Two assessors independently evaluated the included articles using the AGREE II instrument. As stated in the instructions of the AGREE II instrument (5), specific items may not be applicable to particular guidelines. We had to adjust this instrument (which focuses on clinical guidelines) to the given context and thus excluded several items. Items 11, 16, and 21 were excluded due to their inapplicability to non-clinical guidelines and recommendation papers. Items 10, 13, 14, 17, 20, 22, and 23 were excluded as the in this assessment included articles do not meet the methodological or formal requirements needed in order to apply these items. Each item was rated on a 7-point scale (1- strongly disagree to 7- strongly agree). Scaled domain scores were calculated as percentages of the maximum possible scores, according to the AGREE II methodology, using the following formula: (obtained score-minimum possible score) / (maximum possible score–minimum possible score), where the “obtained score” is the sum of the appraisers scores for each domain [see paragraph IV. Scoring the AGREE II (5)]. Hence, the discrepancies between the two assessors were considered during the process of evaluation. As reported in other studies using the AGREE II instrument, we considered a value >60% using the modified scale as a sufficient quality score and a value >80% as a good quality score (6, 7).

To provide an overview on the types of recommendations given, two assessors extracted the recommendations and grouped them into different categories and within each category, into different topics. During a mutual process, key topics were derived from the given recommendations and recommendations were assigned to their respective key topic. However, recommendations that were mentioned in <5 out of the included 41 articles and could not be assigned to any of the existing key topics were excluded from the presented overview. The type of categories and topics are reported, together with the total number of articles including each recommendation.

Table 1 provides a detailed overview of the outcomes per article.

Specific recommendations have been identified and were clustered into four different categories. These categories and their respective key topics are displayed in Figure 2. A detailed overview on which key topic was included in which article can be found in Table 2. Out of the included articles, physicians were mentioned as the target occupational group in 11 articles, while nurses were mentioned in 12 articles. Management employees were mentioned in only one article, as well as specialist interns and patients or family members of patients. Allied health professionals, such as midwives or paramedics, were among the target occupational groups in 5 articles. Unfortunately, the majority of the articles (21 out of 41) did not further specify the term HCWs.

In most of the selected publications, authors emphasized the need for promoting better self-care of HCWs during this pandemic. Indeed, a basic component is meeting physical daily needs, such as supplementation of healthy nutrition and hydration, fitness, rest, and sleep. In the current setting (e.g., shortage of staff and time), these self-care aspects might often fall short for HCWs. Amongst the strategies for self-care practices on the individual level, diaphragmatic breathing (26, 28, 29, 32, 43), meditation (42, 43), maintaining a positive mind set (16, 29), mindfulness-, relaxation-, and problem-solving training (16, 26, 29, 32, 36, 39, 43, 45, 47) as well as maintaining personal interests, activities, and the connection to loved ones (16, 28, 32, 36, 43, 45–47) were mentioned frequently. Interestingly, avoiding maladaptive coping strategies (e.g., excessive alcohol consumption, overeating and prescription drugs) were only mentioned in two articles (46, 47). Interventions to encourage self-care on the organizational level included well-being courses, yoga or gym classes (12, 22, 23, 40) and providing opportunities for staff to talk about their experiences to enhance support and team cohesion (9, 14, 19, 43, 46). As reviewed, these recommendations on self-care are quite diverse and affect physical, psychological and social well-being of HCWs. It is possible that employees may not take up or use these offers due to lack of time or motivation after a long work shift. Strategies to overcome these potential barriers were not discussed in most articles. Finally, one must note that mental conditions like anxiety or depression itself of a certain degree can also result in barriers for the affected individual to promote self-care.

Recommendations addressing access to mental health services consist of (1) early identification of “at-risk” individuals (due to pre-existing experiences or mental health issues) so that plans can be put in place to support them, (2) actively monitoring anyone who has been exposed to a potentially traumatic event, (3) available access for staff members to psychologic or psychiatric support (e.g., helplines, online self-help programs, trauma focused PTSD treatment) (8–10, 12, 14–16, 19–24, 26, 31, 33, 35–38, 40–48). Providing psychological care and monitoring staff who are at higher risk of developing a mental disorder after the pandemic begins to recede were recommended in only five articles (14, 18, 24, 25, 37). Most guidelines seem to neglect the potentially ongoing stress and burden HCWs might face after pandemic (e.g., postponed surgeries and treatments, structural changes in the healthcare systems, staff shortage). Barriers and limiting factors for the use of mental health services, such as lack of anonymity or accessibility, were scarcely addressed in the included articles.

Interventions to improve mental well-being through social and structural support were also mentioned across many articles, which overlap to some extent with the strategies and topics mentioned above. Authors stressed the pivotal role of an appropriate appreciation, acknowledgment, and professional validation within the team and in particular as an integral part of the leadership style (12, 16, 19, 24–26, 30, 40, 43, 47, 48). Individual strategies include basic rules for respectful interaction, such as “giving thanks” (29). However, the majority of authors remained vague about specific strategies to actively show and promote appreciation, acknowledgment and validation. Organizationally, leaders are required to listen, learn and act (34). Not only in times of crisis is an ethical, inclusive and effective leadership required (e.g., leading by example, providing personal and professional support, involving staff in decision-making and action plans, establishing a human connection by validating an individual's feelings and thoughts) (8, 20, 34). Other strategies, such as providing free food and drinks, to show appreciation might not be sufficient. Authors highlight the role of support, both at the individual (e.g., family, friends, communities) and at the organizational level (peer support programs, online support, support from leaders). In respect to the recommendations concerning social support, it is noted that this aspect greatly depends on the support system and the resources of each individual. On that basis, the University of Minnesota Medical Center proposed an approach taken from the military framework. They developed a psychological resilience intervention founded on a peer support model (Battle Buddies) with 2 key elements: A Battle Buddy to provide peer support and a mental health consultant assigned to the unit (9).

In times of crisis, it is important to provide high-quality and transparent communication and accurate information updates to all staff (14). Existing research shows that uncertainty leads to stress and anxiety (49). Stress increases with high work demands but co-occuring low work control (43). Therefore, leaders should provide staff with transparent and current updates so they are best prepared for what they are going to face and reflect on the risks and challenges (14, 43). That is especially important at the beginning of a crisis. Surprisingly, avoiding unreliable news sources and social media is recommended only by eight articles (9, 28, 32, 36, 39, 41, 43, 46). That raises the question whether the influence of social media and news is underestimated in this context. Authors further emphasize the need of listening to staff input and feedback (9, 11, 13, 14, 16, 19, 20, 28, 31, 33, 34, 36, 37, 40, 43, 44). Leaders should provide the opportunity to talk to them and implement regular feedback mechanisms. For implementation, it requires practical strategies. Once again, the shortage of time, exhaustion and staff shortage might be limiting factors.

Authors often mentioned the need of adequate organizational support through the implementation of a safe and employee-oriented work environment. This includes the provision of complete and quality Personal Protective Equipment (PPE) and supplies to prevent infection, provision of accurate and timely information regarding the disease, employing best-practice protocols and guidelines, and implementation of infection control trainings (10, 12–14, 18, 21, 22, 31, 36, 38–41, 46, 47). Moreover, a few authors took into consideration that an employee-oriented work environment should promote professional development and provide specialized job training (16, 18, 22, 26, 35). Where possible, work environment should be optimized to support appropriate nutrition, rest (e.g., “take a minute” room) and sleep periods. While mandatory training and supervision programs (on the clinical skills required to deal with COVID-19 as well as on the potentially traumatic situations) might be beneficial for the team, some team members may have negative feelings and doubts toward mandatory interventions. Individual attitudes, preferences and sentiments might therefore have adverse effects on the team and counteract these interventions. This is not addressed in most articles. Another overlooked, yet crucial factor, is ensuring an adequate income as well as appropriate working hours for all occupational groups working in the healthcare system. While recommendations concerning these aspects might appear too obvious to be mentioned, it is necessary to stress their importance and potential consequences.

Overall, recommendations on how to improve mental health in HCWs during the COVID-19 pandemic were targeted at various levels: from societal aspects to senior management and healthcare professionals. The relevance of protecting and promoting HCWs mental well-being must be viewed as a worldwide problem, as studies show the negative effects pandemics and epidemics have on the mental health of HCWs in several countries across Asia, Northern America, Middle East, Europe and West Africa (1). Additionally, a recent review showed an increase in the turnover intention in nurses in post-pandemic studies, posing the risk of further aggravating staff shortage (50). These circumstances can lead to a vicious cycle, putting more pressure on those remaining in their professions. Only few of the reviewed publications included intervention programs specifically designed to enhance mental health care for HCWs to face psychological challenges during the pandemic. Interestingly, most articles only focus on the time of the acute crisis and neglect what might happen when the crisis is over. However, we must be aware that the COVID-19 pandemic is rather a marathon, not a sprint (48). Against this background, we were surprised not to find a relevant number of recommendations regarding the prevention of mental health burden after the end of the pandemic. In that regard, only few of the reviewed publication described strategies and interventions to support HCWs' mental health after the crisis. Another pivotal, but so far neglected factor might be an adequate income for HCWs. Without ensuring a fair income, there might be little incentive to choose the profession and this in turn might cause staff shortage in the future. The potential consequences include e.g., higher workloads and more working hours for HCWs in the future, which might affect the work-life balance and cause even more physical and mental strain. Breaking this vicious circle is long overdue and should be addressed by political decision-makers. Government should provide healthcare organizations with sufficient resources (and appropriate income) to implement recommendations that fit their needs and adapt them to their context.

This review should be considered in light of some limitations. First, available publications for this review varied greatly regarding the publication type, which makes direct comparisons difficult and prevented us from conducting a formal systematic analysis. As reported by some of the authors, few of the recommendations and suggestions have substantial empirical evidence to support them. Second, our initial search may have neglected certain terms, however, re-inspecting our search by adding other potentially relevant key terms provided no additional articles suitable for our assessment. Further, we initially limited the search to the database Pubmed. Adding the databases Cochrane Library and EMBASE during the process of revision yielded in no additional publications. Moreover, we did not pre-specify our reported outcome categories and did not register this work in PROSPERO, as scoping reviews are not accepted for registration by PROSPERO. However, we instead registered our review on OSF Registries. The reason for our inductive approach was that no previous work was available, and we aimed at providing the very first overview assessing the quality of the guidelines. Furthermore, checking all included articles for the aspect of telemedicine showed that, surprisingly, only five articles mentioned telemedicine. This potential limited availability during a pandemic that may be derived from the underreporting in the selected publications, can be challenging for HCWs who are quarantined or who live in areas with limited access to mental-healthcare services. Next, the quality of the analyzed publication was low compared to standard medical guidelines. Especially the low scores for stakeholder involvement and rigor of development are a relevant source of bias.

Nevertheless, the urgency to develop specific psychological support intervention protocols for HCWs is apparent, not only in times of crisis. We believe that there are lots of measures that organizations, individuals, and national societies can take to minimize the impact of COVID-19 on the mental health of HCWs. However, it is essential to implementation some of the described measures to allow for early-detection and early-intervention in HCWs facing tremendous stress and burden. At this stage, one must conclude that despite a huge amount of available publications, evidence- and consensus-based guidelines on how to detect, prevent and treat psychiatric conditions in HCWs are lacking. The reasons for this gap were described in the previous paragraphs. While possible treatment options for HCWs with mental conditions during the pandemic can be extrapolated from available high-quality guidelines on how to treat e.g., anxiety, depression, trauma or alcohol dependency, more research is needed regarding the earlier detection and prevention in the vulnerable group of HCWs. Finally, in a pandemic, the perspective of different healthcare systems and cultures must receive far more attention.