Introduction
Methods
Results
Study characteristics
Outcome analysis
Discussion
Limitations
Conclusion

INTRODUCTION

The term “burnout,” introduced by Freudenberger in 1974, refers to job-related dissatisfaction primarily caused by work-related stress.
¹
The most widely validated tool for measuring burnout among physicians is the Maslach Burnout Inventory-Human Services Survey (MBI-HSS) 22-item tool.
²
The MBI-HSS measures burnout in three subcomponents: emotional exhaustion (EE); depersonalization (DP); and personal accomplishment (PA).
²
Burnout is suggested by a high score in EE and DP, and a low score on PA.
²

There is a palpable gender gap in academic emergency medicine (EM) where female emergency physicians (EP) are less likely to hold major leadership positions, more likely to spend a greater percentage of time in clinical and teaching activities, publish less in peer-reviewed journals, and are less likely to achieve senior academic ranks in their medical schools.
³
Even after adjusting for factors such as race, region, rank, years of experience, clinical hours, core faculty status, administrative roles, board certification, and fellowship training, the mean (±SD) salary of women was found to be $19,418 (±$3,736) less than that of men (P < 0.001).
⁴
This gender disparity can negatively impact the retention of female EPs and predispose them to higher burnout.

Although there are systematic reviews that have described burnout among EPs, none have focused on the gender gap in burnout among EPs.
<sup>5

–

7</sup>
Therefore, the Female Leadership Academy for Medical Excellence (FLAME) members of the World Academic Council of Emergency Medicine performed a systematic review to describe the gender disparity in burnout among EPs at a global level. To the best of our knowledge, this is the first systematic review focusing on gender disparity in burnout among EPs as measured by the validated MBI tool.

METHODS

We performed a systematic review following the PRISMA methods
⁸
using the protocol published in PROSPERO (CRD42024558794).

Search Strategy

We searched two open access databases, PubMed and Epistemonikos on June 30, 2024, for peer -reviewed articles on burnout and emergency physicians. We operationalized different permutations of each keyword as follows:

Burnout: “Maslach burnout inventory” OR MBI OR burnout OR burn-out OR “burned out” OR depersonalization OR “emotional exhaustion” OR “compassion fatigue”

Emergency Physician: “emergency physician*” OR “emergency doctor*” OR “EM physician*” OR “EM doctor*” OR “emergency resident*” OR “EM resident*” OR “emergency consultant*” OR “EM consultant*” OR “emergency faculty*” OR “EM faculty*” OR “emergency professor*” OR “EM professor*” OR “emergency attending*” OR “EM attending*”

We applied the field “All fields” for searching on PubMed and “Title and Abstracts” for searching the same combination of keywords in Epistemonikos.

Screening and Eligibility

We applied a series of inclusion and exclusion criteria. Articles were included if they were 1) written in English, 2) published in a peer-reviewed journal, 3) original articles, and 4) applied any version of the MBI-HSS to measure burnout. They were excluded if they 1) did not describe the results separately by sex, 2) did not include EPs in their study, or 3) were a systematic review.

Extraction and Analysis

Extraction was performed by two investigators independently. The following information was extracted: study characteristics (first author, year of publication, country, number of participants that responded); characteristics of participants (mean age, proportion of males and females); and outcome data (proportion of high burnout in males and females, proportion of males and females with high EE, high DP and low PA, mean scores in males and females for EE, DP or PA and odds of burnout or its subcomponents—EE, EP, or PA—in female EPs.

Study Quality

We used the Joanna Briggs Institute’s critical appraisal checklist for evaluation of the quality of the prevalence studies.
⁹
The tool assessed quality using nine questions. A score of 1 was assigned for a “Yes” as an answer, and a score of 0 was assigned for an answer that was “No,” “Unclear,” or “Not Applicable.” The scores were graded as low, moderate or high if the total score was ≤ 4, 5–7, and ≥ 8, respectively. The quality assessment was performed independently by two investigators, and any disagreement was settled by discussion.

RESULTS

Literature Search

Our initial search resulted in 331 articles in PubMed and 13 in Epistemonikos, which were imported into EndNote reference management software (Clarivate Analytics, Philadelphia, PA). Of these 344 articles, nine were found to be duplicates, leaving a total of 335 articles for the screening and eligibility stages (Figure). Of the 335 articles screened, we excluded 226 that did not meet the inclusion criteria, leaving us with 109 articles for retrieval. We reviewed these 109 full texts for eligibility, resulting in the exclusion of the following: •

3 articles that were systematic reviews

•

7 articles that were not peer-reviewed original articles

•

5 articles that did not include emergency physicians

•

41 articles that did not use the Maslach Burnout Inventory for measuring burnout

•

35 articles that did not report their data by sex,

This left a total of 18 articles for the final review. The process of screening and selecting studies is shown in the PRISMA flow diagram (Figure).

STUDY CHARACTERISTICS

We included 18 studies from 10 different countries in the final analysis (Table 1). The total number of EPs studied in these 18 studies was 26,939, including 8,864 (33%) female EPs, resulting in a male-to-female ratio of 2:1. Fifteen of these studies used the 22-item MBI-HSS tool for measuring burnout, while two studies used the two-item tool based on the MBI-HSS, and one included the nine-item MBI tool (Table 1). All were multicentric studies except for one, which was a single-center study. Included were EPs of both sexes in all studies except for one, which included only female EPs. All included studies were conducted in the last 10 years (2014 to present), except for one study conducted in 1996. Six studies were scored as high quality, while the remaining 12 were moderate quality studies. The response rates varied from 30–94.1%.

OUTCOME ANALYSIS

Burnout calculated by Maslach Burnout Inventory

Seven studies reported the burnout proportions separately in male and female EPs (Table 2). The studies that included the 22-item MBI-HSS tool had a total of 1,181 male and 542 female EPs, with an average pooled proportion of high burnout at 58.7% and 58.3%, respectively (P = 0.8). Two studies used the two-item MBI, including a total of 4,868 male and 2,736 female EPs, with an average pooled proportion of high burnout at 27% and 34%, respectively (P < 0.001).

Subcomponents of MBI

Seven studies reported individual components of the MBI-HSS, specifically the proportion of participants with high EE, high DP, and low PA separately for male and female EPs (Tables 3 and 4). Four of these studies reported the proportion of participants with subcomponents suggestive of burnout (Table 3), and three of these studies reported the mean and standard deviations of the MBI subcomponents (Table 4). Female EPs had proportionately higher EE and lower PA than male EPs (P < 0.001). The DP levels were similar among male and female EPs (Table 3). The combined mean EE score was higher in female EPs (P < 0.001), while the combined PA score was higher and the DP score was lower in female EPs compared to male EPs (P < 0.001) (Table 4). Four studies presented their results in the form of odds or relative risk (Table 5), and three of these showed higher odds among females of burnout while one had lower odds. Three of these studies reported only one subcomponent of MBI-HSS (ie, EE) (Table 5).

DISCUSSION

The stressful environment of EM is a known contributor to the negative impacts of burnout.
¹⁰
Burnout can be the result of good-intentioned physicians who strive for perfection at work.
¹⁰
Emergency physicians become frustrated when their work environment falls short of supporting well-meaning goals, leading to EP burnout.
¹⁰

Moral injury is now recognized as a significant factor contributing to burnout among EPs. They often face challenging decisions such as prioritizing care in life-or-death situations, dealing with resource limitations, and frequently witnessing suffering and death. The emotional toll of moral injury can lead to symptoms of depression, anxiety and stress disorders, which are closely linked to burnout. Female EPs, in addition, face sex-based discrimination, bias, unequal treatment, and fewer opportunities for career advancement.
³
They are more burdened with balancing professional and family responsibilities. Females are expected to display more empathy and provide emotional support to patients and colleagues, which can increase emotional labor and moral injury if they are unable to meet these expectations.

The MBI assesses the severity of the three primary symptoms of burnout: exhaustion; depersonalization; and lack of personal accomplishment. Developed in the 1970s, it has become the gold standard for measuring burnout across various professions and industries. The MBI-HSS is specifically designed for use in human services professions such as social work, counseling, and healthcare.
²
A burnout survey of 7,288 US physicians from all specialties, using the MBI-HSS tool, showed that high burnout was reported by 38% of US physicians and that burnout is more common among physicians than other US workers.
¹¹
Among all specialties, EM had the highest burnout rates, with over 60% of EPs reporting high burnout levels.
¹¹
Emergency medicine is a frontline specialty, and several factors contribute to high burnout in EPs, including night shifts, sleep disorders, job-related strain, fear of making mistakes, and workplace violence.
¹²

In recent years, more female physicians have entered the workforce, leading to increased data availability for studying sex differences in burnout symptoms. In some cultures, female patients preferentially ask for female EPs to attend to them in the ED.
¹³
A recently conducted systematic review of US physicians found that women physicians have a higher likelihood of experiencing burnout compared to male physicians, particularly with respect to the EE dimension of burnout.
¹⁴
Studies focusing on EPs corroborate this finding, indicating that female EPs are at higher risk of burnout compared to male EPs. Additionally, females have higher attrition rates compared to their male counterparts.
<sup>3

,

4

,

15</sup>
Factors cited as contributing to the discrepancy in burnout include greater levels of work-family conflict, greater tendencies to emotionally invest in patients/work, and greater discrimination in salaries and promotions by female physicians.
¹⁴

In contrast to the above studies, findings from a systematic review of 16,016 physicians from the Eastern Mediterranean region indicated no significant difference in burnout rates between male and female physicians.
¹⁶
However, none of the studies in Doraiswamy’s systematic review were designed to compare differences by sex, limiting the interpretations of the findings.
¹⁵
Another systematic review that included 109,628 physicians concluded that inconsistencies in definitions and assessment methods for burnout across studies prevented a reliable determination of the association between burnout and the sex of the physician.
¹⁷

These reviews highlight variability in findings across different regions and contexts, suggesting that the relationship between burnout and sex may vary depending on factors such as cultural norms, healthcare system characteristics, and study methodologies. Therefore, while some studies may indicate a gender disparity in burnout, others may not find such differences, emphasizing the need for nuanced interpretation and context-specific understanding of burnout in healthcare professions.

We conducted this systematic review to address inconsistent data on the gender gap in burnout among EPs. Rotenstein et al have highlighted in their review the inconsistencies in the definitions and assessment methods of burnout; therefore, we focused specifically on studies that used the MBI-HSS tool for assessing burnout.
¹⁷
Despite using a common assessment tool, the included studies employed various methods for reporting burnout scores. Of the 18 studies included in our review,
<sup>18

–

35</sup>
seven reported burnout as the percentage of participants with high burnout; four reported the percentage of participants with high individual components of burnout (EE, DP, PA) but not overall burnout; three reported the mean scores for EE, DP, and PA; and four reported odds ratios and relative risks for EE (Tables 2

–5). This approach allowed us to comprehensively examine and compare the gender disparities in burnout among EPs across different studies.

We collated data from studies reporting burnout as a percentage of the population having high burnout and found that of 1,181 male EPs and 542 female EPs, high burnout was reported in 58.8% and 58.4%, respectively, which indicates nearly equal rates of burnout between male and female EPs
<sup>22

–

24

,

26

,

27</sup>
(Table 2). The two-item tool, known for its brevity and ease of administration, can effectively identify at-risk EM residents showing early signs of burnout.
³⁶
This tool uses two questions from the MBI-HSS and enables consistent, widespread, and longitudinal monitoring of burnout among EM residents at local, regional, and national levels.
³⁶
In our systematic review, we included two studies that used the two-item tool to measure burnout among 4,778 male and 1,502 female EPs. The reported burnout rates were 31% among male EPs and 33% among female EPs (P = 0.05).
<sup>29

,

32</sup>
This slight difference suggests a trend toward higher burnout among female EPs, as indicated by these specific studies using the abbreviated MBI tool.

Four additional studies included in our systematic review, totalling 317 male and 399 female EPs, reported individual components of burnout (EE, DP, PA) as percentages of participants with high or low scores (Table 3).
<sup>18

,

20

,

33

,

34</sup>
These studies found that the proportion of female EPs with high EE and low PA was significantly higher compared to male EPs (P < 0.001). However, levels of DP were comparable between both genders (P = 0.77). This indicates that female EPs may experience greater EE and lower PA, highlighting potential areas of concern for gender-specific burnout interventions in emergency medicine.

Lastly, among the remaining studies that reported odds ratios or relative risks for burnout, 3 of 4 studies indicated a higher risk of burnout among female EPs, while one study showed a higher risk among male EPs (Table 5).
<sup>19

,

25

,

30

,

31</sup>
Specifically, female EPs were found to have a higher risk for the EE component of burnout compared to male EPs, as suggested by these findings. This underscores the gender disparity in burnout risk within the EM profession, emphasizing the need for targeted interventions to mitigate these disparities and support the well-being of all EPs.

In our systematic review, we encountered challenges in performing a meta-analysis due to the variability in how authors reported their findings using the MBI tool to measure burnout among EPs. Specifically, there were limitations stemming from the limited number of studies that reported results separately for male and female EPs, as well as the diversity in how parameters of burnout were reported across these studies. The variation in reporting included differences in the following: •

Whether burnout was reported as overall scores or individual components (EE, DP, PA)

•

The specific metrics used to define high burnout

•

The methods used to analyze and present data (percentages, means, odds ratios, etc).

This variability makes it challenging to aggregate data across studies for a meta-analysis, which typically requires a consistent approach to data reporting and statistical measures. As a result, while our review provides valuable insights into the gender disparities in burnout among EPs, the heterogeneity in reporting prevents a quantitative synthesis of the findings.

Moving forward, standardizing the reporting of MBI-HSS results and burnout parameters in future studies would facilitate more robust meta-analytical approaches to further elucidate the gender gap in burnout among EPs and inform targeted interventions to mitigate burnout in this critical healthcare specialty.

As per the originators of the MBI-HSS the pre-2016 versions of the tool used arbitrary classifications of high burnout, dividing the normative population into tertiles labelled as high, moderate, and low burnout. This approach was later acknowledged as a mistake, leading to the removal of these cutoff classifications from all versions of the MBI-HSS starting with the fourth edition in 2016. This change allowed researchers to view burnout as a continuum within the context of specific populations.
³⁷

In our systematic review, despite including 14 studies conducted after 2016, only three studies reported the mean (±SD) scores of individual components of burnout separately for male and female EPs.
<sup>21

,

28

,

35</sup>
Notably, these studies highlighted significant differences: females exhibited significantly higher EE scores, while males showed significantly higher DP and lower PA scores (P < 0.001).
<sup>21

,

28

,

35</sup>
It is important to note that the study by Yan et al in China included a large cohort of 15,243 participants, which may skew the overall findings toward this study’s results.
³⁵

In most other studies included in our review, researchers derived cutoffs by combining results from individual components (EE, DP, PA), which could explain why overall burnout scores appeared equivocal across these studies (Table 2). This variability in reporting underscores the ongoing challenge of harmonizing burnout measurements across different studies and emphasizes the need for standardized reporting practices to facilitate clearer comparisons and meta-analyses in future research.

Twelve of the articles included in our study were published in 2020 or later, and five studies were conducted during the peak of the COVID-19 pandemic in 2020–2021. It is likely that the pandemic caused unusual fatigue and burnout, resulting in different burnout levels among males and females across different regions.

LIMITATIONS

This systematic review encountered several limitations. One major challenge was the variability in how studies reported burnout, even when using the MBI-HSS tool. Some studies focused on overall burnout, while others reported individual components such as EE, DP and PA. Additionally, studies employed different metrics (percentages, means, or odds ratios), making it difficult to aggregate findings for a meta-analysis. The limited number of studies reporting gender-specific data further constrained our ability to make definitive conclusions about the gender gap in burnout among EPs. Moreover, some studies used outdated versions of the MBI-HSS, which relied on arbitrary cutoffs for high burnout, affecting the accuracy of burnout classification. Finally, several studies were conducted during the COVID-19 pandemic, a period marked by increased fatigue and burnout, potentially skewing the findings and limiting their generalizability beyond that time frame.

CONCLUSION

This review highlights that the science of examining physician burnout is complex and influenced by a wide range of factors. No two studies are perfectly comparable, even when using similar assessment tools, such as the Maslach Burnout Inventory-Human Services Survey. Variables like the type of work, workload, acuity of tasks, job satisfaction, hours worked, frequency of rapid decision-making, critical thinking demands, work-life balance, competing interests, cultural beliefs, and societal norms—such as power distance and hierarchy—all impact burnout differently.

Despite the findings that female emergency physicians report higher emotional exhaustion and lower sense of personal accomplishment than their male counterparts, it is essential for societies and countries to delve deeper into this issue, tailoring studies to their specific contexts and cultures. This review underscores the need for gender-specific strategies to combat burnout among EPs. While both men and women experience significant burnout, the nature of their burnout differs, necessitating targeted interventions to support the well-being of all EPs.

Recommendations for Research

1.

Uniform Reporting of Results: Future studies on burnout among EPs should uniformly report results, including the mean (±SD) for the individual components of the MBI-HSS tool— EE, DP and PA—for both male and female EPs. This standardized approach will facilitate a clearer understanding of the gender gap in burnout across different settings and populations.

2.

Further Research Focus: There is a critical need for further research to delve into the underlying reasons behind the disparities in EE, DP, and PA between male and female EPs. Understanding these factors is essential for developing targeted interventions that address the specific needs of each gender, thereby effectively mitigating burnout.

3.

Move Away from Arbitrary Cutoff Scores: The practice of using arbitrary cutoff scores (high, moderate, low burnout) to categorize burnout levels should be abandoned. This approach, discouraged by the originators of the MBI-HSS tool, does not accurately capture the nuanced experiences of burnout and may lead to misleading conclusions.

4.

Holistic Assessment of Burnout: Rather than focusing solely on overall burnout scores, future studies should emphasize the detailed assessment of EE, DP, and PA. This holistic approach provides a more comprehensive understanding of burnout dynamics among EPs and allows for targeted interventions based on specific components of burnout.

Call to Action by FLAME

Based on the observations regarding burnout among women in EM and female EPs, we are proposing the following measures: 1.

Increased Awareness and Education:
○

Enhance awareness and recognition of burnout within the EM community.

○

Incorporate burnout-related sessions into EM residency training and core curriculum.

2.

Proactive Faculty and Leadership:
○

Faculty and EM leadership should closely monitor all staff, including both female and male EPs, for signs of work-related stress, cognitive overload, and other relevant commitments.

○

Emphasize addressing burnout as a work-related issue rather than a gender-related one, especially when there is no confirmed evidence of gender-specific causes.

3.

Open Discussions on Burnout:
○

Foster an open, non-judgmental dialogue about burnout during departmental peer-review sessions.

○

Ensure that staff at all levels can share their experiences while maintaining psychological safety.

○

Address burnout openly to prevent it from becoming a “silent crisis.”

4.

Psychological Wellness Initiatives:
○

Leadership and management should implement psychological wellness initiatives, such as “Joy @ Work,” iTHRIVE initiatives, and wellness grants.

5.

Peer-Support Committees:
○

Establish interprofessional peer-support committees or teams.

○

Encourage staff to discuss burnout with peers, who may be more approachable, and share best practices for managing burnout.

Footnotes

Section Editor: Elisabeth Calhoun, MD, MPH

Full text available through open access at http://escholarship.org/uc/uciem_westjem

Address for Correspondence: Suman Thakur, MD, Indira Gandhi Medical College & Hospital, Department of Emergency Medicine, Shimla, India, 171001. Email: Suman_thakur13@yahoo.com
03 / 2025; 26:338 – 346

Submission history: Revision received July 16, 2024; Submitted October 24, 2024; Accepted October 28, 2024

Conflicts of Interest: By the WestJEM article submission agreement, all authors are required to disclose all affiliations, funding sources and financial or management relationships that could be perceived as potential sources of bias. No author has professional or financial relationships with any companies that are relevant to this study. There are no conflicts of interest or sources of funding to declare.

REFERENCES

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

Share on Facebook

Tweet

Share via email