Burnout in Surgical Trainees: a Narrative Review of Trends, Contributors, Consequences and Possible Interventions

Studies have linked higher burnout in surgeons with their overall attitude towards their training programme, reflected in poorer ratings on programme satisfaction [37] and their likelihood of recommending their programme to others [38]. Links have also been made with specific aspects of training programmes. A recent study in US plastic surgery residents found that higher burnout was not linked with age, gender, race or relationship status but was linked with feeling less involved in programme-related decisions [38]. Similarly, a survey of US general surgery residents found no links between burnout and gender or relationship status, but did report that burnout was higher in residents reporting fewer programmatic social events and lower in residents receiving programme mentorship [30]. Links between receiving mentorship and reporting lower burnout have also been reported in other studies of surgical trainees [32, 39, 40]. However, there is no evidence to suggest that burnout varies consistently according to training stage, with studies showing mixed and variable findings in burnout rates across training years [32, 36, 39, 41].

Higher burnout has been linked with greater number of working hours overall [38, 40, 42, 43] and working more weekends per month [30]. One study in 665 US general surgery residents found that 69% of respondents were burnt out, and those reporting high burnout worked on average 3 h more per week than those reporting low burnout [40]. Associations between workload factors and burnout have not replicated consistently, with some findings indicating no association between number of work hours or the delivery of night shifts and burnout in trainees [44, 45], and a study from Saudi Arabia reporting no significant association between number of on-calls per month and emotional exhaustion in plastic surgery residents [46].

Burnout has been found to be lower when surgery trainees report the presence of more work support factors. For example, in plastic surgery residents in France, lower risk of burnout was associated with the occurrence of regular staff meetings and weekly ward rounds by senior surgeons [44]. In a study of US general surgery residents, lower risk of burnout was linked with autonomy at work, feedback, perceived social support and opportunities for development [41]. Conversely, factors which could generate a sense of being unsupported have been associated with higher risk of burnout. In a study of digestive surgery trainees in France, it was found that perceiving a lack of gratitude from senior staff, an insufficient amount of practical training and an overly high level of responsibility were all associated with a greater risk of reporting burnout [47].

Colleague factors are also important to burnout. Studies have found that burnout is lower in surgery trainees when they report being friends with their co-workers [48] and when they feel a sense of social belonging in their specialty [49]. Conversely, burnout is higher when trainees report negative colleague interactions, such as poor interactions with nurses [50], being stressed by relationships at work [32] or being shamed and humiliated at work, for example by being banished from the operating room or being called ‘stupid’ [51].

Gender is frequently tested in relation to burnout in consultant and attending surgeons, with an overall conclusion that burnout risk is higher in women [52]. However, the picture is less clear in surgical trainees. While some studies have reported that burnout overall [41, 43] or the emotional exhaustion facet in particular is higher in women [40, 49], this finding has not always replicated. Several studies have reported no association between gender and either emotional exhaustion [53] or overall burnout [27, 30, 38] and other studies have highlighted that burnout overall [45] or the depersonalisation facet is in fact higher in men [40, 53]. As undergraduate medical degrees are increasingly taking higher proportions of female students [54] but surgical training remains a less popular choice among this group [4], understanding links between gender and burnout and making improvements based on this will be important for the sustainability of surgical disciplines in the future.

Of all the studies which have investigated associations between burnout and personality characteristics in surgeons, the majority of these have been in trainee groups [35]. These have suggested that higher emotional intelligence [41, 55, 56] agreeableness, conscientiousness and emotional stability [41], grit, resilience [57, 58] and dispositional mindfulness [53] are associated with lower burnout. However, the investigation of personality in surgeons and surgical trainees is contentious; the usefulness of such knowledge is unclear and some commentators have highlighted its potential for misleading and deterring prospective surgical trainee candidates [59].

A range of studies have investigated family factors but findings have been mixed. While being single has been identified as a risk factor for burnout in attending and consultant surgeons [35] and this has occasionally replicated in studies of surgical trainees [27, 50], several studies have reported no association between relationship status and burnout in trainee groups [38, 40, 44, 60]. Furthermore, having children was found to be associated with lower levels of burnout in a study of surgical residents in Pakistan [45] but this has not always replicated [40, 44]. Similar with studies into personality variables, it could be suggested that understanding links between family factors and burnout has limited use and could potentially stigmatise individuals based on characteristics which are not related to their work and which they may have limited latitude to change.

Research clearly indicates the influence of the COVID-19 pandemic on surgeon burnout: in a UK survey, 86% of consultant and trainee surgeons reported negative effects related to the pandemic, including emotional exhaustion [11]. In the Netherlands, surgeons assigned to a COVID ward reported high burnout symptoms [61]. In India, 1 in every 2 orthopaedic surgeons reported experiencing either mild, moderate or severe anxiety in May 2020 [62]. However, further research is needed which focuses on surgical trainees; this group may be particularly vulnerable to burnout given the pandemic’s impact on their training and education [11].

Broader reviews of surgeons have identified burnout as an associate of numerous detrimental personal consequences, including a higher risk of reporting a mental health disorder and poorer physical quality of life [35, 63]. In surgical trainees in particular, burnout is associated with higher levels of depression [27, 53, 64, 65], anxiety [53], PTSD [66] and suicidality [53]. One study reported no association between burnout and suicidality in US general surgery trainees, but the sample size was relatively small (n = 92) and, given the rarity of suicidal thinking, may be attributable to lack of statistical power [64].

Burnout has also been associated with unhealthy behaviours in surgical trainees such as alcohol misuse [32] and engaging less frequently in exercise [32, 46, 64], although these findings have not always replicated [44, 48]. While it could be suggested that burnout is a contributory factor to these symptoms and behaviours, most studies have been cross-sectional and so it is not possible to establish causality. Instead, it may be more useful and accurate to view burnout as a warning sign which could indicate that a wider pattern of poor mental health and unhelpful behaviours could be present. This is supported by studies linking higher burnout to poorer self-reported general health [3], physical health quality of life [32, 48] and happiness levels [48].

The first systematic review investigating associations between burnout and patient safety in healthcare professionals was published in 2016 [67]. Across the 30 studies which were included, 21 (64%) reported a significant association between burnout and patient safety outcomes [67]. Since this time, this research field has flourished with numerous further original studies being reported. This growth has enabled the subsequent publishing of reviews focused on specific professional groups such as physicians and primary healthcare providers [68, 69]. It has also allowed for a wider range of patient care outcomes to be synthesised including provider communication and quality of care indicators [68‐70]. In India, higher burnout in doctors has been associated with poorer communication with patients, including being more likely to shout at patients [34].

No systematic review has investigated the association between burnout and patient care outcomes in surgeons or surgical trainees specifically, but results from a limited number of original studies in surgical trainees reflect findings in the wider literature. For example, in US plastic surgery residents, significant associations were found between higher burnout and having made an error that could have resulted in patient harm or a lab error [38]. There was a trend towards burnout being associated with medication errors but no association was found between burnout and having made an error that did result in patient harm [15]. In OTS residents in France [27] and in US abdominal transplant surgery fellows, higher burnout was associated with a greater risk of having made a medical error in the past 3 months [48]. In a study of US general surgery residents, it was found that overall burnout and each of the emotional exhaustion and depersonalisation facets were significantly associated with a higher self-reported risk of having made a medical error that resulted in harm, and making a near-miss that did not result in harm [71].

In terms of surgeons and medical error, one area of research that may be particularly important is the relationship between burnout and cognitive impairments [72]. For example, two studies observed significant associations between burnout and cognitive decline only in complex cognitive tasks whereas performance on less complex tasks was within the normal range [73, 74]. Interestingly, burnout is associated with poorer performance on cognitive tasks tapping attention and visuo-spatial constructional ability [75]. Additionally, a study revealed that healthcare professionals (doctors, nurses, psychologists) with stress-related exhaustion take longer to switch their attention between tasks [76]. Moreover, research with internal medicine residents and faculty members show that high depersonalisation scores were associated with a decreased blood oxygenation level dependent (BOLD) effect in the right dorsolateral prefrontal cortex (dlPFC) and middle frontal gyrus, while high exhaustion scores were associated with more right posterior cingulate cortex and middle frontal gyrus BOLD, three brain regions that are associated with executive functions, memory and attention, respectively [77]. Error monitoring impairments, as indicated by event-related potential (ERP) patterns, have also been observed among burnout populations [78]. Disrupted ERP patterns indicative of a high cognitive workload have also been observed among surgical residents [79]. Overall, the impact of burnout on cognitive functioning most relevant to surgical practice (i.e. attention, executive functions and visuo-spatial abilities) is likely to have negative impacts on patient safety.

Together, these findings indicate that surgical resident burnout is likely to be associated with patient care outcomes. However, these have been cross-sectional and mainly conducted in the USA; further research is needed to establish the direction of causality between burnout and patient care outcomes, to examine whether this is present in other countries and to investigate a wider range of outcomes. This literature has also been criticised for an over-reliance on self-reported rather than objectively measured errors [35].

It can be concluded from the existing literature that interventions, overall, are effective for reducing burnout in physicians [82, 88], but effect sizes are small and there is a need to improve these. We are not aware of any systematic review that investigates burnout reduction interventions in surgeons or surgical trainees in particular, but in a 2017 review of burnout reduction interventions in residents across specialties, a total of 19 original studies were identified including 6 randomised controlled trials and 13 cohort studies [89]. Of the included studies, 3 were in surgical specialties and 1 was in obstetrics and gynaecology residents, and 9 focused upon the 2003 and 2011 Accreditation Council for Graduate Medical Education duty hour restrictions [89]. Overall, the review found that the duty hour restrictions were associated with reductions in emotional exhaustion and depersonalisation. They also reported limited evidence supporting meditation and self-care workshop interventions [89].

A small number of original interventional studies have been published since the publication of this review which aimed to reduce burnout in surgical trainees and measured this as an outcome. In a controlled study of 21 US surgery residents, mindfulness meditation training was associated with post-intervention improvements in perceived stress and mindfulness in the active group (n = 12) compared with the control group (n = 9) but no differences were observed for burnout [90]. However, it must be noted that the sample size for this study was small and null findings could be attributed to a lack of statistical power. In a cohort study of 8 Canadian otolaryngology-head and neck surgery residents, a formal mentorship training programme was associated with reductions in emotional exhaustion and depersonalisation at 12-month follow-up, but as there was no control group, it is possible that this could be attributed to the effects of time [91].

There is a clear need for more and higher quality studies testing burnout interventions in surgeons and surgical trainees. Most studies to date have been cross-sectional, had a small sample size or have not included a control sample, which prevents firm conclusions being drawn regarding intervention effectiveness. There has also been a preponderance of research from the USA, and there is a need for studies which are conducted in other nations, particularly in low- and middle-income countries (LMIC). Despite this, several promising avenues for future research can be identified. Findings from studies of risk factors related to surgical trainee burnout suggest that changes made to training programmes could be useful targets for reducing burnout and could be manageable to implement, such as the introduction of formal mentorship [30]. Changes to work conditions, such as improving working hours, have also been reported to have tangible benefits for burnout [89]. Such interventions can be distributed widely to all trainees prophylactically, which could reduce the risk of trainees choosing not to engage due to the stigma of reporting burnout.

The intense and demanding nature of surgical work promotes high levels of identification within departments and units—creating the phenomenon of small organisations (i.e. surgical units) within the bigger one (i.e. hospital). One approach that is particularly suited to comparing across units is the areas of worklife model, which specifies six areas in which the job-person match is critical: workload, control, reward, community, fairness and values [92]. It a useful framework for identifying possible ways to improve the job environment by enabling focus on problems that people care about and are willing to work on, until there is an improvement—which then can cross-pollinate to other areas. For example, organisations where staff have set their own working patterns (affecting the areas of fairness and control) have shown improved recruitment and staff satisfaction [93]. An intervention approach that focuses on ‘what can be done’ is possibly more likely to fit with the ethos and mindset of pragmatic surgery staff.

Furthermore, it is possible to draw on the wider literature when identifying potentially useful burnout interventions. For example, there has been a growing awareness of the contribution of technology advancements and associated technology-related frustrations to increasing healthcare professional burnout [94]. However, a recent systematic review of 81 studies in physicians found that interventions which optimise technologies, such as by providing training, reducing the time spent on electronic documentation or improving workflow processes were associated with burnout reductions [94]. These interventions can be simple to identify, design and deliver, and could become increasingly beneficial in a time where medical and surgical interventions are becoming more and more digitised.

Another interesting intervention to consider is yoga [95]. An increasing evidence base supports a link between physical health and burnout, but in a systematic review of 11 articles, yoga was associated with lower levels of stress and burnout in healthcare workers [95]. However, it should be noted that only seven of the studies included in the review were controlled trials, and there is a need for further, higher-quality evidence to gather more evidence and identify which forms of yoga may be most acceptable and effective [95].