General surgery residents in the state of North Carolina completed the Maslach Burnout Inventory and the Patient Health Questionnaire–9 depression screen to assess for burnout (A) and depression risk (B), respectively. This cohort of general surgery residents and their attendings were asked to estimate the prevalence of burnout and depression among general surgery residents. Most residents and attendings underestimated the prevalence of both conditions. Blue arrow denotes actual prevalence.
Customize your JAMA Network experience by selecting one or more topics from the list below.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
Williford ML, Scarlet S, Meyers MO, et al. Multiple-Institution Comparison of Resident and Faculty Perceptions of Burnout and Depression During Surgical Training. JAMA Surg. 2018;153(8):705–711. doi:10.1001/jamasurg.2018.0974
What is the prevalence of burnout and depression among general surgery trainees in North Carolina, and do residents’ and attendings’ perceptions of these conditions differ?
In this cross-sectional survey of 92 general residents and 55 attendings, the prevalence of burnout and risk of depression among general surgery residents were high (58 of 77 [75%] and 30 of 76 [39%], respectively). Residents and faculty members significantly underestimated the prevalence of burnout and depression but identified the same barriers to seeking treatment.
Discrepancies exist in actual and perceived levels of burnout and depression among residents and attendings, but a common understanding of barriers to care provides an opportunity for the development of practical interventions.
Prior studies demonstrate a high prevalence of burnout and depression among surgeons. Limited data exist regarding how these conditions are perceived by the surgical community.
To measure prevalence of burnout and depression among general surgery trainees and to characterize how residents and attendings perceive these conditions.
Design, Setting, and Participants
This cross-sectional study used unique, anonymous surveys for residents and attendings that were administered via a web-based platform from November 1, 2016, through March 31, 2017. All residents and attendings in the 6 general surgery training programs in North Carolina were invited to participate.
Main Outcomes and Measures
The prevalence of burnout and depression among residents was assessed using validated tools. Burnout was defined by high emotional exhaustion or depersonalization on the Maslach Burnout Inventory. Depression was defined by a score of 10 or greater on the Patient Health Questionnaire–9. Linear and logistic regression models were used to assess predictive factors for burnout and depression. Residents’ and attendings’ perceptions of these conditions were analyzed for significant similarities and differences.
In this study, a total of 92 residents and 55 attendings responded. Fifty-eight of 77 residents with complete responses (75%) met criteria for burnout, and 30 of 76 (39%) met criteria for depression. Of those with burnout, 28 of 58 (48%) were at elevated risk of depression (P = .03). Nine of 77 residents (12%) had suicidal ideation in the past 2 weeks. Most residents (40 of 76 [53%]) correctly estimated that more than 50% of residents had burnout, whereas only 13 of 56 attendings (23%) correctly estimated this prevalence (P < .001). Forty-two of 83 residents (51%) and 42 of 56 attendings (75%) underestimated the true prevalence of depression (P = .002). Sixty-six of 73 residents (90%) and 40 of 51 attendings (78%) identified the same top 3 barriers to seeking care for burnout: inability to take time off to seek treatment, avoidance or denial of the problem, and negative stigma toward those seeking care.
Conclusions and Relevance
The prevalence of burnout and depression was high among general surgery residents in this study. Attendings and residents underestimated the prevalence of these conditions but acknowledged common barriers to seeking care. Discrepancies in actual and perceived levels of burnout and depression may hinder wellness interventions. Increasing understanding of these perceptions offers an opportunity to develop practical solutions.
Most US physicians experience burnout, a syndrome in which emotional depletion and maladaptive detachment develop in response to prolonged occupational stress.1 Burnout appears to increase the risk of psychiatric conditions, including depression, anxiety, posttraumatic stress disorder, and suicidality.2 Thus, burnout among physicians has garnered widespread attention within the health care sector and is considered by many to be a developing public health concern.3
Burnout has increased since its recognition in the 1970s.3 In a study of more than 6000 US physicians from 2011 to 2014,3 the prevalence of burnout increased from 45% to 54%. Surgeons appear to be particularly vulnerable, with the proportion affected exceeding the national average for physicians.4 Surgical residency may represent a time of elevated risk, given the significant duty hours, patient care demands, and chronic sleep deprivation. Quiz Ref IDIn 2 recent national studies,4,5 the prevalence of burnout among surgical residents was 69%, which exceeds the rate for residents in nonsurgical specialties.
Most studies addressing burnout and depression among surgeons have focused on prevalence. However, developing effective solutions to reduce burnout and depression among surgeons requires a deeper understanding of perceptions, risk factors, and barriers to seeking treatment. Until these data are available, efforts to reduce burnout and depression will be insufficient.
The aims of this study were 2-fold. First, we sought to establish the prevalence of burnout and depression among surgery trainees across the state of North Carolina. Second, we aimed to characterize how resident and attending surgeons perceive burnout and depression to identify potential barriers to seeking care that may be amenable to change.
Quiz Ref IDAll residents and faculty members of the general surgery training programs at Carolinas Medical Center, Charlotte, Duke University, Durham, East Carolina Brody School of Medicine, Greenville, New Hanover Regional Medical Center, Wilmington, University of North Carolina at Chapel Hill, and Wake Forest University, Winston-Salem, in North Carolina, were identified as potential participants. At each institution, the general surgery program director was asked to contact these individuals via email to request participation. Involvement was voluntary, and no incentives were offered. Surveys were completed from November 1, 2016, to March 31, 2017. This study was approved by the institutional review board of the University of North Carolina at Chapel Hill, which did not require informed consent for this anonymized survey.
Unique electronic surveys were created for residents and attendings using Qualtrics survey software (Qualtrics LLC). The 36-question resident survey used validated tools to identify and characterize burnout (Maslach Burnout Inventory [MBI])6 and the risk of major depressive disorder (Patient Health Questionnaire–9 Depression Screen [PHQ-9]; possible scores range from 0-27, with higher scores indicating severe risk of depression).7 Burnout was defined as a high emotional exhaustion (possible scores range from 0-54, with higher scores indicating higher degrees of emotional exhaustion) or depersonalization (possible scores range from 0-30, with higher scores indicating higher degrees of depersonalization) score according to the MBI. We did not use the personal achievement score (possible scores range from 0-48, with higher scores indicating higher degrees of personal achievement) as a criterion because this score has not been used in previous studies.8 The attending survey did not assess for burnout or depression among attendings but instead focused on their perceptions of these conditions among their residents. Both surveys included questions regarding potential causative factors for burnout and depression and barriers to seeking care.
Given the small size of individual training programs and the stigma associated with burnout and depression, surveys were administered anonymously. With the exception of level of training, we did not collect data that could be used to identify residents and attendings such as age or sex. We believed collecting minimal demographic information would reduce the risk of deductive disclosure, thereby leading participants to answer survey questions more truthfully.
Descriptive statistics were calculated (means and SDs for continuous variables and counts and proportions for categorical variables) for the outcomes of the emotional exhaustion score, depersonalization score, personal achievement score, burnout, PHQ-9 score, and suicidal ideation among residents. In addition, descriptive statistics were calculated for covariates, including familiarity with burnout syndrome, any weekly physical activity, involvement in meaningful activities outside the hospital, and mean number of hours worked per week (in the hospital and at home). Linear regression models were used to assess the association of covariates with the continuous outcomes of emotional exhaustion, depersonalization, personal achievement, and PHQ-9 scores. Logistic regression models were used to assess the association of covariates with the outcomes of high emotional exhaustion, high depersonalization, low personal achievement, burnout, depression, and suicidal ideation. In all models, covariates were screened for univariate correlation with the outcome of interest, and only covariates with a significant correlation were included.
To compare median scores on Likert scales, Mann-Whitney tests were used to identify differences between resident and attending physician responses. Comparisons between resident and attending physician responses to binary questions were made using 2-proportion z tests or Fisher exact test when appropriate. In all analyses, P < .05 was used to assess statistical significance. All analyses were performed using R software (R, version 3.2.2; R Core Team).9
A total of 158 general surgery residents were invited to participate, with 92 (58.2%) responding. A total of 137 attending surgeons were invited to participate, with 55 (40.1%) responding. These response rates are significantly higher than those for larger studies on burnout recently published in the literature (8.6%-19.2%).3-5 Fifteen residents (16% of resident respondents) did not complete the MBI or PHQ-9 portions of the survey and thus were excluded from analyses of these outcomes. One resident completed the MBI portion but not the PHQ-9 portion and was thus excluded from analyses of PHQ-9 and suicidal ideation.
Quiz Ref IDFifty-eight residents (75%) met criteria for burnout. Median (interquartile range [IQR]) personal achievement score was 37 (33-41); median (IQR) emotional exhaustion score, 31 (23-40); and median (IQR) depersonalization score, 15 (11-21). Forty-two residents (55%) had high emotional exhaustion and high depersonalization scores. Significant associations were revealed between high emotional exhaustion and high depersonalization scores (42 of 77 residents [55%]; P < .001) and high emotional exhaustion and low personal achievement scores (37 of 77 residents [48%]; P = .005). Further classification of dimensions of burnout according to the MBI appears in Table 1.
Thirty of 76 residents (39%) who completed the PHQ-9 portion of the survey had a PHQ-9 score of 10 or greater, consistent with moderate to severe risk of depression. Median PHQ-9 score was 8, with an IQR of 4 to 11, consistent with mild depression. Nine of 77 residents (12%) endorsed suicidal ideation within the 2 weeks before completing the survey. Further representation of PHQ-9 scores is found in Table 2.
Burnout was associated with a 6-point increase in PHQ-9 score (coefficient [SE], 6.08 [1.41]; P < .001). Twenty-eight of 58 residents (48%) with burnout also met criteria for depression (P = .03). We found no significant association between burnout and suicidal ideation (9 residents with suicidal ideation and burnout, 49 with burnout only, 0 with suicidal ideation only, and 18 with no suicidal ideation or burnout; P = .11), although all residents acknowledging suicidal ideation met criteria for burnout and depression. We found no association between level of training and burnout, depression, or suicidality.
Eighty-five of 90 residents (94%) and 55 of 57 attendings (96%) were familiar with burnout (P = .71). Sixty-one of 87 residents (70%) and 41 of 56 attendings (73%) agreed or strongly agreed that burnout is a legitimate medical condition (P = .69). Twelve of 83 residents (14%) incorrectly believed that burnout is a reportable condition to the North Carolina State Medical Board. Sixty-one of 83 residents (73%) were aware of resources available for physicians experiencing burnout, whereas only 32 of 55 attendings (58%) were aware of resources (P = .09).
Residents were asked to rank 10 causative factors of burnout identified by previous studies based on the level of frustration that they caused.8 Thirty-nine of 82 residents (48%) identified lack of time to exercise, for self-care, and for doing things that they enjoy as most frustrating. Eleven of 82 (13%) identified conflicting responsibilities among work, home, and family, and 11 of 82 (13%) identified feeling underappreciated.
Quiz Ref IDForty-one of 73 residents (56%) identified the top barrier to seeking care for burnout as inability to take time off from work to seek treatment; 19 of 73 (26%) identified ambivalence, avoidance, and/or denial of the problem; and 6 of 73 (8%) identified negative stigma. Among attending surgeons, 15 of 51 (29%) identified ambivalence, avoidance, and/or denial as the top barrier to seeking care; 13 of 51 (25%) identified negative stigma; and 12 of 51 (24%) identified inability to take time off from work to seek treatment. Further reporting of perceived top barriers to seeking care for burnout is given in Table 3.
Quiz Ref IDMost residents (40 of 76 [53%]) correctly estimated that more than 50% of residents were at high risk of burnout, whereas only 13 of 56 attendings (23%) correctly estimated this prevalence. The median resident believed that attendings would recognize 0 to 25% of residents as experiencing burnout, whereas the median attending believed that 25% to 50% of residents experience burnout (P < .001). However, residents expected attending physicians to estimate a lower prevalence of burnout among residents than they actually did. Most residents (42 of 83 [51%]) and attendings (42 of 56 [75%]) underestimated the true prevalence of residents at elevated risk of depression (P = .002). The Figure represents resident and attending perceptions of burnout and depression.
We used linear regression models to analyze the association of covariates with emotional exhaustion, depersonalization, personal achievement, and PHQ-9 scores. Residents participating in any physical activity on a weekly basis had a 5.7-point decrease in emotional exhaustion score (coefficient [SE], 5.70 [2.69]; P = .04). Residents involved in activities that they believe to be meaningful outside of the hospital scored 5.0 points lower on the PHQ-9 (coefficient [SE], −5.01 [1.28]; P < .001).
We used logistic regression models to analyze the association of covariates with high emotional exhaustion and depersonalization scores. The mean number of hours worked per week was associated with increased odds of high depersonalization (odds ratio, 1.04; 95% CI, 1.10-1.09; P = .02). Every additional 10 hours worked per week was estimated to be associated with increased odds of high depersonalization by a factor of 1.53 (95% CI, 1.10-2.34).
The well-being of surgeons and other health care professionals is essential for a high functioning health care system. Burnout and depression can harm surgeons and patients. Previous studies have established an association between burnout and medical error, reduced levels of empathy, and early departure from the workforce.10 Given the potential of burnout to hinder the delivery of care, the wellness of physicians has become a discrete goal within the larger mission to improve the quality of US health care.11
The prevalence of burnout among surgical residents in our study (75%) is higher than recently published estimates of burnout among general surgery residents (69%) and all US physicians (54%).3-5 Of note, our resident survey response rate of 58.2% is significantly higher than the response rates in these other studies.3-5 Thus, our study data may more accurately reflect the prevalence of burnout among surgery residents.
Based on their PHQ-9 score, 40% of residents in our study were at risk of moderate to severe depression. Comparatively, a 2015 systematic review and meta-analysis including 9447 US resident physicians2 found that 20.9% of individuals were at risk of depression. Our study is consistent with a growing body of literature that suggests a higher prevalence of depression among physicians compared with the general population, which is estimated to be 7.6% for Americans 12 years and older.12
Physician suicide has gained national attention during the past several years. An estimated 300 to 400 physicians die by suicide each year.13 In 2008, a cross-sectional survey14 found the prevalence of suicidal ideation among US surgeons to be 6.3%. From 2000 to 2014, suicide was the second leading cause of death for resident physicians.13 In our study, 12% of residents endorsed suicidal ideation in the 2 weeks preceding the survey.
Our study compared perceptions of burnout and depression between residents and attending surgeons. Given the considerable harms related to burnout and depression, it is critical for attending physicians, particularly those participating in resident education, to learn to identify residents with these conditions accurately. Our data suggest that attendings perceive fewer cases of burnout and depression among residents than are present. We consider the following 3 potential reasons for this discrepancy: (1) attendings may be unfamiliar with the signs and symptoms of burnout and depression in residents, (2) attendings may perceive the signs and symptoms of these conditions as normal within the surgical culture, and (3) residents may conceal burnout from their attendings. We believe that addressing these issues through formal education of surgical faculty will improve identification of and response to residents who experience burnout and depression. In addition, attendings who appreciate the high prevalence of burnout and depression among surgery residents are more likely to be supportive of implementing interventions designed to ameliorate these conditions.15
Residents also underestimate the prevalence of resident burnout and depression. Despite 75% of residents scoring positive for burnout, only 53% estimated that most residents were affected by it. Likewise, although 40% of residents were at elevated risk of depression, more than half underestimated the prevalence of depression. We consider the following 4 potential reasons for these discrepancies: (1) residents may be unfamiliar with the signs and symptoms of burnout and depression in their coresidents, (2) residents may perceive the signs and symptoms of these conditions as normal within surgical culture, (3) residents are practicing in isolation, and (4) residents fear being labeled as inadequate, underperforming, or the weakest link. These beliefs may result in compensatory behaviors, such as concealing their true status from their attendings and resident colleagues.
More than 90% of residents and attendings were familiar with burnout, and more than 70% strongly agreed that burnout is a legitimate medical condition. Despite this reported understanding, the data reveal that both cohorts underestimate the prevalence of resident burnout and depression.
An improved understanding of risk factors for burnout and depression and barriers to seeking care will facilitate the creation of targeted interventions. Residents and attendings must acknowledge burnout as a legitimate medical condition and become aware of resources available for treatment. That said, awareness alone does not ensure the use of resources. Of interest, residents and attendings were unified around the same top 3 barriers to seeking care, which highlights an opportunity for shared understanding and targeted interventions.
Our data demonstrate that the number of hours worked per week by residents is associated with burnout and, more specifically, the likelihood of depersonalization. In our study, residents reported working a mean of 95 hours per week (inclusive of work in the hospital and at home), which exceeds the current Accreditation Council for Graduate Medical Education Association guidelines of 80 hours per week averaged across 4 weeks. Although these hours were self-reported and we did not collect values reported in duty hour logs (which are known to be inaccurate), this finding represents a potentially modifiable factor in addressing burnout. Attendings estimated that residents work a mean of 81 hours per week, which is more consistent with current duty hour restrictions. The discrepancy between attending and resident estimates in our study may indicate that residents are not accurately reporting duty hours to their faculty or may reflect a lack of appreciation by faculty of work performed at home. In addition, a difference in perspective between residents and faculty as to what is considered work compared with education might be a factor, with faculty potentially considering education not to be part of what should be included as reported work hours. Differing perceptions about resident work hours may explain why residents identified time constraints as the top factor limiting care for burnout, whereas attending physicians did not.
Although we found no significant association between personal health behaviors and burnout (except weekly exercise), the data indicate the need for improvement in self-care among residents. Thirty-one of the 85 residents (36%) did not exercise at all, 41 of 76 (54%) only ate 2 meals per day, 61 of 76 (80%) slept 5 to 6 hours per night, mean (SD) consumption of caffeinated beverages was 2.5 (0.8) per day, and 9 of 76 (12%) were at risk for heavy alcohol consumption. Further reporting of resident lifestyle factors is found in Table 4.
It is important to study how an individual’s responses to objective markers such as the MBI and PHQ-9 change throughout residency and in response to interventions designed to reduce burnout. Going forward, we plan to obtain longitudinal data regarding burnout and depression throughout the course of a surgeon’s maturation from student to resident to attending.
In North Carolina, the results of this survey have galvanized an initiative to respond to and prevent burnout and depression during surgical training. Although not featured in this study owing to our desire to prevent deductive disclosure, we reported program-specific prevalence of burnout, depression risk, and suicidal ideation to all of the program directors at all participating sites so that immediate intervention was possible.
Our study indicates that programs focused on prevention, identification, and treatment of burnout among surgeons are necessary for ensuring the health of the surgical workforce. In North Carolina, these programs are developing and expanding. For example, in 2012, the University of North Carolina School of Medicine developed the Taking Care of Our Own program for responding to burnout and associated conditions in residents and other health care professionals. This innovative program is dedicated to “increasing awareness, providing psychoeducation and offering assessment and treatment in a confidential and supportive setting that is optimized to destigmatize seeking help for emotional distress.”16 In 2016, the program launched a Peer Support Program designed to connect health care professionals affected by unanticipated patient outcomes with trained colleagues.17 More recently, the Department of Surgery of the University of North Carolina at Chapel Hill has partnered with Taking Care of Our Own program leaders to develop an educational series to promote awareness of burnout and its sequelae, share department-specific prevalence data, and provide information for surgeons in need of mental health services.15 Given the collective interest in burnout syndrome across the general surgery programs in our state, we plan to share resources and strategies designed to assist surgeons experiencing burnout. As we continue to implement innovative programs to improve wellness, longitudinal use of our survey tool will be needed to assess effectiveness.
The survey was only administered to surgery trainees in North Carolina, which is a limited demographic with potentially unique risk factors and health behaviors. However, the risk factors and barriers to care observed within our population are consistent with previously reported themes in the literature.8 In addition, 1 of the 6 general surgery programs invited to participate did not submit any resident or attending survey responses and was not represented by our analysis.
We hypothesize that symptoms of burnout and depression may vary based on surgical rotation and time of year or in parallel with other life variables. Our cross-sectional study cannot capture variability over time. As with all surveys, we cannot determine whether questions were answered truthfully. Stigmatization of burnout and mental illness, as well as a perceived disclosure of these conditions to state licensing agencies, may lead to inaccurate reporting of symptoms because of fear of privacy violations.
We did not examine perceived causative factors of burnout among residents. Instead, residents were presented with a list of 10 factors identified by previous studies as contributing to burnout and asked to rank-order them based on how frustrating they believed them to be.8 However, a more thorough understanding of what surgical residents perceive to be the underlying drivers of burnout, which may not directly equate to reported frustrations, is essential for successful development of strategies focused on prevention and treatment in surgical trainees.
Despite the work that has been performed to characterize and mitigate the harmful effects of burnout and depression in physicians, the challenge continues.18 Our study agrees with existing evidence that the prevalence of burnout and depression among surgical residents is high. The presence of suicidal ideation should be a call to action.
Our study expands the existing literature by comparing perceptions from residents within this high-risk group with those of their attendings. We found that residents and attendings underestimated the true prevalence of burnout and depression; however, both cited the same barriers to seeking care. Systematic improvements can be directed toward increasing awareness and facilitating access to confidential treatment.
Surgeons must support a culture that identifies residents at risk for burnout and depression and offers them practical solutions to prevent and treat these conditions. The practice of surgery demands a high level of mental, physical, and emotional investment. Tension exists between strategies to reduce burnout and depression and the rigorous nature of the work of surgeons. If we are to be successful at reducing these conditions in this group, continued discussion and collaboration between surgery residents and attendings is vital.
Accepted for Publication: February 11, 2018.
Corresponding Author: Michael L. Williford, MD, Department of Surgery, University of North Carolina at Chapel Hill, 9 Piccadilly Ct, Durham, NC 27713 (email@example.com).
Published Online: May 2, 2018. doi:10.1001/jamasurg.2018.0974
Author Contributions: Dr Scarlet had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Williford, Fine, Clancy, Meltzer-Brody, Farrell.
Acquisition, analysis, or interpretation of data: Williford, Scarlet, Meyers, Luckett, Fine, Goettler, Green, Hildreth, Meltzer-Brody, Farrell.
Drafting of the manuscript: Williford, Scarlet, Meyers, Fine, Green, Meltzer-Brody, Farrell.
Critical revision of the manuscript for important intellectual content: Williford, Scarlet, Meyers, Luckett, Goettler, Clancy, Hildreth, Meltzer-Brody, Farrell.
Statistical analysis: Williford, Scarlet, Luckett, Fine.
Administrative, technical, or material support: Meyers, Meltzer-Brody, Farrell.
Study supervision: Meyers, Green, Hildreth, Meltzer-Brody, Farrell.
Conflict of Interest Disclosures: None reported.
Create a personal account or sign in to: