Burnout in modern-day health care: Where are we, and how can we markedly reduce it? A meta-narrative review from the EUREKA* project

Abstract

Background

Burnout is disrupting the health care workforce, threatening the livelihoods of health care workers and the probability of safe and effective patient care.

Purposes

The aims of this study were to describe the evolution and gaps in burnout research and identify next steps to advance the field and reduce burnout.

Methodology/Approach

We formed a learning community of burnout scholars and Chief Wellness Officers, sought recent review articles for a meta-narrative synthesis of themes on health care worker burnout, and conducted focus groups with learning community members.

Results

In 1,425 systematic burnout studies found in a Medline database search of systematic reviews published since 2018, 68 were retained for analysis. Many focused on individual interventions (e.g., mindfulness), paying inconsistent attention to (a) what comprises burnout, (b) prevalence and contributors, (c) theories underlying it, (d) presence in marginalized populations, and (e) innovative research methods. There was consensus that burnout poses a global crisis, but there was no agreement on how to address it. Focus group participants noted that although burnout research is now “mainstream,” health systems commit insufficient resources to addressing it. They proposed that emphasizing organizational finances and patient safety may make burnout a priority for health systems.

Practice Implications

Despite burnout's progressing unabated, many organizations do not employ known burnout indicators (worker dissatisfaction or turnover) as wellness metrics. Research into organizational contributors to burnout, rigorous evaluation of interventions, and organizational adoption of research findings into systemic action are urgently needed. A well-supported international research agenda is required to quickly move the field ahead and reduce or ultimately eliminate burnout.

Burnout is a process of demoralization and decreased energy and enthusiasm for work, stemming largely from overwork, obstacles to feeling a sense of mission and purpose, or the lack of alignment of values with one's organization (Rotenstein et al., 2023). Exhaustion and distancing from work may result and, when persistent, often culminate in one leaving a job to find more sustainable work elsewhere (Prasad et al., 2021) or leaving the field altogether. Despite a global burnout epidemic (Dean et al., 2024), with burnout in some countries' health workers exceeding 50%, there is limited evidence to guide systems in preventing it.

Maslach et al. (2001) defined three components of burnout: emotional exhaustion, depersonalization, and lack of personal accomplishment. Our team's recent work suggests health care workers (HCWs) often define burnout considering other work and home matters, suggesting a more expansive definition. Distress signals, such as betrayal by an authority figure leading to moral injury (Dean et al., 2019), overwork and compassion fatigue (Cohen et al., 2023; Lee & Cha, 2023), and vicarious trauma from caring for those with trauma, can correlate with burnout but are separately defined (Mantri et al., 2020), thus representing different constructs.

Burnout is a warning signal, pointing to a dysfunctional organization (Linzer et al., 2009; NASEM, 2019). A system with excessive work in chaotic environments without suitable controls or attention to aligned values is associated with lower quality of care and burnout (Linzer et al., 2009). HCWs in such systems have less favorable experiences, report more errors, and often desire to leave. To avoid downward spirals of demoralization and departure, a better understanding is required of burnout and knowledge of means to combat it.

The Eliminating burnout through a Unified REsearch approach: Knowledge to Action (EUREKA) project is supported by the Agency for Healthcare Research and Quality (AHRQ) to determine remediable sources of burnout and scientifically supported approaches to reducing it. A scholarly learning community (LC) in EUREKA is adding to this meta-narrative review with focus groups and creating a national agenda for burnout studies. EUREKA is also building a cadre of scholars for this work through a learning health system (LHS) network, supported by AHRQ and PCORI (Patient-Centered Outcomes Research Institute).

This paper's meta-narrative review (Greenhalgh et al., 2005) describes how burnout research has evolved with time, including themes and knowledge gaps, supplemented by focus groups with wellness leaders and burnout scholars to propose a national agenda to move the field ahead. This paper will address (a) how our understanding of burnout has progressed in the past 20 years and (b) areas of research emphasis that will allow a consistent and effective approach to eliminate burnout and create more sustainable work lives for all HCWs.

Theory

The most prominent theory underpinning burnout research in the psychology and management literatures, among the few studies that actually do cite theory, is job demands–resources (JD-R) theory (Bakker et al., 2023). Yet, JD-R theory is more a theory of job design, which focuses on worker well-being and its pathways between job design and performance (Bakker et al., 2023). In today's health care work context with well-known job demands and chronic lack of resources, a more practical theory might be conservation of resources (COR; Hobfoll, 1989). COR posits loss of resources as a stress trigger and proposes that individuals try to minimize loss of resources; even anticipation of resource loss can cause stress. An important tenet of COR is gain and loss spirals: Those who have more resources can protect themselves from future losses, whereas those with less have a harder time gaining new resources. Thus, COR predicts HCWs with fewer resources will be vulnerable to burnout (Williams et al., 2020).

In burnout literature, most studies conceptualize burnout as a static state that is an outcome of poor working conditions or as a predictor of subsequent negative consequences (Williams et al., 2020). In contrast, COR theory allows conceptualizing burnout as a process, with a resource loss spiral consistent with the “burnout cascade” described in occupational health literature (Weber & Jaekel-Reinhard, 2000). Early stages in the spiral include reduced activity (emotional exhaustion), followed by distress and withdrawal (depersonalization) and culminating in despair (reduced personal accomplishment). Under this framework, interventions can target specific stages individuals face in the cascade; this may help explain why some interventions work and others do not. For example, Maslach et al. (2001) argued that it is often the best workers who experience emotional exhaustion because when they face barriers, they often increase energy expenditures or become hyperactive to achieve goals. If hyperactivity is seen as the first phase in the burnout cascade (downward spiral), then early theory-driven interventions could be developed to target high-quality active workers.

COR theory also posits that individuals benefit from resource caravans and resource pools (Hobfoll et al., 2018). Caravans are bundles of resources that are correlated, such as education, financial resources, and health, or psychological resources such as job security and optimism. Resource pools (individual and organizational resources) are important because those with richer pools can find the resources necessary to invest for future gain or prevention of loss.

Examining COR's nuances can lead to interventions targeting burnout stages with particularly effective interventions. For example, a systematic review proposed that antecedents of burnout start at the organization level, but protective factors occur at individual and relational levels (Sibeoni et al., 2019). Using COR, organizations could focus resource pools for workers to reduce burnout antecedents and develop interventions to reduce loss spirals.

Method

Meta-Narrative Review Study Design

The approach (as in Greenhalgh et al., 2005) is a systematic review of reviews. It was chosen to identify themes that may have entered the literature at different points in time, rather than quantitatively assessing multiple heterogeneous, often small sample size, single-site studies. A Medline search was conducted for systematic reviews published since 2018 on HCW burnout and mental health (Appendix A, Supplemental Digital Content, http://links.lww.com/HCMR/A168). We searched for reviews published between 2018 and 2024, and most reviews examined articles published since the inception of burnout studies.

Inclusion/Exclusion Criteria

Included articles were systematic reviews of studies examining burnout in nonspecific HCW populations. After removing duplicates and articles not written in English, 1,390 articles remained; titles and abstracts were reviewed for 703; and after title and abstract screening, 488 were retrieved for possible full review. Articles were excluded if they were not systematic reviews, were focused on specific HCW specialties (e.g., oncologists), or did not focus on burnout, resulting in 68 included studies (Figure 1). If included, a coauthor summarized main findings of the identified article and any observations by the review team.

Figure 1.

PRISMA flow diagram of systematic reviews included in the study

Focus Group Study Design and Participants

During the first EUREKA conference in Durham, NC, in April 2024, we conducted two focus groups; participants had a national presence in burnout research, a position as a Chief Wellness Officer (CWO), or a leadership role in the AHRQ/PCORI LHS network where we plan to develop avenues for dissemination of future scholarship. We designed four focus group questions based on project goals and experience of project leads, probing participants' thoughts on strengths and weaknesses of current burnout research and what new studies are needed. The Hennepin Healthcare Institutional Review Board exempted the study from human subjects review. Project leads (M.L., E.C.O.) each led a 45-minute focus group with seven to eight participants (total participants = 15). Focus groups were recorded, transcribed, and deidentified.

Qualitative Analysis

Two coauthors skilled at qualitative analysis (S.M., E.S.) completed an initial (open coding) review of transcripts prior to determining a coding strategy. Given the focus group design, data were deductively analyzed by question to identify themes related to (a) strengths and weaknesses of burnout research, (b) future studies needed to address burnout, and (c) funding required for future research. The two coauthors independently coded focus group data, identified themes in the main areas, and reached consensus on reporting results.

Results

Meta-Narrative Results

Our search yielded 1,425 records, 68 of which were eligible systematic reviews of empirical burnout studies and included in our final analysis (Figure 1). Although most studies had slightly different research questions, most focused on either predictors or consequences of burnout in HCWs, or interventions to reduce burnout. Because the burnout literature does not have a standard approach vis-à-vis underlying theory, the overall results lean in numerous, unconnected dimensions. The overwhelming majority of studies noted in the reviews employed cross-sectional study designs.

In addition, we found a recent trend of systematic reviews using specific, validated tests for quality and bias in their reviewed studies. Notably, those who conducted such tests found many included studies were not of the highest quality. Even intervention studies, including randomized controlled trials, were often considered of lower quality due to problematic study components. Details are in Table S1 in Supplemental Digital Content. Themes (main findings by authors and observations by the review team) follow:

1. Where we are now in understanding trends over time and across professional groups. There has been a steady rise in the number of articles on burnout, with a sharp rise just before and during the pandemic. Most articles address physicians and nurses, with few in other HCWs.

2. Prevalence. Burnout prevalence in physicians was close to 27% in 2000 (Linzer et al., 2000) but rose to over 60% in U.S. physicians during the late pandemic (Linzer et al., 2022; T. D. Shanafelt et al., 2022), more recently falling to around 50% in physicians. There are less certain prevalence statistics outside the United States (Doraiswamy et al., 2021) and in other HCWs, although fairly consistent recent numbers in nursing are around 60% (Prasad et al., 2021). Some studies suggest burnout prevalence may be high among other clinical staff (Mallick et al., 2024; Rotenstein et al., 2023). Higher prevalence was noted (Forrest et al., 2021) where COVID risk was high.

3. Predictors (aggravators and mitigators; Table 1). There is a general agreement on workload as a predictor; other predictors include work pace, work control, time pressure (e.g., for clinical notes), home electronic health record (EHR) time, and culture of the organization around trust, quality, values, communication, and cohesion (Linzer et al., 2009). A previously unrecognized and very strong mitigator recently identified (Linzer et al., 2022; Prasad et al., 2021) was workers feeling valued by their organizations, with six recommended strategies to improve this (Stillman et al., 2024): (a) ensure safety; (b) address financial challenges; (c) provide transparent, frequent communication; (d) create effective teams; (e) train empathetic leaders; and (f) provide organizational support. In terms of equity, challenges are found in association with race, bias, and feelings of being unappreciated, especially when working in historically marginalized areas with high rates of death associated with one's race and community

TABLE 1.

Themes from the meta-narrative analysis: predictors and consequences of burnout

	Themes	Illustrative citationsa
Predictors	Individual factors (age, gender, underlying mental health status, social support, lack of sleep)	Claponea et al. (2022) Membrive-Jimenez et al. (2020)
	Work-related factors (job role, lack of respect at work, lack of perceived support from leadership, frontline vs. non-frontline settings)	Kesarwani et al. (2020) Meredith et al. (2022)
	Environmental factors (exposure to workplace violence, lack of access to protective equipment, working in an economically deprived area)	Chemali et al. (2019) Wright et al. (2022)
Consequences	Individual health outcomes (anxiety/depression, suicidality, poor sleep, substance abuse)	Finnerty et al. (2022) Ryan et al. (2023)
	Professional outcomes (low job satisfaction, turnover, decreased job performance)	Dall'Ora et al. (2020) Lo et al. (2018)
	Patient care outcomes (lower quality, lower patient safety, increased reported medical errors)	Hodkinson et al. (2022) Tawfik et al. (2019)

^aSee Supplemental Digital Content for a comprehensive list of references included in the meta-narrative review.

4. Consequences of burnout (Table 1) include turnover, self-harm, depression, disengaged teams, demoralized leaders, decreased quality of care, compromised patient/provider safety, increased absenteeism, reduced efficiency, and increased costs. Strong links exist between burnout and intent to leave, with about one third of those intending to leave doing so (Hann et al., 2010). Depression and impaired mood and anxiety co-occur (Meredith et al., 2022; Prasad et al., 2021; Ryan et al., 2023), with a moderate correlation with burnout (r = .4). Leader burnout is also described but infrequently studied (Niinihuhta & Häggman-Laitila, 2022; Sullivan et al., 2023). There are demonstrated relationships of burnout with lower quality of care and patient safety (Linzer et al., 2009); although with clinical outcomes, it is often not the burned out clinicians giving lower quality care (Rabatin et al., 2016) but rather the organization with high rates of burnout, which is associated with adverse patient outcomes (the “clinician as buffer” hypothesis; see Linzer et al., 2009). Organizational-directed interventions require further study as per DeChant et al. (2019), who also note that some high-quality studies identified benefits from interventions in workplace processes, team-based care, and the use of scribes.

A summary of findings from the meta-narrative review using Kuhn's “structure of scientific revolution” (from Greenhalgh et al., 2005) shows (a) a conceptual consensus on burnout as a meaningful metric; (b) theoretical relationships between burnout's components and consequences supported by models examining burnout loss spirals (Williams et al., 2020), resource pools, and caravans (Hobfoll et al., 2018); (c) methodologic gaps in terms of rigorously tested ways to solve it (Craig et al., 2021); and (d) a variation in burnout metrics used (Meira-Silva et al., 2022), with suggestions that which method is used may be less important (Brady et al., 2022; Tawfik et al., 2019) than an organization choosing one and staying with it.

Focus Group Results

Participants were well versed in the burnout literature given their roles and were able to add their perspectives on ways of “operationalizing” the literature into their roles and work lives. In particular, they noted several strengths of current burnout research, with an agreement that burnout research is now “mainstream” and that there is general awareness and a “burning platform” to address an issue that affects patient care and HCW retention. There was a sense that health systems know there is a problem (echoing findings in the review) but do not have (or want to commit) resources to fix it. One focus group suggested that implementing literature-based suggestions to address burnout would be facilitated by clarifying known links between burnout, finance (reduced turnover), and patient safety outcomes. Participants felt we were missing opportunities to address the issue, that interventions were mainly reactive, and that burnout is an issue in need of long-term investment for sustainable solutions. Current research at participants' institutions included (a) identifying underlying causes and risk factors (e.g., workload, lack of resources, and coping strategies), (b) learning how to manage burnout and foster wellness, and (c) highlighting relationships between burnout and patient outcomes. Participants and the literature both embraced a multidimensional approach, integrating psychological, social, and organizational perspectives in addressing burnout.

Participants also called attention to a need for longitudinal studies to track burnout over several years. This could reveal patterns, factors contributing to resilience, and the long-term impacts on career trajectories and patient care. They also sought studies on favorable impacts of technology (as in Craig et al., 2021, or Nguyen et al., 2021) and the influence of local culture and leadership styles to provide insights into mitigating burnout rates. Finally, they sought further qualitative research to capture HCW experiences and barriers to seeking support. Focus group respondents noted a need for increased federal grant support for intervention research; one respondent said funding could be more likely for studies connecting burnout to patient outcomes.

Discussion

In this meta-narrative review of systematic reviews of burnout in HCWs in varied roles across the world, supplemented with observations by thought leaders and CWOs, we found substantial inconsistencies among definitions, findings, and approaches. In part due to these inconsistencies, as well as lack of broad uptake by organizational leaders, and despite support by prominent national organizations, the rate of burnout has risen from 25% in physicians in 2000 to over 60% in 2021, with comparable numbers in nurses (Prasad et al., 2021) and far-ranging consequences for HCWs (Rotenstein et al., 2023) as well as the patients and systems they serve. Furthermore, the costs are enormous, with close to $5 billion per year lost in the United States due to turnover from physician burnout (Han et al., 2019) and $5,000 per nurse, which could be saved annually with a burnout reduction program (Muir et al., 2021). There is consensus that burnout's contributors include but are not limited to time pressure, work control, chaotic workplaces, and unsupportive organizational cultures and that consequences include turnover and mental health symptoms, sometimes serious. A deeper understanding of this complex area can build a foundation for intervention. Here, we review historical contexts, propose future research, and provide clear suggestions for mechanisms to reduce burnout within modern-day health care.

Contextual Features—A Link to Patient Safety

Focus groups raised the importance of linking wellness to patient safety. Although the literature frequently addresses this, this outcome is less often linked operationally to wellness programs. With the escalating amount of chronic illness care, pressure has built upon HCWs, with risks for their ability to deliver empathic care, build trust with patients, and be rewarded emotionally by therapeutically connecting with their patients (Rathert et al., 2024). Amidst a prevailing emphasis on productivity, there has been a concerning devaluation of essential clinical practices such as direct patient engagement and compassionate responsiveness as well as how these outcomes and experiences can relate to work conditions resulting in burnout. Articles in this review demonstrate the eroding effects of time pressure and provide evidence for how to foster working conditions for HCWs that allow response to patients' adverse life circumstances (Allwood et al., 2022).

Historical Threads in the Evolution of Burnout Studies

Due to the meta-narrative review being a review of reviews, a sense of individual contributions to the burnout field remained submerged. In this section, we provide a brief glimpse of how leaders in the field contributed over the course of the past three decades and how their work relates to constructs identified within the meta-narrative review. Christina Maslach's landmark work began in the 1980s, culminating in the “gold standard” Maslach Burnout Inventory, with three components: emotional exhaustion, depersonalization, and lack of personal accomplishment. The 27-item measure appears extensively in this review, although it lacks a cutoff for determining burnout and is proprietary, sometimes limiting its regular use. Similarly, Bakker and colleagues have led this work globally and outside of health care for decades (Bakker et al., 2023). Linzer's team pioneered briefer metrics (Freeborn, 2001) with recent data showing HCWs define burnout more expansively than categories found in the Maslach Burnout Inventory. Structural models predicting burnout have been robust over time (Linzer et al., 2001), predicting close to 50% of burnout variance. Gender differences identified in 1999 are still in urgent need of attention (Linzer et al., 2024).

Sinsky and colleagues have performed landmark time and motion studies (Sinsky et al., 2016) and identified challenges of the EHR. Their practice-changing research demonstrated ways to structure primary care more sustainably (Sinsky et al., 2013). The “Coping with Covid” study (Prasad et al., 2021; Rotenstein et al., 2023) defined wide-ranging impacts of work conditions on burnout. Shanafelt and colleagues have championed national surveys and models determining burnout prevalence and contributors among varied roles, including physicians and trainees (T. D. Shanafelt et al., 2022). Dyrbye et al. identified burnout contributors and impact in learners (Dyrbye et al., 2021), and West et al. have studied, among other change mechanisms, individual interventions such as meditation with good effects. Wallace et al.'s foundational paper in 2009 on wellness as a missing quality indicator predated creation of CWOs and set the stage for wellness research for the past two decades (Wallace et al., 2009).

O'Brien and colleagues in the HERO (Healthcare Worker Exposure Response and Outcomes) registry demonstrated the degree of burnout throughout the health care workforce (Forrest et al., 2021). This epidemiologic approach to burnout set the standard for incorporating large numbers (>50,000) of HCWs of varied roles into national databases to assess and reduce burnout across the workforce. Sullivan and colleagues brought primary care's challenges to light by highlighting the voices of 8,000 primary care clinicians during the pandemic (Sullivan et al., 2023). Sullivan et al.'s work also encompasses burnout among rural leaders, some of whom serve in dual roles as clinicians and leaders, two groups at a particularly high risk.

A Burning Question in the Meta-Narrative: Why Is What Is Known to Reduce Burnout Not Used?

HCWs observe that burnout studies reflect their work conditions and evolving challenges. What remains unclear is why so many organizations have not responded effectively. Due in part to fiscal concerns, organizations find it challenging to know how exactly to address burnout without “taking one's foot off the gas pedal.” Most systems do not formally integrate HCW well-being into their strategic plans nor align it clearly with their mission, vision, and values. Crucially, they do not incorporate measures of burnout and wellness as key performance indicators that hold leadership accountable. Without this strategic commitment, decisions made at the operational levels inevitably prioritize other elements of the Quadruple Aim—patient experience (Bodenheimer and Sinsky, 2014), population health, and costs—over HCW wellness. This imbalance perpetuates a cycle where organizational decisions consistently disadvantage wellness. Meanwhile, the federal government has empowered state and professional organizations to review and advise, but few systematic standards for wellness have emerged, such as have existed in occupational health and safety for decades. This translates into a call for action. We remain optimistic that with the current focus on burnout from the National Academies of Science, Engineering, and Medicine report (National Academies of Science, Engineering, and Medicine, 2019) and the Surgeon General's report (Office of the Surgeon General, 2022), the time is right for a national shift, as Wallace et al. proposed 15 years ago, to making wellness a quality indicator for health systems. This may include requiring implementation of evidence-based approaches to burnout reduction to qualify for state or federal funding, at the same time as future high-quality studies are conducted to extend our knowledge of what works best for these purposes.

Innovations and New Developments

CWOs

Approximately 10 years ago, the concept of the CWO began to permeate the wellness world. Foundational work from Stanford (Shanafelt et al., 2020) in a course for CWOs has spread the concept, and an increasing number of CWOs are being appointed across the country. Amidst the burgeoning field of worker wellness, there is a compelling need for dedicated leadership that can stay current with best practices, collaborate effectively with other executive leaders to craft and implement comprehensive workforce wellness strategies, oversee wellness performance metrics, and serve as an effective organizational influencer to ensure that wellness initiatives are prioritized and integrated into the broader strategy and operations of the health system. With a seat at the table alongside other C-suite colleagues, the CWO can ensure initiatives are not fleeting trends but rather sustainable measures. The CWO, supported by staff and resources, can serve as a key resource to creatively navigate tensions among competing priorities. Without a CWO, organizations may fall into reactive approaches favoring personal over systemic change.

Wellness and the Pandemic

The HERO Registry (NCT04342806), created near the start of the pandemic to assess impact on frontline workers, found that 54% of HCWs were experiencing burnout (Forrest et al., 2021) and that burnout varied by role with greater likelihood among nurses (Galanis et al., 2021), physical therapists, respiratory therapists, and medical assistants. Burnout was associated with intention to leave, particularly among women. Data from HERO (Forrest et al., 2021) and Coping with COVID (Prasad et al., 2021) demonstrated the value of national registries in creating databases to monitor burnout and its consequences.

Breaches in and Promoting Trust

A history of intentional and unintentional harm created by health care toward a variety of groups, along with work conditions driven by health care as a business, has created a legacy of mistrust between communities and organizations as well as HCWs and their organizations. Intentional work on how organizations can repair and redress mistrust has begun (e.g., a U.S. national project led by the Institute for Healthcare Improvement) and shows promise in improving HCW wellness while enhancing trust from communities.

Ensuring Equity in Wellness

Interventions to improve wellness among HCWs have come to reflect the increased focus on equity seen in the wider health care domain. The field is now positioned to create work environments that can identify risk factors among underrepresented groups in medicine, having adapted methodologies meant to determine association between nonmodifiable factors (e.g., age, race/ethnicity, sex/gender), work conditions/culture (e.g., autonomy, leadership, scheduling), and wellness. However, disparities in burnout rates, decreased professional achievement, and varying clinical outcomes (e.g., mental health) persist among women, younger HCWs (especially medical residents), and parents as well as between professions and specialties. New projects to evaluate the efficacy of existing screening tools and a redoubling of institutional commitment to change are being seen in efforts to close these gaps. Insights from the LHS framework are paving the way by (a) unifying efforts of researchers, system leaders, and implementers of interventions and (b) emphasizing the necessity of robust systems to collect and communicate data to key stakeholders

Identifying and Addressing Moral Injury

A notable gap in the research to date is the lack of emphasis on the problem of moral injury. Although the concept of moral distress had been considered in nursing for many years (Hamric, 2012; Lee et al., 2024), the 2019 National Academies of Science, Engineering, and Medicine report identified moral distress as an important antecedent to burnout. Some scholars have asserted that what scholars had been calling burnout was in fact moral injury (Dean et al., 2019). Similar to moral distress, moral injury occurs when clinicians are required to participate in or observe activities that violate their “deeply held moral beliefs” (p. 400). Framing this as moral injury as opposed to burnout puts the onus on a health system, which is considered to have forced clinicians into ethical dilemmas due to myriad competing stakeholders in broken systems. Moral injury came into focus during the pandemic (Cartolovni et al., 2021), and Moral Injury Prevention Programs are piloting novel measures specifying injury components such as nonbeneficial care near the end of life, psychological and physical safety, insufficient resources for patent care, and witnessing outcomes from societal racism; there are also proposals for interventions with psychological support and system restructuring to reduce moral distress and injury (Linzer & Poplau, 2021).

Gaps in Knowledge and Action

There are several gaps that we feel can be rapidly addressed with sufficient national and organizational support. Contributors to burnout are known, along with effective mitigators. It is less clear what will motivate health systems to implement this knowledge, alter systems causing distress, and embrace wellness as a quality metric. The industrial model of care has fallen short by overlooking the intrapersonal, human needs of providers and patients; a system focus on these as desirable outcomes would be a step forward in building systems of empathic care, supported by adequately funded studies to determine high-yield interventions.

Future Research

There is a clear need for longitudinal studies as well as large-scale intervention trials (Tjasink et al., 2023; Linzer et al., 2015) and additional qualitative studies (Sibeoni et al., 2019). Sophisticated study designs such as comparative effectiveness trials, optimization trials (e.g., sequential, multiple assignment, randomized trials), and rapid randomization trials can utilize methodological strengths to promote efficiencies and translatability of findings. These studies require large sample sizes and thus are not likely to be resourced within single health systems. Some have called attention to a need to study primary care (Verhoef & Biomme, 2022) and medical trainees (Zhou et al., 2020). We encourage funders and journal editors to fund and publish rigorous studies of sufficient size to produce actionable findings. Focus group participants emphasized the need to assess favorable impacts of technology, organizational culture, and qualitative studies to better capture HCW lived experiences.

Limitations and Strengths

By studying reviews of reviews, we were able to see many studies, but the focus of individual studies within individual disciplines was lost. In particular, any attempt to evaluate the presence of theory-driven work is limited by the absence of focus on the original studies themselves. Meanwhile, the evolution of themes (the main purpose of a meta-narrative) was difficult to assess given the lack of flow of rigorous data from randomized trials. All systematic reviews are limited by the particular studies retrieved and the level of accuracy in reporting them. A large number of included studies were deemed to be of poor quality, with poorly described interventions and variables. For strengths, EUREKA's numerous leaders in the field were able to comment on themes and offer additional contributions, a main component of a meta-narrative, and these findings were a main driver of our nine bold practice implications below. In addition, we had the benefit of incorporating a deductive thematic analysis of focus groups by national leaders speaking to gaps in the field and best ways forward.

Practice Implications: Nine Bold Proposals for Where the Field Can Go

Based on our review, focus groups, and the EUREKA team's experience, we offer the proposals discussed below.

Identify and Close “Knowing–Doing” Gaps

We know many of the causes of burnout (time pressure, lack of control, chaotic environments, and lack of supportive organizational cultures); the key is often to link wellness and patient outcomes in the mind of organizational leaders, using the literature to support it. We now need the will and international support to implement evidence-based programs that address these with infrastructures (ongoing wellness committees and periodic surveys) and resources (committed leaders with dedicated time and cultural openness to system changes) to support sustainable changes in worker wellness. Some regions may have little awareness and require more urgent attention (Dubale et al., 2019; Ghahramani et al., 2021). We also require thoughtfully designed randomized trials of system interventions, underpinned by solid theorizing, to confirm best practices to prevent burnout. A target of eliminating burnout may be within reach; although distress is a part of medicine, there is nothing inherent in a health system that says this must result in burnout.

Use the LHS Framework to Introduce Burnout Reduction as a System Metric

One mechanism for transforming health systems around wellness is through LHSs. Embedding scholars and leaders (e.g., CWOs) within health systems to oversee wellness structures and make sure data on existing work conditions and longitudinal worker outcomes are incorporated into system change is a natural evolution within LHSs. Projects addressing user interface with the EHR would fit well in this rubric (Craig et al., 2021), although higher quality studies are required. AHRQ has supported, with PCORI, a national LHS network with embedded scholars dedicated to health system evaluation and system transformation. The EUREKA project will create a focus through published wellness competencies (Yilmaz et al., 2023) and the creation of a cadre of scholars named after early AHRQ Director Dr. John Eisenberg. These scholars will participate in studies to effectively integrate evidence-based strategies into health systems to promote HCW wellness. Competencies for these scholars may include the ability to (as in Yilmaz et al., 2023):

• Assess HCW burnout using standard metrics while identifying remediable predictors.

• Elucidate components of burnout and how to adjust work climates to become more ethical, with attention to equity in wellness.

• Employ new methods (e.g., appreciative inquiry, with a focus on positive aspects of work life).

• Structure training programs to emphasize burnout prevention.

• Embed burnout prevention into an LHS to benefit researchers, HCWs, and patients.

Use Theory-Driven Methods to Address Burnout in all HCWs

Only a few of the highest quality articles in this meta-narrative review were theory driven, and even fewer employed a theoretical approach to understand the unique aspects of burnout in multiple HCW roles (see, e.g., Bazargan-Hezani et al., 2021, using the PERMA model of positive psychology). Likewise, theory-driven scholarship is more likely to provide enduring findings. Most studies that had a theoretical basis used JD-R or COR (Hobfoll et al., 2018). Although there are numerous theories to drive studies, these two have been used most frequently and provide varied ways to tie theory to designing interventions. Gain and loss spirals, burnout cascades, and resource pools (Williams et al., 2020) may be helpful in guiding future studies.

Identify and Implement Strategies to Radically Reduce Burnout Among Health Care Leaders

The importance of accessible and supportive leadership has been repeatedly emphasized, but few studies note the profound pressure upon health care leaders, especially in the industrial model where department and clinic leaders are mandated to maintain productivity targets based on quantity of care rather than empathy, satisfaction, and successful care. Industry, regulatory, and financial pressures can be high for leaders, yet the literature has largely failed to address these pressures or what can be done to alleviate them. Leader training in self-empathy (Riess, 2017) has been available but may not often be utilized, whereas the effectiveness of coaching for burnout reduction has been only occasionally studied (Goelz et al., 2020). Health care leaders, often perceived as insulated from frontline stressors, confront a myriad of challenges including staffing shortages, a disengaged workforce, escalating costs, and declining reimbursement rates. Midlevel managers, tasked with executing health system strategies, shoulder many responsibilities yet frequently lack relief from non-value-added tasks. Effective leadership necessitates alleviating burdens to enable leaders to focus on core responsibilities. Establishing leader standard work focused on wellness is essential, as is providing leaders with clear expectations and equipping them with resources to support their teams.

Provide Quickly Implementable Strategies to Improve Recruitment, Retention, and Understanding of Unique Experiences of Diverse HCW and Trainee Groups

Current burnout metrics may not be tapping into the unique health care experience of trainees and workers of color (Lawrence et al., 2022). Concerns around anonymity can create barriers to disclosing one's race or visible identity on burnout surveys, which may lead to lowering of the number of burned out minoritized individuals in samples. More effort needs to be directed at measuring and improving the work lives of underrepresented gender and racial groups, with more comprehensive measures of the work and social environments in which they work and live. By doing this, we can better build a diverse cadre of HCWs with sustainable work lives.

How Do We Quickly Begin to Use the Science We Have?

This relates to opportunities of the LHS: By developing systems that improve organically, based upon scientific findings from within and without, organizations can incorporate new evidence and build more responsive and sustainable systems. Combinations of interventions can be considered (Aryankhesal et al., 2019). Implementation and improvement scientists may hold value in contributing to transforming systems to reduce burnout, improve retention, and build responsive care systems that prevent recurrence of adverse work conditions and continuously improve care. The value of implementation science in furthering burnout reduction should be explored.

There Is No Need to Choose Between Resilience Training and “System Change” to Reduce Burnout

There are concerns that individual approaches around resilience “blame” the individual. Yet, that is simply not the case. Some are effective, although system change may be more effective (Canadas de la Fuente et al., 2018; West et al., 2016). Because system change takes time (to accumulate data and assess impact), individual-focused actions are worth considering for persons so motivated. Several reviews in this meta-narrative summary (e.g., Wang et al., 2023) showed reasonable short-term impact, especially for mindfulness and yoga. The challenge lies in avoiding the perception that personal resilience efforts are a substitute for addressing systemic issues. HCWs rightly resist implications that they lack resilience and often feel that these resilience-focused efforts are aimed at masking deeper organizational flaws. To navigate this complexity, resilience initiatives can be approached with sensitivity and transparency. Leaders must ensure such efforts are not seen as a substitute for meaningful systemic improvements but rather as part of a comprehensive strategy to support HCWs in demanding environments.

Integrate HCW Wellness Departments Into Academic Institutions Devoted to Health of the Public (e.g., Within Schools of Public Health, Medical Schools, or Nursing Schools)

To our knowledge, there are no Departments of Healthcare Workforce Wellness or Workforce Protection in schools of public health or academic health centers. Yet, there are few projects that could reach more people and potentially impact the health of the public more quickly. If humane systems of organizing work environments and standards for health care delivery were enacted globally, millions of patients could be rapidly affected. One reason this transformation seems unattainable is the lack of dedicated programs to oversee improvement and set standards; they are needed due to individuals scattered in numerous departments and settings with an inability to achieve a critical mass to expand the science. Schools of public health, medical schools, and nursing schools would be ideal settings for moving this forward; interdisciplinary efforts around workforce wellness in cross-cutting, interdepartmental centers would be welcome. These units would focus on work life science and implementation of new findings, housing scholars devoted to burnout and moral injury reduction, health policy, and workforce protection.

Devise a Long-Term Plan for the End of Burnout in Postindustrial Health Care

We propose a new vision in which burnout is markedly reduced as care flourishes. Leaders and funders must recognize the core purpose of health systems is to create conditions for care. HCWs and leaders can then be trained to notice the problematic human situation of patients and communities seeking care and ensure that systems that promote safe and reliable action are in place. They also must note that patients present with complex conditions for which no guideline-directed care plan exists; plans must be co-created within unhurried conversations, mandating the creation of environments in which such care can occur. Navigational barriers—physical, cultural, and administrative—must be reduced to lessen disruption in patients' lives. These changes constitute a fundamental change in health care's structure and purpose, a revolution from health care as an industrial process toward health care as an investment in human flourishing (Montori, 2018).

Conclusions

Although the understanding of burnout and HCW well-being is rapidly evolving, improving the structure and focus of this work has an opportunity to meaningfully impact HCW, patient, and organizational outcomes. Many of the thousands of studies to date have been done at single sites on limited budgets and have served to boost visibility of the importance of burnout in HCWs, which in many ways has led to the EUREKA project. Thus, EUREKA calls attention to a lack of funding for rigorous studies, lack of a structured outline for what studies are needed, and relatively few scholarly mentors in this comparatively new area of study. As the EUREKA group continues to review content areas and methods for future studies, as well as ways to involve funders, journals, and new scholars over the next 2 years, we advocate for funding high-quality research, national standards for burnout reduction tied to financial incentives, and support for mentored scholars working toward known research gaps. With a global scientific approach to studying the means to extinguish burnout, we believe meaningful progress can be made.