The COVID-19 pandemic presented one of the greatest challenges faced by already stretched healthcare services world-wide. Healthcare workers (HCWs) had to adapt to unfamiliar roles and environments with new schedules and pressures, while constantly needing to rapidly upskill themselves in order to provide clinical care in an area where the scientific knowledge was and still is, constantly evolving (Gavin et al, 2020; Wu et al, 2020). Added to this, they were contending with the biopsychosocial consequences of the COVID-19 pandemic in the form of social distancing, lockdown or quarantine (Gavin et al, 2020; Wu et al, 2020).
The additional stress of a pandemic has the potential to detrimentally impact the health of HCWs, as was associated with the 2003 SARS pandemic (Maunder, 2004; Marjanovic, 2007; Lancee et al, 2008). Burnout is a debilitating condition, which is predictive of long-term sickness (Peterson et al, 2011) and individuals leaving the profession (Rudman et al, 2014) and has an impact on quality of patient care (Kohnavard et al, 2015; Khodaveisi et al, 2017; Turner et al, 2017), with high burnout levels often being linked with compassion fatigue (Green and Walkey, 1988; Bakker et al, 2004; McKinley et al, 2020).
HCWs' resilience has been shown to play a part in their likelihood of developing burnout, with resilient individuals handling occupational and other stresses well, in addition to being better at adapting to changes around them Arrogante, 2014; Sull et al, 2015). It is, therefore, not surprising that interventions based on resilience training have been shown to have a significant effect in reducing burnout levels in HCWs (Montero-Marin et al, 2015; Sull et al, 2015).
Methods
The authors invited health professionals to complete an electronic survey, an outline of which can be found in Box 1. The survey was created using Microsoft Forms, and was distributed to HCWs internationally through social media and via healthcare colleges and associations, which advertised the survey link on their websites or emailed it to their members. Data collection took place from 16 April 2020 to 18 May 2020. This study was compliant with ethical standards, ethical approval was not required. Informed consent was provided to each participant before they engaged in the survey. Participants were made aware that data would be potentially used anonymously in publication.
Box 1.Outline of survey
- Demographic information: gender, age, country
- Professional information: role, specialty and subspecialty
- Role/duties changed due to pandemic?
- Work schedule changed due to pandemic?
- Frequency of exercise (at least 30 minutes/day)
- Relationship status
- Previously received resilience training? Format?
- Years of experience, private/public sector?
- Oldenburg Burnout Inventory questions
- Brief Resilience Scale questions
- Connor-Davidson Resilience Scale questions
Oldenburg Burnout Inventory
This survey used the Oldenburg Burnout Inventory, a well validated tool for measuring burnout, which measures the two main components of burnout – ‘exhaustion’ and ‘disengagement’ from work (Sinval et al, 2019). Exhaustion can be defined as the consequence of intense physical, affective, and cognitive strain caused by prolonged exposure to certain demands of a job, whereas disengagement refers to one's distancing from one's work environment (including work object and content) (Rudman and Gustavsson, 2012; Khodaveisi et al, 2017). The inventory is composed of 16 negatively and positively framed questions to assess exhaustion and disengagement. The participants are asked to respond to these questions using a scale from 1 (strongly agree) to 4 (strongly disagree). Average scores of ≥2.1 in the disengagement domain and ≥2.25 in the exhaustion domain are used as cut-offs, with those who are both disengaged and exhausted being classed as having burnout.
Brief Resilience Scale
This is a well-validated 6-item scale that measures an individual's perception of their ability to ‘bounce back’ and recover from stress (Smith et al, 2008; Windle et al, 2011; Gavin et al, 2020). For this survey, it was used to measure general resilience. It is scored on a 5-point scale, with the participants being asked to respond to questions using a scale from 1 (strongly disagree) to 5 (strongly agree). Total scores range from 6 (minimum resilience) to 30 (maximum resilience). This is then averaged to give a score, which is interpreted as:
- Low resilience: 1.00–1.99
- Normal resilience: 2.00–4.30
- High resilience: 4.31–5.00.
Connor-Davidson Resilience Scale
The 10-item Connor-Davidson Resilience Scale specifically measures an individual's assessment of their resilience over the previous month, thus, it was used here to measure resilience during the COVID-19 pandemic as the pandemic had been ongoing for more than 30 days at the time participants completed the survey. It is scored on a 4-point scale, with responses from 0 (not true at all) to 4 (true nearly all the time) (Connor and Davidson, 2003; Campbell-Sills and Stein, 2007; Arrogante, 2014; Sull et al, 2015). Total scores range from 0 (minimum resilience) to 40 (maximum resilience). It is also well validated with multiple reference scores in varying populations. A cut-off value of 24 was applied as this has been found to predict more than moderate post-traumatic stress disorder (PTSD) (Sull et al, 2015; Shin et al, 2018).
The relationship between the variables was examined via univariate and multivariate logistic regression analysis. Statistical significance was defined at two-tailed P<0.05. Statistical analysis was carried out using R software, version 1.2.5033 (RStudio, Boston). Sample size estimation using a 95% confidence interval, 90% power, and a standard deviation of 0.5 estimated an adequate sample size of 194.
Results
A total of 1370 individuals completed the survey, 164 were male and 1206 female. The main age range of respondents was from the category 25-34 years, with 32.6%. Considering exercise, 40.0% of respondents exercised 2–3 times a week, whereas 14.2% never exercised. In terms of experience and sector, 11.3 % had 0-2 years of experience in health care and 32.1% had more than 16 years of experience; 73% of respondents worked in the public sector, 15% in private and the remaining in both. During the pandemic, 71% of respondent's roles changed and 73% said their schedules changed; 15.8% said they had resilience training.
Respondents were mainly from the UK (n=702, 51.2%) and Ireland (n=199, 14.5%), with other core groups from Zimbabwe (n=175, 12.8%), Australia (n=148, 10.8%), South Africa (n=72, 5.3%) and the USA (n=51, 3.7%).
Primary outcomes
The primary outcomes are illustrated in Table 1. Burnout was present in 802 respondents (58.5%). For general resilience scores, 942 (68.8%) respondents had normal general resilience, 80 (5.8%) had high general resilience, and the remainder, 348 (25.4%), had low general resilience. In terms of COVID resilience, 390 (28.5%) were not resilient, whereas 980 (71.5%) were resilient.
Table 1. Breakdown of primary outcomes
| Number | Percentages | ||
|---|---|---|---|
| Total respondents | 1370 | ||
| Burnout | Burnout | 802 | 58.5% |
| Not burnout | 568 | 41.5% | |
| Exhaustion | Exhausted | 1124 | 82% |
| Not exhausted | 246 | 18% | |
| Engagement | Engaged | 425 | 21% |
| Disengaged | 945 | 69% | |
| COVID resilience | Resilient | 980 | 71.5% |
| Not resilient | 390 | 28.5% | |
| General resilience | Low resilience | 348 | 25.4% |
| Normal resilience | 942 | 68.8% | |
| High resilience | 80 | 5.8% | |
Univariate analysis
The results of univariate analysis are demonstrated in Table 2 and Table 3
Table 2. Univariate analysis Fisher's Exact Test odds ratios and 95% confidence intervals for key outcomes – binary factors
| Odds ratio (95% CI) | P value | ||
|---|---|---|---|
| Outcome = Burnout | Female gender | 1.48 (1.05–2.08) | 0.022* |
| Role change | 0.83 (0.63–1.08) | 0.166 | |
| Schedule change | 0.90 (0.69–1.18) | 0.458 | |
| Resilience training – not done | 1.38 (1.02–1.87) | 0.035* | |
| Outcome = Exhaustion | Female gender | 2.03 (1.36–2.98) | <0.001* |
| Role change | 0.89 (0.64–1.26) | 0.500 | |
| Schedule change | 1.30 (0.91–1.90) | 0.144 | |
| Resilience training – not done | 1.42 (0.98 – 2.05) | 0.05 | |
| Outcome = Disengagement | Female gender | 1.48 (1.05–2.08) | 0.590 |
| Role change | 0.98 (0.74–1.31) | 0.943 | |
| Schedule change | 0.88 (0.66–1.18) | 0.387 | |
| Resilience training – not done | 1.38 (1.00–1.89) | 0.045* | |
| Outcome = Low COVID resilience | Female gender | 1.06 (0.73–1.56) | 0.783 |
| Role change | 0.89 (0.66–1.20) | 0.468 | |
| Schedule change | 0.82 (0.60–1.11) | 0.387 | |
| Resilience training – not done | 1.16 (0.83–1.65) | 0.411 | |
| Outcome = Low general resilience | Female gender | 1.15 (0.77–1.73) | 0.566 |
| Role change | 0.93 (0.68–1.27) | 0.650 | |
| Schedule change | 1.02 (0.74–1.39) | 0.938 | |
| Resilience training – not done | 1.60 (1.10–2.38) | 0.011* |
Table 3. Univariate analysis Fisher's Exact Test odds ratios and 95% confidence intervals for key outcomes
| Odds ratio (95% CI) | P value | |
|---|---|---|
| Low COVID resilience and burnout | 3.42 (2.59–4.54) | <0.001* |
| Low general resilienceA and burnout | 3.39 (2.54–4.57) | <0.001* |
| Low general resilienceN and burnout | 3.11 (2.32–4.20) | <0.001* |
| Low general resilienceH and burnout | 0.10 (0.05–0.18) | <0.001* |
| Low COVID resilience and exhaustion | 5.23 (3.27–8.77) | <0.001* |
| Low general resilienceA and exhaustion | 4.94 (3.02–8.50) | <0.001* |
| Low general resilienceN and exhaustion | 4.31 (2.63–7.45) | <0.001* |
| Low general resilienceH and exhaustion | 0.06 (0.03–0.12) | <0.001* |
| Low COVID resilience and disengagement | 3.34 (2.45–4.62) | <0.001* |
| Low general resilienceA and disengagement | 2.73 (1.99–3.78) | <0.001* |
| Low general resilienceN and disengagement | 2.46 (1.79–3.42) | <0.001* |
| Low general resilienceH and disengagement | 0.12 (0.07–0.20) | <0.001* |
| Low general resilienceA and COVID resilience | 10.07 (7.57–13.44) | <0.001* |
| Low general resilienceN and COVID resilience | 9.13 (6.86–12.21) | <0.001* |
| Low general resilienceH and COVID resilience | 0 (0.00–0.026) | <0.001* |
denoting low resilience vs normal resilience + high resilience
Ndenoting low resilience vs normal resilience
Hdenoting low resilience vs high resilience
Burnout
On univariate analysis, burnout was significantly associated with female gender (OR 1.48, 95% CI 1.05–2.08; P=0.022), low general resilience (OR 3.39, 95% CI 2.54–4.57; P <0.001), low COVID resilience (OR 3.42, 95% CI 2.59–4.54; P<0.001) and lack of prior resilience training (OR 1.38, 95% CI 1.02–1.87; P=0.035).
Resilience
Low COVID resilience was significantly associated with having low general resilience (OR 10.07, 95% CI 7.57–13.44; P<0.001) and lack of prior resilience training (OR 1.60, 95% CI 1.10–2.38; P=0.011).
Multivariate analysis
The results of multivariate analysis are demonstrated in Table 4 and Table 5
Table 4. Multivariate analysis odds ratios and 95% confidence intervals for burnout
| Multivariate analysis | |||
|---|---|---|---|
| Odds ratio (95% CI) | P value | ||
| Outcome = Not burnout | Age – 55–64 | 2.72 (1.03–7.49) | 0.045* |
| COVID resilience – Resilient | 2.36 (1.75–3.20) | <0.001* | |
| Outcome = Burnout | Country – Ireland | 2.01 (1.08–3.76) | 0.028* |
| Country – South Africa | 2.88 (1.35–6.39) | 0.007* | |
| Exercise – Once a month | 2.98 (1.60–5.87) | <0.001* | |
| Relationship – Widowed | 5.60 (1.42–26.50) | 0.018* | |
| General resilience – Low resilience | 2.10 (1.53–2.89) | <0.001* | |
| Outcome = Not exhausted | Country – Zimbabwe | 1.99 (1.07–3.78) | 0.032* |
| Experience – 12–15 years | 3.03 (1.29–7.51) | 0.013* | |
| COVID resilience – Resilient | 3.26 (2.03–5.48) | <0.002* | |
| Outcome = Exhausted | Exercise – Never exercise | 2.23 (1.12–4.14) | 0.007* |
| Exercise – Once a month | 5.00 (1.43–17.97) | 0.004* | |
| Exercise – Once a week | 2.02 (1.24–3.25) | 0.007* | |
| General resilience – Low resilience | 2.75 (1.67–4.74) | <0.001* | |
| Outcome = Not disengaged | Age – 65–74 | 10.93 (1.39–116.69) | 0.029* |
| Role – Counsellor | 6.19 (1.33–33.53) | 0.025* | |
| COVID resilience – Resilient | 2.56 (1.83–3.61) | <0.001* | |
| Outcome = Disengaged | Country – South Africa | 3.38 (1.46–8.47) | 0.006* |
| Exercise – Once a month | 2.21 (1.14–4.59) | 0.024* | |
| General resilience – Low resilience | 1.64 (1.17–2.33) | 0.005* | |
Table 5. Multivariate analysis odds ratios and 95% confidence intervals for resilience
| Multivariate analysis | |||
|---|---|---|---|
| Odds ratio (95% CI) | P value | ||
| Outcome = Not COVID resilient | Country – Zimbabwe | 2.45 (1.33–4.61) | 0.005* |
| Country – UK | 2.11 (1.26–3.64) | 0.004* | |
| Exercise – Once a month | 1.69 (1.11–3.32) | 0.029* | |
| Exercise – Never | 1.88 (1.05–2.57) | 0.014* | |
| Outcome = COVID resilient | Exhaustion – Not exhausted | 2.28 (1.31–4.12) | 0.005* |
| Engagement – Not disengaged | 1.99 (1.17–3.45) | 0.015* | |
| General resilience – Resilient | 8.03 (6.02–10.78) | <0.001* | |
| Outcome = Not generally resilient | Relationship status – Widowed | 5.69 (1.52–22.58) | 0.010* |
| Outcome = Generally resilient | Resilience training – Done | 1.79 (1.17–2.80) | 0.009* |
| Burnout – Not burnout | 2.28 (1.23–4.44) | 0.011* | |
| COVID resilience – Resilient | 8.01 (6.00–10.75) | <0.001 | |
Burnout
Burnout was associated with working in South Africa (OR 2.88, 95% CI 1.35–6.39; P=0.007) or Ireland (OR 2.01, 95% CI 1.08–3.76; P=0.028), being widowed (OR 5.60, 95% CI 1.42 -26.50; P=0.018), exercising once a month (OR 2.98, 95% CI 1.60-5.87; P<0.001) and low general resilience (OR 2.10, 95% CI 1.53-2.89; P<0.001).
Resilience
Lower resilience was associated with being based in Zimbabwe (OR 2.45, 95% CI 1.33–4.61; P=0.005) and the UK (OR 2.11, 95% CI 1.26–3.64; P=0.004). Lower resilience was also linked to never exercising (OR 1.88, 95% CI 1.05–2.57; P=0.014).
Discussion
This study is one of the first to demonstrate a strong relationship between resilience and burnout among HCWs during the COVID-19 pandemic. With respect to the primary outcomes, a large majority of participants were suffering from an element of burnout (exhaustion 81%, disengagement 71.5% and burnout 58.5%).
McCain et al (2018), in a smaller single-centre Irish study, reported similar burnout rates in doctors despite high levels of resilience, during non-pandemic times. Recent, larger studies have shown significantly lower (McKinley et al, 2020) or simply, moderate (Sull et al, 2015) resilience levels in doctors as well as high burnout rates of up to one-third of their participants (McKinley et al, 2020). Burnout rates in the present study are significantly higher than those published in the literature – 58.5% vs 22% (Linzer et al, 2001) and 33% (McKinley et al, 2020) – a difference which is likely explained by the COVID-19 pandemic but may also be explained by the fact that this study considered all HCWs and not only doctors. Nevertheless, the burnout rate for doctors alone in the present study was, notably, still 57%. The effect of the COVID-19 pandemic on burnout may also be reflected by the higher percentage of HCWs who had low COVID resilience (28.5%) compared with general resilience (25.4%); suggesting that some HCWs were less resilient during the period and may have experienced burnout due to the pandemic. This is particularly important when considering the relationship between resilience and PTSD, ie, the ability to adapt successfully in the face of stress and adversity being protective against PTSD secondary to acute or chronic stressors, which includes the COVID-19 pandemic in addition to general work-related stress (Ha et al, 2014; Streb et al, 2014; Horn et al, 2016). However, studies carried out following the SARS pandemic did not report higher rates of PTSD among HCWs (Maunder, 2004; Marjanovic, 2007; Lancee et al, 2008; Maunder et al, 2008), with some arguing that psychological preparedness through exposure, in this case already dealing with medical emergencies regularly, was protective (Streb et al, 2014).
Burnout is associated with sleep deprivation and medical errors. A broken system can be masked whereby high priority clinical tasks are maintained yet secondary tasks, such as reassuring patients, are neglected. This can lead the public to believe that HCWs do not care about the emotional wellbeing of patients. It can therefore be inferred that burnout should be used as a systematic tool indicator for healthcare quality (Hockey, 1997).
The work-life model focuses on six areas in which the job–person match is critical: workload, control, reward, community, fairness, and values. Burnout arises from workload problems combined with mismatches across the five other areas (Montgomery et al, 2019). The extent of mismatch in the five remaining categories impacts on the individual's level of experienced burnout.
Solutions targeted at an individual level, rather than those that are systematically rooted, will lack longevity and sustainability; they suggest the belief that this failure is personal to the individual rather than the organisation as a whole (Montgomery et al, 2019). Burnout is best confronted as a shared responsibility with interventions at an organisational and individual level, such as increased support for clinical work and mindfulness, which have demonstrated benefits in reducing burnout (West et al, 2018).
Low general resilience levels correlate with low resilience levels during the COVID-19 pandemic. This supports the hypothesis that improving HCWs' resilience could help with reducing burnout within the healthcare workforce. Unfortunately, the authors could find no studies in the literature looking at training with respect to burnout.
Workshops and small-group problem-solving have been demonstrated to have resilience-building effects (Rogers, 2016). Strategies that promote characteristics of personal resilience such as optimism, flexibility, and tolerance (Matheson et al, 2016), as well as dealing with workplace challenges such as workload and challenging patients (Matheson et al, 2016), are likely to succeed at building HCWs' resilience.
Blended learning using ‘SMART’ strategies (specific, measurable, agreed upon, realistic, and timely) in reducing stress and burnout in nurses has been found to significantly reduce anxiety, stress and burnout while increasing resilience, happiness, and mindfulness (Magtibay et al, 2017). A resilience-building intervention is more beneficial for those with a low resilience baseline to begin with, over a longer period (Goldhagen et al, 2015).
Building resilience may help an individual adapt and deal with stress encountered at work but does not address the stress the workplace exerts on HCWs. Healthcare organisations should ensure that their governance and management systems maximise the participation of clinical staff in setting priorities and solving problems and effectively aim to eliminate stressors such as staff shortages, and working overtime, as well as offering personalised resilience training (Card, 2018).
McCain et al (2018) found that non-clinical issues such as complaints, litigation and lack of senior-level support were the main factors leading to low resilience. Maunder et al (2006) looked at the long-term effects of the SARS pandemic in Toronto, demonstrating the immediate impact of dealing with infected patients, changes in working schedules or roles did not necessarily have a significant effect on the long-term distress of HCWs; what did have an impact was the perceived adequacy of training, moral support and protection provided by the organisation before and after the pandemic. The present study supports this as HCWs' changes in roles and schedules, although experienced by 71% and 73% of participants, respectively, did not have a significant effect on resilience or burnout. This suggests that there are wider organisational stressors that are also important to consider.
HCWs in the public sector were more likely to have low general resilience and low resilience during the COVID-19 outbreak as well as suffer from exhaustion and burnout. For example, Lim and Pinto (2009) found that radiologists working in the public sector in New Zealand reported significantly higher levels of work stress, lower job satisfaction and higher rates of burnout compared with those working in the private sector. These findings were closely related to their work environments, ie, the chronic staff shortages, increased workload, multiple commitments, and lack of autonomy faced by those working in the public sector. This contrasts with the almost complete autonomy over the general work environment and workflow efficiency enjoyed by those in the private sector.
Raftopoulos et al (2012) found a similar relationship when looking at emotional exhaustion and burnout of nurses working in Cyprus and argued that the increased workload and understaffing present in the public sector may be to blame. Remarkably, Pavlakis et al (2010) found that physiotherapists in the private sector suffered from higher rates of emotional exhaustion and burnout in Cyprus. Raftopoulos et al (2012), who published their research 2 years later, had also hypothesised higher burnout rates for nurses in the private sector. They argued that the recent changes made to the healthcare system in Cyprus had improved working conditions in the private sector, brought equilibrium between the two sectors and offered an explanation for the variability in outcomes. A study carried out in Bangladesh found that organisational support was the strongest predictor adversely affecting job satisfaction, turnover and burnout in doctors in both public and private sectors (Roy et al, 2017). The difference noted in the Cypriot studies may also be owing to the two studies looking at different roles, ie, physiotherapists and nurses. Our study has shown that, even in the same healthcare systems and countries, a respondent's role was strongly related to experiencing burnout, exhaustion, and disengagement. Indeed, on multivariate analysis, counsellors were found to be six times less likely to be disengaged.
There may also be a role played by the differences between societies and cultures. On multivariate analysis, individuals based in Ireland and South Africa were twice as likely to experience burnout, with respondents from the latter being three times as likely to be disengaged. Similarly, respondents based in the UK and Zimbabwe were twice as likely to have low resilience during the COVID-19 pandemic period, whereas, in contrast, respondents based in Zimbabwe were less likely to experience exhaustion. It does, however, provide some insight into the potential role of larger systematic factors on these outcomes.
Exercise has been shown to be protective against burnout (Gerber et al, 2013; Wolf and Rosenstock, 2017). Gerber et al (2013) looked at the effects of a 12-week exercise programme on individuals suffering from burnout and found that exercise reduced overall perceived stress as well as symptoms of burnout and depression. Similarly, Wolf and Rosenstock (2017) found that decreased exercise frequency was significantly correlated with lower professional efficacy in medical students. General resilience and resilience during the COVID-19 outbreak were unaffected by exercise, suggesting that exercise does not affect burnout by increasing resilience – highlighting the importance of not relying only on resilience training as a method of tackling burnout.
Individuals in relationships are more likely to be emotionally and psychologically supported and therefore, be more resistant to burnout (Wade et al, 2013). Single respondents are more likely to have low general resilience and suffer from exhaustion and burnout than those in a relationship. Studies have shown higher levels of exhaustion, and cynicism, in unmarried vs married individuals (Ahola et al, 2006; Cañadas-De la Fuente et al, 2018).
Females are more likely to develop exhaustion than males (Ahola et al, 2006; Pavlakis et al, 2010; Purvanova and Muros, 2010), which was reflected in our results, with female burnout and exhaustion rates being significantly higher than those of males. Other studies have shown no difference in overall burnout between the genders (Carlson et al, 2003; Purvanova and Muros, 2010), with some demonstrating a higher propensity of either exhaustion or depersonalisation in females and males respectively (Purvanova and Muros, 2010); making it difficult to draw conclusions. Perhaps the fact that males are less likely to engage with burnout research, representing only 12% of our participants, a trend that has been observed in other studies (McKinley et al, 2020), plays a role in this discrepancy.
Multivariate models demonstrated that being a ‘resident/fellow’ was associated with increased odds of burnout and being a medical student with increased odds of depressive symptoms. Compared with the population control samples, medical students and residents/fellows were more likely to be burned out (all P<0.0001). At each stage, burnout and symptoms of depression were more prevalent among medical students and doctors than among their peers pursuing other careers (both P<0.0001). Training appears to be the peak time for distress among physicians, further reinforcing the argument that tools for building resilience should be incorporated much earlier on in an individual's career (Dyrbye et al, 2014).
Experience had a significant effect on burnout, exhaustion, and disengagement, with greater experience being protective against the three. Similarly, on univariate analysis, age was significantly associated with burnout, exhaustion, and disengagement, with respondents aged 18–25 years being most affected across all three outcomes. Correspondingly, multivariate analysis revealed the 55–64 and 65–74 age groups to be the least affected by burnout and disengagement. These findings are in concordance with previous studies, which showed a negative correlation between age and burnout (Brewer and Shapard, 2004; Randall, 2007). Less work has been published looking at the independent relationship between experience and burnout, partly because many studies consider it to be analogous to age. This may lead to incorrect conclusions, given that, for example, Randall (2007) found that although age was protective against burnout, the number of years a clergyman had been in the ministry did not have an effect. With that said, we did find increasing experience and age to both be protective of burnout. It therefore appears that HCWs, with time and experience, managed to develop skills important in the prevention of exhaustion, disengagement and burnout (Ahola et al, 2006). This does not, however, change the fact that many participants did suffer from exhaustion, disengagement and burnout in all age and experience groups. Moreover, with no or at most very weak correlations between age and experience with resilience (Sull et al, 2015) or mindset, one could argue that the coping mechanisms that a worker develops over time are:
- Not developed by all individuals
- Irrelevant in the protection of younger HCWs
- Insufficient in building good levels of resilience and in preventing exhaustion, disengagement, and burnout.
Limitations
As a self-reported survey-based study, there is potential for reporting bias, though this is mitigated by using validated psychological tools. This study was conducted at the start of the COVID-19 pandemic. Although the number of respondents is high, the study is still open to non-response bias given first, the use of social media, and second, the fact that many HCWs would have been occupied with dealing with the COVID-19 pandemic. Moreover, there is potential for selection bias, given the relatively low proportion of male respondents (12%) and the predominance of UK-based (702) participants compared with other countries.
Conclusion
Low resilience levels are associated with an increased likelihood of developing exhaustion, disengagement, and burnout. The burnout rate of HCWs was higher during the COVID-19 pandemic than previously reported in the literature, with more HCWs reporting low resilience during the pandemic than they normally would. This suggests that the COVID-19 pandemic has influenced both resilience and burnout levels of HCWs. There is additional complex interplay between personal and systemic variables that directly impact HCW resilience and burnout. The role of organisational and systemic factors must not be underestimated. Early personalised resilience training, together with better institutional support, could have a significant positive impact in reducing burnout in HCWs, to address this ever-growing issue.
KEY POINTS
- Low resilience levels are associated with an increased likelihood of developing exhaustion, disengagement, and burnout
- The COVID-19 pandemic has influenced both resilience and burnout levels of healthcare workers
- The role of organisational and systemic factors must not be underestimated
- Experience is a protective factor against burnout, exhaustion, and disengagement
- It is essential to recognise the need for early personalised resilience training in healthcare workers, to address the ever-growing issue of burnout
CPD reflective questions
- Stress is an integral part of working in health care, but when does ‘stressed’ become ‘burned out’? What are the red flags?
- Consider the factors that increase resilience. What might help you and members of your team to build up resilience?
- What else could the nursing community do to reduce workplace stress and burnout?