Introduction

Health system resilience is defined as a health system’s ability to “anticipate, prevent, prepare for, absorb and adapt in response to, and recover from a wide variety of shocks and stressors while delivering high-quality individual and population health services as needed” [1]. Resilience is a complex construct and can be applied to different contexts, including to individuals, groups, organisations, and systems [2]. While political sciences, environmental sciences, history, disaster planning and other fields have largely described resilience as a system’s ability to absorb disturbances and perform when faced with shocks [3], psychology primarily uses the term to describe an individual’s ability to deal with, and recover from, a shock or strain [4, 5]. It can thus be argued that resilience can be assessed at both individual and system level. Blanchet et al.’s resilience framework categorises the ability of a health system to manage a shock according to three capacities [6]: (i) the absorptive capacity ensures that the system can maintain health services using previously existing resources and capacities; (ii) the adaptive capacity refers to the system’s ability to maintain health services with fewer or different resources by making organisational adaptations; and (iii) the transformative capacity describes the system’s ability to transform its functions and structure to respond to changes [6]. While recognising the need for a health system to absorb, adapt, and transform, recent debates on health system resilience also increasingly highlight the need for a health system to further evolve through learning from shocks [7]. It is anticipated that such evolution can eventually lead to increases in the system’s resilience and an improved capacity to deal with future threats of a similar nature [7]. This literature recognises that a shock requires a system to substantially change and that it is therefore unlikely that the system will simply “bounce back” to its original state. Instead, a resilient health system will transform by implementing lasting systemic change and improvements after the shock [7].

Studying resilience can be of practical value as it provides insights into the ways in which components of a system can facilitate the successful management of a shock for the system as a whole and the implementation of system-wide adaptations. If what contributes to resilience is understood, it is possible to support a system in advance of, and during, a shock [8] as well as in post-shock transformation. Studies that investigate how a health system behaves during a shock can thus help to identify potential for strengthening the system. Given that the COVID-19 pandemic resulted in an–in modern times–unprecedented shock to health systems worldwide, it provides a unique opportunity to study health system resilience. We propose here to study health system resilience by focusing on the ways in which changes in one of the health system building blocks, namely the health workforce, have system-wide effects.

Our study investigates health system resilience by analysing the system-related factors that supported health care workers (HCWs) during the COVID-19 pandemic [9]. The health workforce as one building block of a health system [10] lends itself as an object of study of resilience in pandemic times for two significant reasons. First, the health workforce plays a critical role in ensuring health services are maintained during pandemics, which has once again become apparent during COVID-19 [9]. Resilient HCWs are thus key contributors to health system resilience [9]. Second, empirical analyses of the situation of HCWs during the COVID-19 pandemic provide evidence that HCWs were greatly affected with impact on their psychological wellbeing [11–15]. This means that the pandemic posed a challenge to HCWs’ resilience. Germany provides a good case to study health system resilience with a specific focus on the health workforce as a building block as the German health system had to deal with a comparatively high number of COVID-19 cases, while also struggling with a chronic shortage in HCWs [16–18]. Box 1 presents a short summary about the COVID-19 pandemic, including the number of cases and fatalities among the general population and among HCWs, impact on the health system. Understanding which factors of the German health system supported HCWs during the pandemic can retrieve lessons that will help prepare the system and those working within it for future public health crises.

Materials and methods

We conducted focus group discussions and individual interviews with HCWs as well as individual interviews with key informants involved in decision-making around health service provision as part of an international, mixed-method study assessing HCWs’ experiences during the COVID-19 pandemic. Qualitative methods were chosen as they allowed a detailed investigation of the views of HCWs themselves as well as of key informants familiar with the situation of HCWs and in-depth insights into the system-related factors that supported HCWs during the COVID-19 pandemic [22]. Interviews were used to elicit detailed descriptions of experiences and views from both HCWs and key informants and helped to gain a variety of perspectives. During the FDGs, some aspects that had been raised or had emerged as topics of contention during the interviews were discussed. The FDGs thus provided more detail on selected aspects, helped to clarify some issues, and shed light on controversies and disagreement. HCWs were recruited between 1 February and 30 November 2022 through a quantitative survey on stress and coping among HCWs during the COVID-19 pandemic which was conducted as part of the overarching study. Survey participants were recruited via the health facilities in which they worked or via professional networks and organisations. HCWs who completed the survey and expressed an interest to participate in the qualitative arm of the study were contacted individually via email. HCWs were included in the study if they were: (i) aged 18 or above, and (ii) had been working in a health facility during the COVID-19 pandemic.

Data were collected between April and December 2022. HCWs who participated in the interviews and FGDs included medical doctors, nurses, allied health workers, nursing students, and other staff working in health facilities during the pandemic. We invited a wide range of HCWs working across all levels of health care (primary, secondary, tertiary) and in a range of health facilities in Germany, including private practices, hospitals, and care homes. Potential participants received information about the study and were given the option to either participate in an individual interview or a FGD.

Key informants included political decision-makers, health facility managers, representatives of professional associations, representatives of health care providers, academics, and other individuals with decision-making functions. Key informants were sampled through an online search where potential interviewees were identified from media and other reports. Consecutively, individuals who had commented on the situation of HCWs during the pandemic were listed as potential participants. Individuals were considered as key informants if they were (i) aged 18 years or above, and (iii) involved in decision-making around health service provision. Purposive sampling was then undertaken to ensure a diverse selection of key informants from various regions, areas of work, genders, and roles. Selected individuals were contacted between 1 April 2022 and 30 November 2022 via email, provided with comprehensive study information and asked to participate in a semi-structured interview.

All participants were required to be 18 years old or above and to provide voluntary informed written consent. Information sheets and informed consent forms were shared with the participants before the interview or FGD. Other than interactions about logistical arrangements, no relationship was established prior to the interview of FGD. Interviews with key informants were conducted online via videoconference or face-to-face, while all HCWs’ interviews and FGD were conducted online. HCWs received a 25 Euro voucher as compensation for participation in interviews or FGDs, while key informants did not receive compensation. Interviews and FGDs were jointly conducted, usually with one senior researcher acting as interviewee/facilitator (HW, RB, BB) and a junior researcher or administrative assistant as a note taker (RB, SA). All interviewees and facilitators were members of the study team and employed as staff at the institution leading the study. Interviews with HCWs and key informants took on average 55 minutes (min: 37 minutes, max: 71 minutes), whereas FGDs lasted on average 70 minutes (min: 57 minutes, max: 79 minutes). All interviews and FGD were conducted in German and audio-recorded. Field notes were taken for each interview and FGD. Verbatim transcription and translation into English was performed by an external company to facilitate analysis among the international study and the team of researchers. Data collection was concluded as data saturation was reached.

Topic guides were developed based on a literature review [11] and covered the following topics: personal experiences, challenges, coping strategies, factors supporting or hindering HCWs in coping with challenges, support needs, and recommendations for support. Guiding questions were defined for each section and tailored to the data collection method and target group (FGD with HCWs, interviews with HCWs, and interviews with key informants). Topic guides were amended as data collection progressed to account for emerging themes from previous interviews or FGDs and clarify conflicting data.

All data were analysed according to thematic content analysis [23] using NVIVO software 1.7. As a first step, the coding team members familiarized themselves with the transcripts, reading and rereading them, and taking margin notes. A codebook was then developed using a priori, or pre-defined codes that came from the interview topic guides, and in vivo codes that were identified from wording used by respondents [24]. The development of the codebook was an iterative, collaborative process. As part of this, a random sample of five interviews was independently coded by three researchers. Coding was discussed, and the codebook was revised thoroughly according to consensus. The final codebook included a definition for each code, specifications for when to use it, and specifications for when not to use it. All transcripts were systematically coded by six members of the study team (ME, EMM, RB, RU, BB, HW), applying the revised codebook. One third of all transcripts was double coded to ensure coding was consistent. Consistency checks were performed by a senior researcher (HW). The analysis involved the integration of similar thematic coding, with findings being subsequently triangulated across different data sources for each theme. As part of the analysis, particular attention was paid to identifying evidence that appeared to contradict the explanations being developed and to considering alternative explanations. Over the course of data collection and analysis, all members of the research team reflected on their own positioning with regard to the research topic. We particularly reflected on the potential impact of the following aspects on data collection and analysis: (i) our professional backgrounds as researchers and (for selected members of the research team) as health professionals; (ii) our affiliation with an institution that communicated extensively during the COVID-19 pandemic and provided evidence which informed political decisions with major impact on HCWs; (iii) the participants‘ potential intentions to promote a personal or political agenda or to, retrospectively, recount situations in a way that supported their views and opinions.

Results

A total of 60 HCWs participated in the study, with 43 HCWs taking part in 10 FGDs and 17 HCWs in individual interviews (Table 1). Additionally, individual interviews were conducted with 18 key informants. Throughout data collection, it became apparent that HCWs struggled with multiple challenges during the COVID-19 pandemic. Listening to their accounts of what supported them to successfully tackle these challenges and maintain resilience, five overarching, interconnected themes could be identified. The five themes related to the work context, the culture of the organisations HCWs worked in and the health system. In addition, individual factors were identified as influencing how HCWs managed COVID-related challenges. We therefore give a brief summary of the individual factors that were mentioned as mediating HCWs’ reactions prior to providing a detailed description of the five work- and system-related themes. Participants’ quotes are provided as illustrative examples to highlight specific aspects of each theme. Depending on the preferences of the respective interviewee or FGD participant, each quote is allocated to the participant’s personal data, or only generic information is given that the quote was provided by a study participant.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Socio-demographic data of HCWs who participated in the study.

https://doi.org/10.1371/journal.pone.0317231.t001

HCWs reported a number of individual factors and strategies that were of importance in dealing with the pandemic situation. HCWs recalled an inner strength that carried them through the pandemic. One respondent referred to being “resilient” (paramedic, male, 28 years, tertiary level), another described an “inner conviction that one can do one’s part” (medical assistant, female, 35 years, primary level) as a supportive trait. A proactive attitude was further mentioned as conducive to successfully handle difficulties experienced. Believing in oneself and a sense of self-efficacy were further highlighted to ease the situation. A focus group discussant expressed:

“What helped us get through this time? I believe it was the faith in each other. […] faith in people and their work. Having hope that things will improve and we will eventually get through this time.” (focus group discussant)

Respondents further stressed the importance of recreation. Walking in the outdoors, being in nature, doing sports, and other recreational activities were mentioned as helpful strategies to maintain resilience. Family members and friends as well as spending time with others in ways that were in line with pandemic restrictions (e.g. online, with the family or as a household) emerged as important sources of support and comfort. Distancing oneself from work and the pandemic events, and prioritising time for social interactions and recovery were mentioned as useful mechanisms to deal with the challenges and demanding situations at work. HCWs reported that keeping boundaries between work and private life was often challenging given the high work demands, the absence of colleagues and existing staff shortages. Deciding to do so was described as an act of self-care, as the following quote by a focus group discussant illustrates:

“For me, it was, well, self-protection to say at some point: ‘Stop, I’m now keeping my regular working hours, I’m no longer working until I drop, but have now set myself fixed working hours, which I also try to keep as a rule.’ Simply as self-protection, otherwise you really collapse.” (physiotherapist, female, 56 years, primary level)

Discussion

This paper is the first to provide an in-depth analysis of the system-related factors that supported HCWs in Germany during the COVID-19 pandemic. Our analysis identified six categories, including: appreciation and encouragement; team support and communication; collaboration across professions and institutions; informational, material and technical support; leadership and participation; and creativity and innovation. By focusing on supportive factors, the analysis adds a salutogenetic perspective to the large body of evidence focused on stress, mental health and the negative consequences that the COVID-19 pandemic had on the health workforce and the health system [25–28]. Our analysis suggests that HCWs can demonstrate high self-efficacy and have the capacity to adapt in the face of a crisis, particularly when the environment and the system enables and supports such positive action. Through successfully handling the various challenges they faced during the COVID-19 pandemic, HCWs substantially contributed to the continued functioning of the German health system. More importantly, however, our analysis highlights how crucial systemic and organisational factors, including organisational culture, conducive working conditions, supportive interactions between employers, employees and colleagues, and opportunities for participation, innovation and creativity are for health system resilience. These findings are in line with Abimbola and Topp who caution against the tendency to frame resilience solely at the individual level, which encourages individuals to “draw on internal strengths and resources to make up for weaknesses in the health system in the face of acute shocks or chronic stress” [29 p2], as it distracts attention and efforts away from addressing such systemic weaknesses. While we acknowledge that HCWs are not a homogenous cohort and that different types of HCWs experienced COVID-19 in different ways [30], our findings show that supportive factors at the organisational and systemic level are not specific to certain health facility types, professions or workplaces. Rather, organisational support was experienced by HCWs of all professions working in all types of health facilities, with no distinct patterns of more supported HCWs or more supportive workplaces emerging from the analysis. Our analysis, in fact, provides evidence that systemic and organisational support can be provided in any health facility and support any type of HCW.

Our analysis shows that the COVID-19 pandemic–in addition to posing huge challenges–offered extensive opportunities for innovation, creativity, and organisational change and transformation, and thus for health system strengthening. Ostensibly negative developments, like the malfunctioning and breakdown of previous hierarchical structures, enabled those working within the system to break from rigid routines and structures, try out new processes and develop creative and alternative solutions. Due to its cross-sectional nature, the study presented in this paper cannot draw conclusions about the ability of the German health system or the organisations within it to transform pandemic adaptations into long-lasting systemic change or about the sustainability of the changes that were made. It only draws attention to the potential for such transformation and offers a glimpse of what such transformation could entail. Yet, the findings support previous research which highlights that plasticity and transformation are key for resilience [31]. Our analysis describes multiple instances of innovation, adaptation and change during the COVID-19 pandemic, thus testifying to the German health system’s resilience.

Barasa et al. conceptualise health systems as complex adaptive systems, and define resilience as “an active process within a dynamic health system that is constantly navigating challenges” [32 p4]. They argue that, in order to facilitate this process, conditions have to be created that enable “bottom-up dynamics” [32 p3] which allow development and innovation. Our analysis supports the concept of the health system as a complex system that adapts and learns, and the importance of bottom-up dynamics and self-initiative of those working within the system for the release of creativity, innovation and transformation. It further identifies several opportunities for learning from the adaptations that were made during the COVID-19 pandemic, which–if seized—are not only likely to contribute to health system resilience but also to health system transformation, organisational change, and the enhancement of HCWs’ working conditions. Such adaptions would likely improve both the system’s as well as individual HCWs’ capacities to deal with future health system shocks. The implementation of such changes is likely to require the consideration and reform of a range of health systems functions, including governance, financing and human resources. It can be assumed that systemic support is needed in order to maintain any innovations that were initiated during the COVID-19 pandemic. A repeat study assessing current support and the maintenance of the changes described in our study (e.g. with regard to participatory leadership and organisational culture, interactions between employers, employees and colleagues, or working conditions) could investigate whether the necessary support for health system transformation is being provided and adaptations could be transformed into sustainable change.

We identify self-determined action and participation as key areas for health system transformation and learning. The findings presented above suggest that self-determination at work can increase motivation and facilitate creativity and innovation. According to Hackman and Oldman, a job with a high level of self-determination relates to the degree of independence that an individual has within their job and with regard to scheduling the work and determining work procedures [33]. In the context of the pandemic, HCWs positively stressed and valued opportunities for exerting self-determined action, e.g. in relation to decisions about implementing pandemic policies, ensuring the maintenance of health services and collaborating with colleagues from other disciplines. Previous studies on HCWs show that self-determination on the job is positively associated with psychological well-being [34], job satisfaction [35], and willingness and ability to stay in the job [36], whereas a lack of self-determination leads to reduced well-being [37], and thoughts of quitting the job [38]. Conducive work conditions are further assumed to increase self-determination at work and self-leadership [34] among HCWs. In line with this literature, our study identifies the participation of HCWs as crucial. Our analysis further suggests that establishing formats that allow interaction and exchange between those making managerial and regulatory decisions and those providing patient care is likely to increase not only satisfaction, engagement and self-determined acting at work, but also the adoption and implementation of policies and work processes that are actionable and in line with health care standards, and health system resilience. Participatory decision-making also promises to facilitate innovation, not only in pandemic times but also in routine care.

Participation and self-determination can only be achieved if the organisational culture values eye-level interaction and self-initiative [39]. Organisational culture is strongly influenced by leadership and management, so it is unsurprising that leadership styles and managerial support were identified as key supportive factors. Positively perceived assistance ranged from appreciation to practical support in the form of information, resources and technical training and to opportunities for communication, feedback and exchange of views. A supportive organisational culture further included interactions and exchange with team members and colleagues from other professions and institutions. The breaking down of boundaries between professions and institutions was perceived as facilitating collegial and effective working procedures, which helped to master pandemic challenges. There findings are in line with a recent OECD publication which highlights greater interdisciplinary task sharing and team work as a key area to increase health system resilience [40].

This study has number of limitations. Study participants’ accounts of support were not validated via a quantitative questionnaire, which might have allowed a more objective measurement and the comparison of different types of support. Instead, the study provides in-depth insights into the factors that were perceived as supportive. Simultaneously, the cross-sectional study design did not allow the exploration of support over time or the sustainability of change and support. Future longitudinal studies could allow the analysis of support at different pandemic stages or a comparison of support provided to HCWs during the pandemic with support in non-pandemic times. Finally, the sample was restricted to key informants identified via media and other reports and HCWs volunteering to participate in the study. While accounts are likely to be influenced by self-selection bias, we are confident that the data was saturated and we captured key themes through the rigorous collection and analysis of the data. Purposive sampling further ensured that respondents were included from different geographical regions, working in various health facilities and other organisations, and holding a broad range of points of views, which makes it likely that the accounts represent a breadth of realities.

The analysis presented in this paper aligns with the existing literature which highlights the benefits that can be reached by increasing participation and self-determination, changing organisational cultures, and improving the work conditions in the health sector [41, 42]. Our main finding is that systemic changes which enhance participation allow HCWs shape their work environment as well as work processes, and increase respectful and appreciative interaction between employers and employees and between colleagues of different professional groups have the potential to increase health system resilience. Implementing such changes might seem challenging in the hospital sector where hierarchies tend to be steep and rigid and procedures inflexible [43, 44]. However, evidence exists of health facilities successfully adopting alternative, less hierarchical work concepts and styles [41, 42]. This study suggests that the adaptation of such innovative ways of working by other health facilities could increase health system resilience.

Supporting information

Acknowledgments

We thank Smilla Johann, Annika Witzigmann, Nadia Mangold, and Swetlana Muminow for their support in making this study possible while working as student assistants and interns in our research unit. We thank Angela Schuster, Sophie Müller, and Sameh Al-Awlaqi for study coordination and invaluable logistical support. We thank all participants for taking part in the study.