Article Text

Download PDFPDF

Grit and burnout in UK doctors: a cross-sectional study across specialties and stages of training
  1. Laura Halliday1,
  2. Abigail Walker2,
  3. Stella Vig3,
  4. John Hines4,
  5. John Brecknell4
  1. 1Department of General Surgery, St George's University Hospital, London, UK
  2. 2Department of ENT Surgery, St George's University Hospital, London, UK
  3. 3Department of Vascular Surgery, Croydon University Hospital, London, UK
  4. 4London Specialty School for Surgery, Health Education South London, London, United Kingdom
  1. Correspondence to Dr Laura Halliday, Department of General Surgery, St George's Hospital, Blackshaw Road, London SW17 0QT, UK; laura.halliday{at}doctors.org.uk

Abstract

Purpose of the study Grit is characterised by the ability to persevere during difficulties and maintain a sustained effort over an extended period of time. Throughout their careers, doctors will experience many periods of stress and difficulty. This may result in burnout, defined by the presence of exhaustion and disengagement from work. This study aims to characterise the relationship between grit and burnout in doctors and to establish whether there are differences between specialties and levels of training.

Study design A multicentre cross-sectional survey by questionnaire was used. Participants were recruited from training days and an online medical forum. The survey consisted of the Short Grit Scale and the Oldenburg Burnout Inventory, which examine levels of grit and burnout, respectively.

Results 548 responses were collected. We found a weak negative correlation between grit and burnout in UK doctors (r=−0.243, p<0.001). Hospital consultants had significantly higher grit scores than trainees. The highest level of burnout was found among general practitioners (GPs). When GPs were analysed separately, the correlation between grit and resilience was not seen.

Conclusions An understanding of an individual's level of grit may be used to identify doctors at a greater risk of burnout. As a high level of grit is associated with less burnout, interventions to improve grit through resilience training should be examined. Further research is needed to understand how grit levels change during a doctor's career and why GPs experience higher levels of burnout.

  • grit
  • resilience
  • MEDICAL EDUCATION & TRAINING
  • burnout
View Full Text

Statistics from Altmetric.com

Purpose of the study

Defined as ‘perseverance and passion towards long-term goals’, the concept of ‘grit’ is characterised by the ability to persevere during difficulties, having self-motivation to achieve success and maintaining a sustained effort over an extended period of time.1 ,2 An association between grit and achievement has been reported in different populations: in adolescents, grit is correlated with high school academic performance and in the US military cadets it correlates with training retention.2

Resilience, defined by the ability to respond positively and thrive in response adverse situations,3 is an important component of grit. The Short Grit Scale (SGS) has been developed and tested as a measure of grit.2 Half of the questions assess resilience and half relate to persistence and long-term commitment.

A number of factors contribute to a person's resilience: optimism, self-efficacy, impulse control, perseverance, flexibility and emotional awareness.4 ,5 The Penn Resilience Programme proposes that building self-belief and self-efficacy makes an individual more likely to seize opportunities and overcome setbacks.6 They have demonstrated that resilience can be taught to students and this has been adapted into a programme for military recruits in the USA.6 ,7

Burnout is characterised by exhaustion and disengagement from work.8 It has been well documented in both the medical and lay press that burnout among doctors is high and causes a significant burden to their well-being. Up to 28% of practicing doctors report symptoms consistent with a psychiatric illness and the suicide rate in doctors is up to twice that of the general population.9–11 More than 8 in 10 doctors report moderate or high levels of burnout.8 ,12 A General Medical Council report published in 2014 called for resilience training to be provided to medical students to enable them to deal with stresses during their career.13

Several small studies have previously looked at grit and resilience in doctors. Resilience negatively correlated with burnout in general practice trainees and nursing students14 ,15 and low levels of resilience correlated with distress in medical students.16 Research into grit in doctors is in its infancy but a study from 2014 demonstrated correlation between grit and burnout in surgical trainees in America.12 However, there is little published data on levels of grit and resilience among UK doctors.

Throughout their careers doctors will go through many periods of stress and difficulty. The factors that contribute to a doctor's ability to support themselves and sustain well-being through these times are poorly understood. Not everyone who is subjected to long-standing work-related stress will develop burnout. Individual and personal factors may determine how a person responds and thus whether they are likely to experience burnout.

This study first seeks to determine the nature of the relationship between grit and burnout among UK doctors in a large and varied population and second to establish whether there are any differences in grit and burnout between specialties and stages of training.

Study design

Participants

A multicentre cross-sectional survey by questionnaire was used. Participants were recruited from regional postgraduate training days, events at a large teaching hospital and an online medical forum open to all qualified doctors.

Survey design

The survey consisted of two main sections: the SGS and the Oldenburg Burnout Inventory (OLBI). The SGS comprises both positive (high grit) and negative (low grit) phrased questions, scored from 1 to 5, with reverse coding for negative questions (listed in online supplementary appendix 1). It has been studied and validated in a range of populations.2

OLBI assesses two separate dimensions of burnout: exhaustion and disengagement. It features 16 questions in total, with both positive (high burnout) and negative (low burnout) phrasing within each dimension (listed in online supplementary appendix 2). Exhaustion is seen as ‘physical, affective and cognitive strain’ due to prolonged exposure to occupational demands, while disengagement is defined as ‘distancing oneself from one's work’.17 The questions have been validated for measuring burnout.17 High scores on the grit and burnout scales represent high levels of grit and burnout, respectively.

Anonymous demographic details were recorded including age, gender, specialty and grade to allow comparisons between demographic groups. Respondents were subdivided into groups based on their level of training or specialty: consultant, qualified general practitioners (GPs), higher specialist trainees (ST3 and above), core/junior specialist trainees (CT/ST 1–2) and foundation doctors. Foundation training forms the first 2 years of postgraduate experience, with doctors experiencing a wide range of general medical, surgical and community rotations. Following on from foundation is core/junior specialty training (CT/ST), which represents the first 2 years of specialist training (described as junior specialist training in our study). Doctors then complete specialist training programmes which may take up to 6 years, before being recognised as GPs or consultants.

Institutional approval was obtained from the London Postgraduate School of Surgery. Prior to recruitment events, the participants were sent an information leaflet detailing the aims of this project. Participation was voluntary. The same survey was posted in an online research forum for UK doctors. It was open for completion for 60 days, and all responses from doctors practicing in the UK were included in the analysis.

Statistical analysis

SGS and OLBI scores were examined to determine whether a normal distribution was present using Kolmogorov-Smirnov tests, and consequently correlation between grit and burnout was assessed using Pearson correlation tests. Differences in grit and burnout were calculated using an independent sample t-test for differences in gender and an analysis of variance (ANOVA) for comparison between specialty and level of training. When appropriate post hoc tests were conducted using Tukey's comparison to establish the differences within an ANOVA. A significance level of p<0.05 was used throughout. Statistical analysis was performed using SPSS V.22 (IBM, New York, USA).

Results

Demographics

Five hundred and ninety-four responses were collected. Forty-six responses were excluded due to incomplete completion of the questionnaire, giving 548 responses for analysis. Table 1 shows the demographic characteristics of the respondents. As the number of responses from foundation doctors was considerably less than other groups, they were not analysed as a subgroup.

Table 1

Demographic details of respondents

Correlation between grit and burnout

In our study population, the average SGS grit score was 3.7, with an average OLBI burnout score of 36.5. Table 2 lists the average scores for each subgroup.

Table 2

Mean scores for grit and burnout among each subgroup

Overall, there was weak negative correlation between grit and burnout as shown in figure 1 (r=−0.312, p=0.0001), demonstrating that high grit scores were associated (although weakly) with low burnout scores. This negative correlation was also seen when separately analysing grit and disengagement (r=−0.226, p=0.001) and grit and exhaustion scores (r=−0.381, p=0.0001), demonstrating that high grit scores were associated with both low disengagement and exhaustion scores. Table 3 lists the correlation coefficients for each subgroup. When GPs were analysed separately, there was no significant correlation between grit and burnout.

Table 3

Strength of correlation between grit and burnout in each subgroup. (r=correlation coefficient)

Figure 1

Scatter plot of overall burnout score against grit score (r=−0.312, p=0.0001).

Differences in grit

The grit scores for each subgroup are shown in figure 2. ANOVA showed that grit scores were significantly different between subgroups (F (3, 308)=4.617, p=0.004). As shown in table 4, consultants report significantly higher levels of grit than GPs or trainees. We found a weak positive correlation between age and grit score (r=0.116, p=0.01). After adjusting for age, the difference in grit between subgroups is reduced but remains statistically significant (mean difference 0.11, p=0.007).

Table 4

Post hoc comparisons of grit scores between subgroups, indicating the average difference in mean grit scores between the two groups listed

Figure 2

Box and whisker plots of the grit scores for each subgroup. (The box shows the median and IQR. The whiskers display the minimum and maximum scores in each group). GP, general practitioner.

Differences in burnout

Burnout scores for each subgroup are shown in figure 3. ANOVA showed that burnout scores were significantly different between the subgroups (F (3, 311)=11.946, p=0.0001). As shown in tables 57, GPs had significantly higher levels of overall burnout, exhaustion and disengagement than any other subgroup. There was no significant correlation between burnout and age.

Table 5

Post hoc comparisons of burnout scores between subgroups, indicating the average difference in mean burnout scores between the two groups listed

Table 6

Post hoc comparisons of disengagement scores between subgroups, indicating the average difference in mean disengagement scores between the two groups listed

Table 7

Post hoc comparisons of exhaustion scores between subgroups, indicating the average difference in mean exhaustion scores between the two groups listed

Figure 3

Box and whisker plots of the burnout scores for each subgroup. (The box shows the median and IQR. The whiskers display the minimum and maximum scores in each group).

Discussion

This study has shown first that high levels of grit are associated with lower levels of burnout in UK doctors; second that consultants have higher levels of grit than trainees; and third that there are comparatively high levels of burnout in qualified GPs.

Previous research in this field found that surgical residents in the USA with below average levels of grit were more likely to contemplate withdrawal from training.18 A study of 128 Australian GP registrars found that high levels of resilience are associated with less burnout, anxiety and stress.14 A longitudinal study of surgical trainees in America found that those with high grit scores reported significantly less burnout and improved well-being after 6 months when compared with those with lower grit at baseline.12

Our finding of weak negative correlation between grit and burnout is in keeping with previous research.12 We found stronger correlation between grit and exhaustion than grit and disengagement. Previous studies have found that emotional exhaustion has the strongest association with stressors among doctors.19 ,20 Emotional exhaustion is an early symptom of burnout, which leads to depersonalisation and higher burnout scores.21 Improving grit may have a role in targeting and reducing the early stages of burnout and preventing the serious complications that severe burnout has both on the doctor themselves and the care they provide.

Previous studies have found median grit scores for doctors of 3.63–4.07,12 ,18 in keeping with our findings. This is consistently higher than the median grit score of 3.4 for the general population.18 Consultants in our study had significantly higher grit scores than trainees. It is unclear whether this association is due to people with lower levels of grit leaving their profession, the acquisition of resilience during training or a natural process with increasing age. It is likely that the relationship between grit and burnout is not unidirectional. Having greater job satisfaction, and thus lower burnout levels, may lead to greater perseverance and higher grit levels through positive reinforcement. Studies in non-medical populations have found a positive correlation between age and grit.1 ,2 This may reflect a natural process of stabilising personal interests and increasing psychological maturity over time.2

While GPs had lower grit scores than consultants, their results were comparable to the other subgroups. The GPs in our study were on average younger than the consultants (41.7 and 45.5 years, respectively, p<0.001), and as shown in the regression analysis, age partly explains the difference in grit levels. However, the difference remains after controlling for age. Other potential explanations for lower grit scores in GPs compared with consultants include the effect of the longer total training time for secondary care compared with primary care physicians which may allow the development of more resilience, or different personality traits in those attracted to hospital or general practice.

Studies of non-medical populations using the OLBI have reported disengagement scores of 17.7 to 19.4 and engagement scores of 17.4 to 19.7.17 ,22 ,23 GPs in our study had comparatively high levels of overall burnout, disengagement and exhaustion scores. A cross-sectional study of burnout in American doctors found that general practice was in the top three specialties for burnout.24 There are likely to be different factors affecting GPs that contribute to high levels of burnout that are not included in this study. When GPs are analysed separately, we found that the correlation between grit and burnout was not present. It is interesting that the correlation was present in all groups except GPs, who had the highest levels of burnout. In a study of surgeons in the USA, no correlation between grit and burnout was found in subanalysis of general surgery trainees, who have the highest levels of attrition from training in their population.12 We suggest that burnout in GPs is complex and multifactorial; when burnout levels are very high, the protective role that grit can play may be lost.

An understanding of an individual's levels of grit may also be used to allow doctors to identify if they are at a high risk of burnout. Teaching resilience techniques may have many benefits for doctors, both professionally and personally. Several other strategies to reduce burnout in doctors have been trialled. Trainees in the USA perceived little benefit from a faculty-led mentoring programme in reducing burnout.19 Stress management programmes have been investigated but the effect on burnout levels was only temporary and frequent training sessions were needed for the benefit to last more than 6 months.25 A mindfulness-based exercise has also been studied in American residents but no significant benefit was seen in levels of stress or burnout.26 However, a pilot study of resilience training for doctors has shown that a single session results in a significant improvement in stress, anxiety and quality-of-life measures.27 Resilience training may form a part of preventative action to help doctors before they develop serious difficulties. Much more research is needed but improving grit and resilience may be an important mediator to other factors that cause burnout.

Future research

It remains to be seen whether grit can predict success in medical training. In other populations, grit has been shown to predict achievement.6 ,7 A longitudinal study of grit and burnout in doctors is required to have a better understanding of how these factors change during an individual's career and how grit scores relate to achievement in medicine.

A larger study of GPs is needed to elicit the reasons for high burnout in this group and explore whether the ‘protective’ role of grit is lost when burnout is high.

Strengths and limitations

This is the largest study to date of grit among practising doctors. Much of the previous work into grit and burnout is based on American data, and this study provides an important insight into UK practice as different career paths and working environments will produce different patterns of burnout. We have included GPs and hospital specialists, different levels of training and different working environments within one study, which provides good external validity for our findings.

Fifty-seven per cent of our respondents came from an online survey. As this was advertised to members of an online social community, they may not be representative of the overall population of UK doctors. The answers given may be affected by time-specific factors, such as proximity of examinations or annual leave. Respondents may also provide answers that they perceive as desirable traits for a doctor and thus giving high grit scores. To build on the findings of our exploratory study, a more representative sample could be obtained in future work by using national training groups and specialist associations.

Multiple statistical tests have been performed on our dataset, which increases the risk of type I error. Post hoc tests used Tukey's comparison to reduce the risk of this although it is not eliminated. While the differences described have shown statistical significance, it is unclear whether these differences have a practical significance. As further research is carried out using these scales, the significance of such differences will become clearer.

Conclusion

This study has found a weak negative correlation between grit and burnout among UK doctors. We have found that consultants have significantly higher levels of grit than doctors in training or GPs. Further research is required to understand how an individual's grit levels change through their career.

As high levels of grit are associated with less burnout in doctors, interventions to improve grit through resilience training should also be examined. GPs reported the highest levels of burnout among UK doctors in our study and when they were analysed separately the correlation between grit and burnout was no longer seen. We propose that burnout in GPs is likely to be complex and multifactorial; when burnout levels are very high the ‘protective role’ that grit can play could be lost.

Main messages

  • Grit has a weak negative correlation with burnout across a wide population of UK doctors.

  • Consultants display higher levels of grit than trainees or general practitioners (GPs).

  • Qualified GPs report higher levels of burnout than trainees or hospital consultants.

  • Further investigation is needed to establish if the nature of burnout in GPs is different to hospital colleagues and why the relationship between grit and burnout is not seen in this group.

Current research question

  • Further investigation is needed to establish if the nature of burnout in GPs is different to hospital colleagues and why the relationship between grit and burnout is not seen in this group.

Acknowledgments

Evangelia Demerouti is thanked for permitting the use of the Oldenburg Burnout Inventory in this study.

References

View Abstract

Footnotes

  • Twitter Follow Abigail Walker at @abiwalker24

  • Contributors All authors were involved in designing this study, data analysis and manuscript preparation.

  • Competing interests None.

  • Ethics approval London Postgraduate School of Surgery.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles