Nursing staff and supervisors perceptions on stress and resilience: a qualitative study

This study aimed to assess the predisposing risk and resilience factors of the stress of nursing staff as well as to explore the common and distinctive perceptions of these factors between nurses without a managerial position (nursing staff) and employees in a supervising position (nurse managers, ward nurses).

The study reports parts of the needs assessment within the subproject “Healthy Ageing in the Nursing Profession” (“Gesund Altern im Pflegeberuf”). This subproject was one of five subprojects of the project “Mental Health in the Hospital Workplace” (“Seelische Gesundheit am Arbeitsplatz Krankenhaus, SEEGEN”) [26] funded by the General Ministry of Research and Education.

In the needs assessment of “Healthy Ageing in Nursing Profession”, participants were asked about several topics, age stereotypes (already published in Helaß et al. [27]) and stress/resilience factors, which were surveyed independently of each other and evaluated separately, so that no overlaps or thematic mixing could occur. This study followed a generic qualitative approach [28], as the participants were not only asked about their own views, but also the supervisors views of (other) nursing staff. Interviews and focus groups were conducted at a university hospital of maximum medical care (surgical, conservative and mental health departments) in Germany.

The participants were registered nurses, ward nurses and nurse managers of a German university hospital.¹ All registered nurses and supervisors were addressed to take part in the study. For the study, ward nurses and nurse managers were grouped together as “supervisors”. Of 3090 contacted nurses and supervisors, n = 50 participants from nine departments took part in the study (response rate = 1.62%), as described in Helaß [27]. Table 1 presents the sociodemographic data of the participants.

Following theoretical sampling [29], the selection of the participants was initially open but was adjusted in the process of data evaluation by specifically selecting based on sociodemographic factors [30] to achieve an almost equal distribution in terms of gender and hierarchical level and the most heterogeneous distribution possible in terms of age, professional experience and hospital affiliation [31].

Either interviews or focus groups were conducted with the 50 participating nurses. The focus groups were conducted as a supplement to the 19 face-to-face interviews in order to work out common points of view at hierarchical level and to clarify disagreements. To verify this, an attempt was made to conduct two focus groups per hierarchy level. 31 nursing staff were assigned to the focus groups according to their hierarchical level, so that 5 focus groups were held (two focus groups of registered nurses, two of ward nurses, and one of nurse managers). One participant withdrew his written consent.

A literature-based semistructured interview guide of 16 questions on work situation and professional biography was constructed according to the model of Lazarus` dichotomy of risk factors and resources [32]. It was used for face-to-face-interviews and focus groups (see Appendix A), which were conducted in the German language between August and November 2018 in the offices of the research team or nurse managers. The mean duration of the interviews was 32 min (range, 26–42 min) and of focus groups 33 min (range, 30–38 min). One moderator (MH) led the hierarchical-level–grouped focus groups. The participants were encouraged to comment and directly asked questions towards each other [33]. In addition, the mention of physical and psychological complaints was quantified.

Written informed consent was obtained. The Ethics Committee of Heidelberg University reviewed and approved the study (S-005/2018). The study was registered with the German Register for Clinical Studies (DRKS 00013482) on 09 March 2018.

The face-to-face interviews and focus groups were transcribed and analysed by two researchers (MH and IM) according to Mayring’s qualitative content analysis [34] using MAXQDA Analytics Pro 2018 [35]. The analysis was conducted inductively in order to find the most important attitudes of the participants and to identify core variables. The preliminary category system that emerged from the analysis of 10 transcripts was further refined and verified by analysing the other interviews. The completed category system was checked again on the first 10 transcripts and the semantic units were re-categorised where necessary. This process was continuously discussed by the research team until consensus and data saturation [36] were reached. The similarities and differences of the reports of supervisors (SU, ward nurses and nursing managers) and registered nurses (RN, nurses without managerial position) are described for each category.

The current research standards (recommendations of Lincoln and Guba (1986) and the Consolidated Criteria for Reported Qualitative Research (COREQ)) were adhered to and followed in terms of credibility, reliability, confirmability, authenticity and transferability. Extensive notes and self-reflections on experiences and biases toward the research subject discussed in the research team [37] were intended to enhance credibility.

In the qualitative analysis, 594 single codes were identified, from which three main categories and nine subcategories were formed. Category 1 summarises job demands as intrapersonal risk factors and predispositions of stress and health complaints in the nursing profession (C1, 93 codes). External job demands are described in category 2 (C2, 284 codes) on five subcategories (high job demand, lack of autonomy, economical and relational factors, lack of gratification, and lack of social support). Category 3 includes job resiliencies, strategies and resources on four subcategories (C3, 217 codes): behavioural, cognitive, social, and relational. Table 2 provides the category system, including definitions for each main category with (an) exemplary citation(s). Figure 1 shows the job demands and resources factors embedded in the transactional stress model [32]. This model assumes cognitive perspective on the development of stress reactions—not the objective situation per se (i.e., work demands and expectations), but rather the appraisal of the situation based on available resources and coping strategies seems to be the determining factor for stress development. In the case of a confrontation with an acute stressor or a punctual demand, a primary evaluation of the stressor occurs on the basis of available resources and strategies: existing risk factors are weighed against the available resources. If there is a predominance of resources and resilience, the punctual demand can be regarded as a task. If risk and resilience balance each other out, the requirement is seen as a challenge. However, if risk factors outweigh resources and strategies, a stress reaction occurs and active coping becomes necessary. A successfully mastered challenge has an effect on both internal risk factors and resources, in that the former are reduced due to the positive learning experience and the latter are strengthened. Research on Lazarus’s stress model and the nursing profession supports this thesis: stress depends on employees’ individual coping strategies based on qualification, social competence, and subjectively estimated control possibilities with regard to the critical situation and the experienced support [38].

We divided stress factors into internal and external factors (including high job demand, lack of autonomy, gratification, social support), economical and relational factors. The section of internal–cognitive factors showed great agreement between registered nurses and supervisors, especially in overcommitment, external locus of control, low self-efficacy and unplanned behaviour. Both nurses and supervisors evaluated the overcommitment of nurses, that is, hard work and long hours to meet or even exceed the work requirements, as a risk factor for work-related stress. Previous studies in the literature have shown that overcommitment and overidentification have an (indirect) effect on stress, burnout [39], anxiety [40], and depression in nurses [41]. However, in contrast to supervisors, nurses highlighted overcommitment as an internalisation of society’s image of the “serving nurse” and a consequence of high work demands, applied by supervisors (resp. by institution).

In addition to overcommitment, participants considered external locus of control, low self-efficacy, and unplanned behaviour as stress factors. These findings are in line with empirical studies confirming the correlation between locus of control, self-efficacy and burnout [42].

With regard to external stressors, next to high job demands and staff shortage, both groups consistently reported a low esprit de corps within the team and supervisors and low interprofessional cooperation. They suggested that nursing staff often experienced either uncertainty or multiplicity of demands, goals or expectations directed at them because of the flattening of hospital hierarchies. These descriptions can be understood as role ambiguity [43], which is closely related to role conflicts [44]. Role conflicts in turn contribute to high emotional exhaustion [45], reduced well-being [46], job dissatisfaction, low commitment, and high intention to leave the profession [47].

Some nurses pointed out that internal and external stressors influence each other. They reported internal factors as mediators of external stress factors, such as low autonomy, participation and gratification. This is in line with the findings of previous studies, in which professional autonomy was identified as an important predictor of job satisfaction [48], work performance and organisational commitment in nurses [49]. An imbalance between efforts and rewards as a reflection of insufficient gratification was positively related to emotional exhaustion [50].

Overall, nurses and supervisors largely agreed in their assessment of internal and external stress factors. Thus, it can be assumed that the stress of nurses is adequately communicated to supervisors and that supervisors are aware of the stress situation of their followers. This understanding for the employees could have a positive influence on SSR and the social work environment and may buffer nurses’ psychological distress and increase their well-being [21].

RNs and SUs agreed to a large extent about nurses’ behavioural, cognitive, social and relational resiliencies and coping strategies, with highest congruency in behavioural aspects and lowest in social support. With regard to behavioural resiliencies, RNs and SUs in our study reported going into action as helpful (e.g., presentation of nurses’ job situation in the media such as Twitter and Facebook to force awareness of the publics’ and stakeholders’ attention to their situation). Although the activity of the nurses was appreciated by the supervisors, some of them were critical of it, saying that this activity has a negative influence on the population’s attitude toward the nursing profession. They suggested the profession might become less attractive for later applicants, as also described earlier by Donelan et al. [51]. The media portrayal of nurses as “victims of serious workplace problems” [52] seems to be an external risk factor, as it can increase the staff shortage and worsen the relationship with inpatients [53].

With regard to cognitive coping strategies, both groups evaluated cognitive skills such as adaptation ability, solution orientation and reflective ability as highly protective. Looking at the differences, supervisors identified further cognitive skills in dealing with work-related stress. They mentioned the adjustment of performance expectation to the circumstances and the knowledge of sufficient coping strategies. Previous research related both coping strategies directly to nurses’ resilience, in accordance with Cam [54].

Both groups agreed that social support and positive professional and private contact with colleagues and supervisors can be a resource in dealing with work-related stress, as confirmed in previous studies [21]. Good relationships in the workplace, especially with supervisors, have an impact on the well-being of nurses [4]. Keeping its high impact in mind, social support can be promoted through good contact management (e.g., a high level of participation, joint activities with the department).

With regard to relational measures and strategies, both groups agreed that recreational phases during working hours were helpful. In addition, both groups saw support from the employer in the form of supervision, coaching and further training as stress-reducing measures; however, they reported that, to be able to better reconcile family and work, they would also like to see more flexibility in working time. Recent studies support this claim: individual-oriented working time flexibility was positively related to employees’ work–life balance, whereas organisation-oriented working time flexibility had a negative effect on them [55]. There was disagreement with regard to the structural changes that, according to the supervisors, would lead to relief in the long term but would be perceived as a burden by the nursing staff in the short term. This phenomenon may be related to a lack of participation in the decision-making process and thus reduced commitment of the nurses. Shared decision making can improve shared governance as collaborative decision making between nurses at every level and can strengthen the nurse workforce, improve job satisfaction and ultimately improve the quality of care [56]. Therefore, establishing a participative leadership style and early involvement in decision making could increase nurses’ sense of control, and short-term aversive consequences could be contextualised with positive long-term consequences, making it easier for staff to accept them. However, this remains speculative and should be the subject of further investigations.

Implemented measures such as the establishment of projects for the migration of nurses, the deployment of medical assistants, changes in conditions (e.g., merging wards, occupational health management) and expansion of nurses’ scope of action were well intended but seemed to have bypassed the registered nurses’ needs and were rather seen as further stressors by nurses because of the necessary adjustment efforts.

The supervisors in our study assumed that measures for the expansion of nurses’ scope of action and area of responsibility, such as training new employees and deciding on the priorities of tasks to be completed, could reduce power struggles, uncertainties and stress. However, members of the nursing staff experienced the shift of responsibility from the middle management level to followers as reinforcing or increasing uncertainties of responsibility and decision-making authority. A previous study reported that an expansion of responsibility with limited autonomy and lack of career opportunities caused dissatisfaction and conflicts, as further stress factors of nurses [49]. Consequently, to increase job satisfaction and reduce work-related stress, an expansion of responsibility should go hand in hand with an expansion of the scope for action.

As a further example of bypassing the needs of nurses, nurses reported stress due to a lack of knowledge, an assumption not shared by supervisors. Training opportunities to improve knowledge and reduce stress are offered mostly after working hours, which, as nurses reported, would in turn lead to further stress. Empirical studies have shown a divergence between the level of knowledge, attitudes and nurse practice [57] and revealed a negative correlation with stress [58]. Therefore, educational programmes presumably would have a negative effect on the experience of stress and a positive influence on work performance. Offering those programmes during working hours may decrease stress and would not generate additional workload.

Overall, nursing staff and supervisors predominantly agreed regarding their assessments of risk and protective factors of stress, and supervisors showed an understanding of the situation of their followers. In our interviews, nurses made support needs (individual contact, work facilitation, further training and gratifications) clearly visible but tended to hand over the responsibility for psychological well-being and needs to their supervisors and therefore put supervisors and the institution in charge. Conversely, supervisors also returned the responsibility of mental health to nurses. For example, in our study, nursing staff associated the required subordination of their work processes with a lack of supervisor appreciation, which would cause mortifications that resulted in low self-confidence and compensatory overcommitment (see the internal factors above). Nursing staff and supervisors differed with regard to the provider of appreciation. According to supervisors, appreciation should come from the person who caused the injury, not automatically from supervisors or nursing staff. Although the negative influence of low appreciation on job satisfaction [59] and burnout [60] is well known, the question of who must provide the appreciation has not yet been sufficiently investigated.

These results provide a reason to reconsider procedures in practice, research and training approaches. First, it should be positively emphasised that there is little difference in the perceptions of nurses and supervisors, suggesting that there is an awareness of the situation of the nurses. The differences mentioned might be based on the interindividual differences of the nurses that were unknown or not salient to the supervisors. It can be helpful for both parties to take the perspective of the other party and to be understanding and considerate of the motivations. The fact that we still have to assume that we are dealing with two different parties with different motivations shows that nurses and supervisors are only partially understood as a team. It is this front or camp mentality that makes it difficult to work together effectively for the benefit of the staff. Here direct communication seems to be helpful for bringing this into the awareness of the decision makers and to be able to take it into account in the implementation of relational and occupational (health) measures. However, at the same time, it should not go unmentioned that employers and supervisors cannot take into account all of the negative consequences and individual circumstances of the nurses. Rather, this requires a compromise between both parties—a certain flexibility on the part of the nurses in taking up and implementing the relational and occupational (health) measures but also on the part of the institution in implementing the measures. Because of incongruent perceptions, expectations, and disagreements, it is necessary that both groups remain in close contact and exchange with each other to minimise stress for the nursing staff and to improve cooperation.

As implications for further research, quantitative and empirical studies should be applied to verify stress and resilience factors from the perspective of nurses und supervisors. Because we conducted the study at a maximum care hospital, a comparison with hospitals of other care levels is necessary to verify the validity of the data.

In addition to implications for practice and further research, there are also implications for the development of training and education programmes by including the above-mentioned factors in the contents. The content of the training should be adapted to the modern way of working and supplemented with the promotion of personal responsibility (self-development). This means that, in the future, to promote nurses’ self-efficacy and self-confidence and to enable them to use participative offers, modern training should focus on strengthening the resources and skills of nurses on a cognitive, behavioural and emotional level. We hope that the development of excellent training courses for improving the health of nursing staff will continue, to once again make this indispensable and enriching profession more attractive and health conserving.

This study has several limitations that should be considered. First of all, the self-selection bias should be mentioned. Although we attempted to draw a broad representative sample, it is possible that participants triggered by the topic in particular registered. The second point is a possible social desirability bias, as participants were familiar with the topics of stress and stress management. This escalated into a controversial topic in the media at the time and the pressure to conform was so high that dissenting opinions were often not voiced, especially in focus groups. The short duration of the focus groups may have favoured this. The third point concerns generalizability. The participants were employees of a university hospital, so the statements are valid for this setting. Generalization to other work settings must be examined. Fourth, as our data is from 2018, other important risk and stress factors related to the Covid-19 pandemic were not taken into account. As this is the first study on the risk and protective factors of stress from different nursing hierarchy levels and reveals the first important results, we hope to inspire future quantitative work on cross-hierarchy stress-related phenomena to verify the observed phenomena.