Burnout in nursing: what have we learnt and what is still unknown?

Introduction

Several nursing studies focus on burnout as an indicator of adverse work environments but they often overlook what it is, factors that contribute to its development and its effect on nurses, healthcare organisations and patients. This article describes a review of the research examining relationships between burnout and work-related variables. We sought to determine what is known and not known about the causes and consequences of burnout in nursing, and whether these relationships confirm or dispute Maslach’s theory of burnout.

What is burnout?

Referring to fatigue and exhaustion, the term ‘burnout’ is common in everyday language. However, it is unclear whether being tired and feeling demotivated are symptoms of burnout, and whether burnout equates to stress and depression.

Burnout has recently been added to the International Classification of Diseases, 11th Revision as an occupational phenomenon, but it has been researched for the past 45 years (Schaufeli et al, 2009). Maslach was the first to propose a theory about burnout and to measure it as a concept distinct from, for example, stress. She developed the Maslach Burnout Inventory (MBI, Maslach and Jackson, 1981) to measure burnout; this scale is still one of the most widely used internationally. According to her theory, burnout is characterised by:

• Feeling emotionally drained (emotional exhaustion);
• Adverse and cynical detachment from patients, clients and colleagues (depersonalisation);
• A lack of confidence in being able to do one’s job (reduced personal accomplishment) (Maslach, 1999).

She stated that burnout develops when there is a prolonged mismatch between an employee and one or more of the following:

• Workload – for example, having too much work without adequate resources;
• Control – such as not having enough autonomy in how to do one’s job;
• Reward – for example, inadequate pay, poor promotion mechanisms or low recognition of the value of one’s work;
• Community – such as having no sense of community or sense of belonging to a group of colleagues;
• Fairness – for example, unfair processes that mean some groups or individuals are more advantaged than others;
• Values – for example, no mission/vision.

We do not know whether existing studies can help us establish a causal pathway between work characteristics and burnout. Additionally, as burnout is often explored as an endpoint, there is little consideration of what happens to nurses who experience it. As such, we looked for studies to help us understand which factors are associated with burnout in nursing.

Literature review

We undertook a theoretical review (Dall’Ora et al, 2020); this design allowed us to understand the concept of burnout from a theoretical perspective and highlight knowledge gaps, as cited by Pare et al (2015) and Campbell et al (2014). We conducted a search for empirical studies and, due to the volume of those retrieved, synthesised the results by identifying common categories via a coding framework based on the six work areas highlighted in Maslach’s theory.

We identified 91 papers in total; the majority had cross-sectional designs (n= 87) and were survey based (n=84). Most studies took place in hospitals (n=82). The tool most used was the MBI (n=81), but only half of these studies measured all three subscales of burnout as recommended by Maslach (emotional exhaustion, depersonalisation and reduced personal accomplishment).

The studies’ samples ranged between hundreds of hospitals (maximum: 927) with hundreds of thousands of nurses (maximum: 326,750) and a single site with a few nurses (minimum: 73). When assessing the quality of the studies, we found most had limitations, including small sample sizes and a failure to adjust for other variables that may influence the relationship between burnout and the variable under study.

Predictors of burnout

We found a strong association between high workload and burnout. Specifically, we found evidence that high workload is associated with emotional exhaustion, while nurse-staffing levels are associated with all three subscales of burnout.

We found that, when nurses have control over their job and when they experience reward for their efforts, they are less likely to experience burnout. The evidence linking fairness and community to burnout was inconclusive, with only a few studies reporting contrasting results.

Overall, our review highlighted mixed results for the effects on burnout of both working at night and the number of working hours per week. There were more conclusive results regarding the negative association between working long shifts – of at least 12 hours – and emotional exhaustion. We found some evidence that staff who were satisfied with their schedule flexibility were less likely to report emotional exhaustion.

The literature indicated that high job and psychological demands, and role conflict, were related to emotional exhaustion. High patient complexity predicted burnout, while task variety, autonomy and the ability to make important decisions appeared to be protective of burnout.

There was also strong evidence that positive working relationships and having support from colleagues and managers might play a protective role against burnout. In particular, this was confirmed to be the case for positive relationships with physicians, support from the team leader, positive leadership style and teamwork.

In summary, when nurses worked in positive work environments, they were less likely to experience emotional exhaustion. However, none of the organisational characteristics at the hospital level – such as hospital type or Magnet accreditation – was consistently associated with burnout.

Consequences of burnout

There was strong evidence that burnout predicted nurses’ intentions to leave their jobs, but this did not translate into actual staff turnover. Studies reported associations between some subscales of burnout and low job performance, sickness absence, poor general health, missed patient care and job dissatisfaction.

However, the relationship with job dissatisfaction and missed care was observed in multiple directions so it is unclear whether the presumed cause (burnout) precedes the effect (job dissatisfaction or missed care) or vice versa.

Burnout was linked to reduced patient safety and adverse events, including medication errors, infections and falls. When staff experienced burnout, patient dissatisfaction and family complaints increased.

Conclusions of the review

The 91 studies we reviewed and synthesised enabled us to identify which adverse job characteristics are associated with burnout in nursing. It is evident from the available literature that the potential consequences of burnout for staff and patients are severe. We found that the relationships posited by Maslach’s theory were observed in several studies, leading us to conclude that the theory is valid and remains relevant.

We also found that the field has been dominated by cross-sectional studies, which is problematic because temporality cannot be established: it is not possible to discern, for example, whether job dissatisfaction drives burnout or vice versa.

Most studies were also limited by the use of incorrectly applied burnout measures, for example, using only one of the three MBI subscales; the emotional exhaustion subscale was frequently used in isolation, with no theoretical justification for doing so. The statistical models were also often unable to control for important variables, which, if done, might have changed the nature of the association.

As a result of these numerous sources of bias, we cannot reliably identify the causes and consequences of burnout. This makes it difficult – if not impossible – to use the evidence to design interventions to reduce burnout.

Despite the uncertainties, the evidence clearly does not support interventions to reduce burnout that are targeted at individual behaviours – such as mindfulness or resilience training – but, rather, at those that aim to fix mismatches in the work environment. To help address this, we propose three research actions:

• Apply longitudinal designs that follow nurses over time to understand which factors contribute to burnout;
• If using Maslach’s theory, report associations for all three MBI subscales;
• Prioritise the use of empirical data on employee behaviours to study the consequences of burnout (such as absenteeism or staff turnover), rather than the use of self-report intentions.

• This article is based on: Dall’Ora C, Saville C (eds) (2020) Burnout in nursing: what have we learnt and what do we still need to know? Evidence Brief; 16 (University of Southampton, October 2020). Reproduced under the terms of the CC-BY 4.0 licence (creativecommons.org/licenses/by/4.0). The authors received no specific funding for carrying out the review.