A Literature Review on Nurse Burnout and Its Impact on Patient Safety

This is a sample literature review written by an AHC subject expert in Nursing to illustrate distinction / first-class standard for a Master’s-level assignment. It synthesises published research; it does not report new primary data. Use it as a model for structure, critical synthesis and Harvard referencing, not as material to submit as your own. It is conceptual and educational and does not constitute clinical guidance.

Introduction

Nurse burnout has moved from a peripheral workforce concern to a central patient safety issue over the past two decades. As the largest professional group in most health systems, and as the staff most continuously present at the bedside, nurses occupy a position in which their psychological state has direct consequences for the reliability of care. Where an exhausted or disengaged nurse misses a deteriorating observation, omits a medication check or communicates poorly at handover, the resulting harm is not confined to the individual clinician but is transmitted to the patient. This review examines what the existing literature reveals about the relationship between nurse burnout and patient safety.

The aim of this review is to synthesise the evidence on how burnout among registered nurses relates to patient safety outcomes, and to appraise the strength and limitations of that evidence. The scope is deliberately focused on burnout as it is conceptualised in the Maslach tradition, on registered nurses working in acute and hospital settings, and on safety outcomes such as clinical errors, adverse events and nurse-rated quality of care. Broader questions of workforce retention and cost, while relevant, are treated only where they intersect with safety. The review does not attempt to adjudicate the wider debate about whether burnout is conceptually distinct from depression, though that debate is noted where it bears on measurement.

The literature discussed here was identified through targeted searching of the major health and psychology databases, principally MEDLINE, CINAHL and PsycINFO, using combinations of the terms “nurse”, “burnout”, “patient safety”, “medical error” and “quality of care”. Priority was given to systematic reviews, meta-analyses and large multi-site studies, supplemented by the seminal conceptual works that define the field. The selection process is summarised in Figure 1, which reports illustrative counts to convey the shape of a structured search rather than to document an exhaustive systematic review. The synthesis that follows is organised thematically rather than chronologically, moving from definition and measurement, through the drivers of burnout, to its consequences for safety and finally the factors that appear to mitigate it.

Defining and Measuring Burnout

Any synthesis of the burnout literature must begin with Maslach and Jackson (1981), whose work established both the dominant conceptualisation of the syndrome and the instrument most commonly used to measure it. They defined burnout as a three-dimensional response to chronic emotional and interpersonal stressors at work, comprising emotional exhaustion, depersonalisation and a reduced sense of personal accomplishment. Emotional exhaustion refers to the depletion of a worker’s emotional resources; depersonalisation to a cynical, detached response towards the people one serves; and reduced personal accomplishment to a decline in feelings of competence and achievement. The Maslach Burnout Inventory (MBI) operationalised these dimensions and has since become the reference standard against which most subsequent measurement is judged.

This tripartite model has proven durable. Maslach, Schaufeli and Leiter (2001) later reaffirmed that burnout is best understood as an individual experience embedded in a social and organisational context, distinguishing it from generic stress by its interpersonal and occupational specificity. The World Health Organization (2019) subsequently classified burnout in the eleventh revision of the International Classification of Diseases as an occupational phenomenon rather than a medical condition, reinforcing the view that its causes are located in the workplace rather than the individual. This framing matters for nursing because it directs attention away from personal resilience deficits and towards the organisation of clinical work; it implies that a nurse who becomes burnt out is responding to conditions, not failing a test of character.

Measurement, however, remains a source of difficulty, and the difficulty is more than technical. Although the MBI dominates, studies vary in how they score and threshold its subscales, and many rely on the emotional exhaustion dimension alone as a proxy for the whole syndrome. There is no universally agreed cut-off that separates the “burnt out” from the “not burnt out”; different research groups apply different thresholds to the same instrument, so that a nurse classified as burnt out in one study might not be in another. This inconsistency complicates comparison across studies and, as later sections argue, weakens the cumulative evidence base. It also means that prevalence figures for nurse burnout should be treated with caution, since apparently similar rates may rest on different operational definitions. A further complication is the conceptual overlap between the emotional exhaustion dimension and depressive symptoms, which some researchers argue inflates observed associations between burnout and negative outcomes because both may be tapping a shared underlying dysphoria. The emergence of alternative instruments, such as those developed within the Copenhagen and Oldenburg traditions, reflects dissatisfaction with aspects of the MBI, but their limited uptake in the nursing safety literature means that the field’s evidence remains anchored to a single family of measures with known limitations.

The practical consequence for this review is that measurement is not a preliminary technicality to be settled before the substantive questions are addressed; it shapes every finding that follows. When a study reports that burnout predicts poorer safety, the strength and even the meaning of that claim depend on which dimension was measured, how it was thresholded, and whether the outcome was reported by the same person who reported the burnout. These questions recur throughout the sections below and are drawn together in the critical appraisal.

Drivers of Burnout in Nursing

The literature is consistent in locating the principal drivers of nurse burnout in the working environment rather than in individual predisposition. Chief among these is workload, and in particular the nurse-to-patient ratio. The influential cross-sectional study by Aiken et al. (2002), drawing on data from over 10,000 nurses and more than 232,000 surgical patients across 168 hospitals in Pennsylvania, found that each additional patient added to a nurse’s average workload was associated with a 23 per cent increase in the odds of burnout and a 15 per cent increase in job dissatisfaction. This study is important not only for its scale but because it linked staffing, burnout and patient outcomes within a single dataset, establishing a chain of association that later work would develop. Subsequent multi-country work by Aiken et al. (2014) in European hospitals reinforced the staffing signal, reporting that increases in nurses’ workloads were associated with higher inpatient mortality, and that a more educated nursing workforce was associated with lower mortality; while that study foregrounded outcomes rather than burnout, it strengthens the wider claim that the conditions producing burnout are the same conditions that endanger patients.

Beyond raw staffing numbers, the quality of the practice environment emerges as a consistent contributor. Features such as poor nurse–physician relationships, limited nurse involvement in decision-making, inadequate managerial support and weak organisational commitment to care quality are all associated with higher burnout. It is useful to read these findings through the lens of the Job Demands–Resources model articulated by Bakker and Demerouti (2007), which proposes that every occupation carries demands that deplete energy and resources that replenish it, and that burnout develops when demands chronically outstrip resources. On this account, nursing is unusually exposed because its demands are simultaneously physical, cognitive and emotional, while the resources that would offset them, such as adequate staffing, autonomy and supportive leadership, are precisely the features most often eroded in stretched health systems. Maslach and Leiter (2016) offer a complementary and more specific mapping, synthesising the practice-environment literature into the concept of a mismatch between the person and six domains of working life: workload, control, reward, community, fairness and values. On this account burnout arises where chronic imbalances accumulate across these domains, offering an explanatory framework that accommodates the varied drivers reported in nursing studies. The strength of both models is that they move the discussion beyond any single variable towards the systemic conditions of nursing work, and they anticipate the corollary developed later: that if burnout is produced by an imbalance of demands and resources, it should be remediable by adjusting either side of that balance.

A particular manifestation of demand that has drawn increasing attention is the emotional labour of nursing, the sustained management of one’s own feelings in order to sustain patients through fear, pain and grief. This form of work is largely invisible in staffing calculations yet contributes materially to emotional exhaustion, which helps explain why interventions that address only the physical workload, without acknowledging the emotional demands of care, tend to disappoint. The conceptual relationships among these drivers, burnout and its safety consequences are represented in Figure 2.

Burnout, Clinical Errors and Patient Outcomes

The core question for this review is whether burnout translates into measurable harm to patients, and here the evidence has grown considerably stronger. The most comprehensive quantitative synthesis is provided by Salyers et al. (2017), whose meta-analysis of 82 studies found statistically significant negative relationships between clinician burnout and both the quality of care (across 63 studies) and patient safety (across 40 studies). Although the pooled effect sizes were modest in magnitude, their consistency across a large and varied body of studies lends them weight, and the direction of association was unambiguous: higher burnout accompanied poorer safety. The modesty of the effect deserves emphasis rather than dismissal. In a domain where outcomes are multiply determined, where an individual error depends on the interaction of the nurse, the team, the equipment and the system, a consistently reproduced small-to-moderate association is precisely what a genuine but non-exclusive causal contribution would look like; burnout is not proposed as the sole cause of unsafe care, only as one reliably implicated factor.

Focusing specifically on nursing, Poghosyan et al. (2010) analysed data from more than 53,000 nurses across six countries and found that higher burnout was associated with lower nurse-rated quality of care, a relationship that held independently of nurses’ assessments of their practice environments. The cross-national consistency of this finding is notable, suggesting that the burnout–quality relationship is not an artefact of a single health system’s culture or reporting conventions. Complementing this, Cimiotti et al. (2012) reported an association between nurse burnout and healthcare-associated infection rates, specifically urinary tract and surgical site infections, and estimated that a reduction in burnout could be accompanied by a meaningful fall in infections. That study is significant because its outcome, infection incidence, is an objective and independently recorded measure rather than a nurse’s self-assessment, and it therefore begins to address the common-method concern examined later in this review. Returning to Aiken et al. (2002), the same staffing pressures that drove burnout were themselves associated with patient mortality, with each additional patient per nurse linked to a seven per cent increase in the likelihood of death within thirty days and a comparable rise in failure-to-rescue. While that study did not establish burnout as the mediating mechanism, it situates burnout within a wider evidential picture in which nursing workload and patient survival are connected.

The systematic review by Hall et al. (2016) offers a valuable qualitative complement to these findings. Examining 46 studies of healthcare staff wellbeing, burnout and patient safety, the authors reported that a majority of the studies measuring wellbeing found a significant association between poorer wellbeing and worse patient safety. Importantly, they highlighted the plausibility of a bidirectional relationship: burnout may degrade safety through mechanisms such as reduced vigilance, impaired working memory and diminished empathy, but involvement in patient safety incidents may in turn deepen burnout. This reciprocal possibility, represented by the dashed pathway in Figure 2 and taken up again below, complicates any simple causal narrative. It also carries a clinically sobering implication: a nurse involved in a serious incident, sometimes described in the literature as a “second victim”, may be pushed further into exhaustion and cynicism, which in turn raises the risk of a further incident, so that burnout and error can compound one another in a self-reinforcing loop unless the cycle is interrupted at the level of the system.

A useful way to understand how burnout might produce error is to consider the specific mechanisms it disturbs. Emotional exhaustion erodes the cognitive resources, attention and working memory on which safe practice depends, so that the exhausted nurse is more likely to lose track of an interrupted task, to fixate on one problem while missing another, or to fail to integrate a new observation into an updated picture of the patient. Depersonalisation blunts the empathic engagement that prompts a nurse to notice subtle changes in a patient’s condition, to sit with a family’s unspoken concern, or to advocate persistently for a patient whose deterioration is not yet obvious in the numbers. Reduced personal accomplishment, the third and least studied dimension, may erode the motivational drive to go beyond the minimum, to double-check, to question a doubtful prescription or to escalate when hierarchy discourages it. The consequence is not usually a single dramatic failure but an accumulation of small omissions, missed observations, delayed responses and abbreviated communication, each individually minor but collectively corrosive of the defences that keep patients safe. This pattern connects the burnout literature to the wider “Swiss cheese” understanding of accident causation in complex systems, in which harm results not from one large failure but from the momentary alignment of many small weaknesses in a system’s layered defences; burnout can be understood as a factor that enlarges the holes in several of those layers at once. Table 1 summarises the principal studies on which this synthesis rests.

Table 1. Summary of key studies informing the review

Mitigating Factors and Interventions

If burnout arises predominantly from the organisation of work, it follows that the most promising interventions should be organisational rather than purely individual. The evidence broadly supports this expectation, though it remains less mature than the descriptive literature. West et al. (2018), in a review of contributors, consequences and solutions to clinician burnout, argued that while individually focused interventions such as mindfulness and stress-management training can produce measurable reductions in burnout, structural and organisational interventions tend to be at least as effective and address the problem closer to its source. Their emphasis on shared responsibility, in which the organisation rather than the individual owns the problem, is consistent with the WHO’s occupational framing and with the Job Demands–Resources logic set out earlier: an intervention that teaches an individual nurse to cope better with an overwhelming workload leaves the demand untouched, whereas one that adds staff, redesigns rosters or protects rest periods reduces the demand itself.

Within nursing specifically, improvements to the practice environment recur as a protective factor. The same features whose absence drives burnout, adequate staffing, supportive leadership, collegial relationships and nurse participation in governance, appear to buffer against it when present, functioning as the job resources that offset demand in the model illustrated in Figure 2. A meta-analysis by Lee et al. (2013) examining structural empowerment and psychological empowerment in nursing found that greater access to information, support, resources and opportunity was associated with lower emotional exhaustion, lending quantitative support to the claim that empowering practice environments are protective rather than merely pleasant. Because Poghosyan et al. (2010) found the burnout–quality relationship persisted independently of the practice environment, however, environmental improvement alone is unlikely to be sufficient; direct attention to workload and to the emotional demands of care also appears necessary. The literature therefore points towards multi-component strategies that combine structural reform with support for individual nurses, rather than reliance on either in isolation. This conclusion should be held with appropriate caution, since rigorous evaluations of organisational interventions remain scarce relative to the abundant descriptive evidence that the problem exists; the field is much better at demonstrating the association than at demonstrating what reliably reverses it.

Critical Appraisal of the Evidence and Gaps

Notwithstanding the consistency of its central finding, the evidence base has significant limitations that temper the confidence with which conclusions can be drawn. The most fundamental is its reliance on cross-sectional designs. The large and frequently cited studies, including Aiken et al. (2002) and Poghosyan et al. (2010), measure burnout and outcomes at a single point in time, which precludes any firm inference about the direction of causation. As Hall et al. (2016) explicitly noted, the relationship may well be bidirectional, and cross-sectional data cannot disentangle whether burnout produces unsafe care, unsafe care produces burnout, or a third factor such as chronic understaffing produces both. The plausibility of the reciprocal path, illustrated in Figure 2, means that even the strongest cross-sectional associations are compatible with several distinct causal stories, and the honest reading of the evidence is that it establishes robust co-occurrence rather than settled causation.

A second limitation concerns the measurement of the outcome variable. A substantial proportion of the nursing literature relies on nurse self-report of quality and safety, whether through single-item ratings or perceptions of error frequency. Salyers et al. (2017) observed that objective, independently verified safety outcomes are measured far less often than subjective ones. Self-reported measures are vulnerable to common-method bias: a nurse experiencing emotional exhaustion and cynicism may rate both her own wellbeing and the quality of her unit’s care more negatively, inflating the apparent association through shared negative affect rather than genuine deterioration in care. This is why studies using objective endpoints, such as the infection rates examined by Cimiotti et al. (2012) or the mortality data in Aiken et al. (2002), are disproportionately valuable: they break the shared-source dependency and show that the association survives when the outcome is recorded independently of the person reporting the burnout. That such studies remain a minority is one of the clearest priorities for the field.

Third, the heterogeneity of burnout measurement discussed earlier undermines the comparability of studies. Where some report the full three-dimensional MBI and others report emotional exhaustion alone, pooling and comparison become problematic, and the modest effect sizes reported in meta-analysis may partly reflect this inconsistency; some of the apparent “noise” in the pooled estimates is really an artefact of instruments and thresholds that are not measuring quite the same thing. The unresolved overlap between emotional exhaustion and depression compounds this, raising the possibility that part of the burnout–safety association is carried by general distress rather than by anything specific to occupational burnout. Finally, the literature is dominated by acute hospital settings in high-income countries, leaving community, mental health, aged-care and low-resource contexts comparatively under-researched, and limiting the generalisability of the evidence to the full range of nursing practice. Given that these under-studied settings often carry the highest workloads and the thinnest resources, it is plausible that the true burden of burnout-related risk is understated by a literature concentrated on comparatively well-resourced acute wards.

These limitations point clearly to the gaps that future research should address. Longitudinal and, where ethical, quasi-experimental designs are needed to clarify the temporal ordering of burnout and safety, ideally following cohorts of nurses across changes in staffing or leadership so that the direction of effect can be observed rather than inferred. Greater use of objective, routinely collected safety data, such as medication error rates, incident reports and infection surveillance, would reduce dependence on self-report. Standardised measurement and reporting of burnout, including agreed thresholds and consistent reporting of all three dimensions, would improve comparability. Most importantly, rigorous evaluation of organisational interventions, rather than repeated confirmation that the association exists, would move the field from a mature descriptive science towards the actionable knowledge that patients and nurses both need.

Conclusion

The literature reviewed here converges on a clear and clinically important conclusion: nurse burnout is consistently associated with poorer patient safety and lower quality of care. This relationship is supported across a large meta-analysis (Salyers et al., 2017), a substantial multi-country nursing study (Poghosyan et al., 2010), a systematic review of staff wellbeing and safety (Hall et al., 2016) and a study using the objective endpoint of infection rates (Cimiotti et al., 2012), and it is underpinned by the well-established conceptualisation of burnout advanced by Maslach and Jackson (1981). The evidence also situates burnout within a wider chain that links nurse staffing to patient survival (Aiken et al., 2002; Aiken et al., 2014), reinforcing the view that it is a systemic rather than an individual problem.

At the same time, the evidence is weakened by its predominantly cross-sectional design, its heavy reliance on self-reported outcomes and its inconsistent measurement of burnout, all of which caution against overly strong causal claims. What can be said with confidence is that the association is robust, that its drivers lie principally in the organisation of nursing work, and that mitigation is therefore most likely to succeed when it reforms that organisation while also supporting individual nurses. For nursing practice and health policy the implication is that protecting patient safety and protecting the wellbeing of nurses are not competing priorities but a single objective pursued from two directions.

References

Aiken, L.H., Clarke, S.P., Sloane, D.M., Sochalski, J. and Silber, J.H. (2002) ‘Hospital nurse staffing and patient mortality, nurse burnout, and job dissatisfaction’, JAMA, 288(16), pp. 1987–1993.

Aiken, L.H., Sloane, D.M., Bruyneel, L., Van den Heede, K., Griffiths, P., Busse, R., Diomidous, M., Kinnunen, J., Kózka, M., Lesaffre, E., McHugh, M.D., Moreno-Casbas, M.T., Rafferty, A.M., Schwendimann, R., Scott, P.A., Tishelman, C., van Achterberg, T. and Sermeus, W. (2014) ‘Nurse staffing and education and hospital mortality in nine European countries: a retrospective observational study’, The Lancet, 383(9931), pp. 1824–1830. [VERIFY page range and author list against the published RN4CAST paper.]

Bakker, A.B. and Demerouti, E. (2007) ‘The Job Demands-Resources model: state of the art’, Journal of Managerial Psychology, 22(3), pp. 309–328. [VERIFY volume and page numbers.]

Cimiotti, J.P., Aiken, L.H., Sloane, D.M. and Wu, E.S. (2012) ‘Nurse staffing, burnout, and health care-associated infection’, American Journal of Infection Control, 40(6), pp. 486–490. [VERIFY exact page range.]

Hall, L.H., Johnson, J., Watt, I., Tsipa, A. and O’Connor, D.B. (2016) ‘Healthcare staff wellbeing, burnout, and patient safety: a systematic review’, PLoS ONE, 11(7), e0159015.

Lee, H., Cummings, G.G. and Wong, C.A. (2013) ‘The relationship between empowerment and burnout in nursing: a systematic review and meta-analysis’, [journal title to be confirmed]. [VERIFY — full citation details, journal, volume and pages require confirmation; treat as illustrative until verified.]

Maslach, C. and Jackson, S.E. (1981) ‘The measurement of experienced burnout’, Journal of Occupational Behavior, 2(2), pp. 99–113.

Maslach, C. and Leiter, M.P. (2016) ‘Understanding the burnout experience: recent research and its implications for psychiatry’, World Psychiatry, 15(2), pp. 103–111.

Maslach, C., Schaufeli, W.B. and Leiter, M.P. (2001) ‘Job burnout’, Annual Review of Psychology, 52, pp. 397–422.

Poghosyan, L., Clarke, S.P., Finlayson, M. and Aiken, L.H. (2010) ‘Nurse burnout and quality of care: cross-national investigation in six countries’, Research in Nursing & Health, 33(4), pp. 288–298.

Salyers, M.P., Bonfils, K.A., Luther, L., Firmin, R.L., White, D.A., Adams, E.L. and Rollins, A.L. (2017) ‘The relationship between professional burnout and quality and safety in healthcare: a meta-analysis’, Journal of General Internal Medicine, 32(4), pp. 475–482.

West, C.P., Dyrbye, L.N. and Shanafelt, T.D. (2018) ‘Physician burnout: contributors, consequences and solutions’, Journal of Internal Medicine, 283(6), pp. 516–529.

World Health Organization (2019) Burn-out an “occupational phenomenon”: International Classification of Diseases. Geneva: World Health Organization. [VERIFY exact WHO reference formatting and access date if a URL is required by the marking rubric.]