weighted NHS sample. A second strength was that moral injury was examined comprehensively, exploring not only the impact of exposure on wellbeing but also at the factors associated with exposure. However, a limitation of this study is that there is no existing UK validated measure for moral injury. We used the MIES because it has been used internationally [39,40], yet as some of the MIES items include distress related to moral injury (e.g. ‘I am troubled by…’) it is possible this may have conflated effects [41]. Secondly, the MIES was only completed by participants who took part in the longer survey [18]. Evidence from the longer survey, reported in Lamb et al. [18] found women were statistically significantly more likely to complete both surveys than men, as were participants of White ethnicity, those in a relationship, those who were nurses or in non-clinical roles, and those who had not been redeployed. As nursing staff and those who were re-deployed may be at particular risk for moral injury, this may have impacted our findings. Third, the presence of mental health difficulties was evaluated using self-report measures rather than clinical interviews, which is the gold standard for mental illness assessment. Finally, the cross-sectional nature of this study cannot be overlooked, and the direction of effects is unclear.