Family medicine residency training and burnout: a qualitative study.

BACKGROUND
Almost three-quarters of family practice residents in British Columbia (BC) meet criteria for burnout. We sought to understand how burnout is perceived and experienced by family medicine residents, and to identify both contributory and protective factors for resident burnout.


METHOD
Two semi-structured focus groups were conducted with ten family practice residents from five distinct University of British Columbia training sites. Participants completed the Maslach Burnout Inventory (MBI). The data were analyzed using a thematic analysis approach.


RESULTS
Seventy percent of the focus group participants met criteria for burnout using the MBI. The experience of burnout was described as physical and emotional exhaustion, loss of motivation, isolation from loved ones, and disillusionment with the medical profession. Contributory factors included high workload, burned-out colleagues, perceived undervaluing of family medicine, lack of autonomy, and inability to achieve work-life balance. Protective factors included strong role models in medicine, feeling that one's work is valued and rotations in family medicine.


CONCLUSIONS
The high level of burnout in family medicine residents in BC is a multifactorial and complex phenomenon. Training programs and faculty should be aware of burnout risk factors and strive to implement changes to reduce burnout, including allowing residents increased control over scheduling, access to counseling services and training for resident mentors.


Introduction
Burnout is a psychological syndrome characterized by emotional exhaustion, depersonalization or cynicism and a reduced sense of personal accomplishment that develops in response to prolonged occupational stress and depletion of personal coping resources. 1 The Maslach Burnout Inventory (MBI) is a questionnaire that measures burnout along its three dimensions: emotional exhaustion, depersonalization and personal accomplishment, and has been validated for use among physicians and residents. 2 Using the MBI, studies have estimated that between 18% and 76% of medical residents are burned out. [3][4][5] The prevalence of burnout in residents appears to be higher than that of the general population, 6 and high levels have been reported in many countries, across a wide range of medical and surgical specialties, and in all years of training. [3][4][5][7][8][9][10][11][12][13] Burnout has serious consequences for both medical residents and the patients under their care. Residents meeting criteria for burnout in a Dutch study reported poorer physical health than those who were not burnt out. 14 Burnout was also associated with depression in this study; of the residents meeting criteria for burnout, 25% were also depressed, while 96% of depressed residents also met criteria for burnout. Further, the prevalence of suicidal thoughts in burned out residents has been shown to be more than double that seen in non burned out residents. 9 The burnout experienced by residents also affects patient care. Shanafelt et al. showed that residents meeting criteria for burnout were two to three times more likely than their non burned out colleagues to report suboptimal patient care practices at least monthly. 5 Similarly, West et al. also found burnout to be independently associated with higher rates of selfperceived major medical errors. 15 Despite a growing body of literature on burnout, studies report conflicting results regarding potential contributory and protective factors. 3,16 A 2004 review by Thomas was unable to identify any demographic or personality features that could reliably predict residents at-risk for burnout. 8 Some studies report financial stress contributing to burnout, 3,17 and financial security as protective; 18 others, however, find no such associations. 19 A 2009 review by Ishak et al. described time demands, lack of control over time management and inherently difficult job situations as commonly cited contributors to burnout. 16 Lack of support systems 19 and dissatisfaction with time for leisure and exercise 18 have also been identified as contributing to burnout. The impact of family/personal relationships on burnout is more complex; one study showed lack of time with family and friends associated with burnout, 20 19 "time demands" was the factor most often cited as contributing to burnout. 22 A 2013 study of French general practitioners in training reported the following factors associated with burnout: time spent working, rotations in internal medicine, lack of recognition from senior physicians and dissatisfaction with time for family, friends and leisure. 20 A survey of University of British Columbia (UBC) family medicine residents in 2011 found that 74% of e15 residents met criteria for burnout, defined as either a high emotional exhaustion or depersonalization score on the MBI. 23 In order to design interventions to address these high rates of burnout in BC and elsewhere, a thorough understanding of both protective and contributory factors is essential. The purpose of this study is to qualitatively explore the experiences of burnout as they relate to family medicine training, to identify factors that contribute to and protect against burnout, and to elicit suggestions for decreasing burnout during residency training. This study adds qualitative results to the mostly quantitative data landscape on resident burnout and contributes to the limited literature on burnout in family practice training.

Study design
A focus group format was selected for this study, as this method is useful for exploring attitudes and needs of health care workers. 24 With few qualitative studies on the topic, an inductive thematic analysis approach was selected for this exploratory study. A preconceived theoretical framework was not used; the researchers allowed themes to emerge as the data were analyzed. This study was approved by the UBC Research Ethics Board.

Participants
All 237 family medicine residents in British Columbia were invited to participate in this study via email. After reviewing the study information and focus group times, ten were able to participate. Inclusion criteria involved being a current resident in the UBC Department of Family Practice. Self-identifying as experiencing burnout was not an inclusion criteria. Participants signed a written consent form prior to participating in a focus group. Residents outside of Vancouver were offered the opportunity to participate by teleconference.

Data collection
Two

Data analysis
Data were analyzed for themes using a thematic analysis approach. Transcripts were independently read and coded by the two authors. Discussion and comparison of coding led to the identification of themes and the development of a preliminary thematic framework. This was reviewed by the research team and revised through debate and discussion with ongoing review of the transcripts in the context of larger emerging themes. A final code book was agreed upon that included clear definitions of themes and sub-themes. Transcripts were independently coded by the two authors who met to review and discuss any differences in coding. A final, unified coded dataset was negotiated and data collected under each theme summarized into an analytic memo.

Results
The ten study participants represented five UBC training sites; demographic data are presented in Table 1. Using the MBI, 70% of the residents participating in the focus groups met criteria for burnout. This is similar to the overall burnout rate of 74% reported in a 2011 program-wide survey. 23 Mean scores for residents meeting the criteria for burnout in 2012 and the corresponding categories are shown in Table 2. MBI results for both groups are compared in Table 3. Themes identified during data analysis are presented according to the four main research questions. Figure 1 shows a schematic representation of the study findings.  For many residents, feeling burned out related to e18 isolation from friends and family. One aspect was physical isolation due to long work hours. Some participants described missing important events and celebrations, while one resident moved across the country for residency and felt she neither had time to make new friends, nor to keep in touch with old friends. Isolation was also described as a growing emotional distance between participants and their loved ones, who were unable to relate to their residency experiences. [

Risk factors for burnout
Participants described the medical culture they worked in as contributing to burnout in a variety of ways. One participant noted that burnout seemed to be ubiquitous at work. Some residents were disillusioned by the behavior and comments of their colleagues. Rotations where residents felt a mismatch between their level of supervision, knowledge and responsibility were also identified as contributing to burnout. These rotations added to feelings of inadequacy, guilt and anxiety. The discord between frequent rotation changes and the goal of continuity of care in family medicine was underscored. Residents reported frustration during rotations where discrepancies arose between their learning goals and opportunities. One resident gave an example from his ICU rotation.

I would forget about my objectives of learning. Okay 'I came here to do few central lines and to learn some other monitoring stuff.' Uh, but it feel[s] divert[ed] into what the others -my senior, or my attending -is expecting because at the end of the day, I have to pass this rotation. (P2)
Finally, residency programs' unresponsiveness to resident concerns and feedback was also identified as contributing to burnout. One resident described such unresponsiveness as:

Factors protective against burnout
Residents identified having a supportive medical community with positive role models and mentors as protective against burnout.

Something I've found helpful in terms of trying to, um, keep cynicism away is having… role models around. For example, my family preceptor, she has a very good work-life balance. (P8)
Residents also felt protected against burnout when they knew their work was valued by their patients, preceptors and coworkers. The culture and nature of family medicine was also described as protective against burnout. Participants felt the discipline attracted physicians who valued and strived to achieve work-life balance. A supportive family was identified as protective against burnout. Family members helped residents debrief their day and engage in activities unrelated to medicine.

Suggestions for decreasing rates of resident burnout
Most suggestions corresponded with mitigating the contributing factors or bolstering the protective factors described above. Increasing the length of training for family medicine residency was suggested by both international medical graduates. Other participants felt that burnout could be lessened by a medical culture that placed higher value on work-life balance and self-care. This included the need to address excessive work hours and allowing residents increased control over their schedules. One resident suggested offering 'flex days' which could be used at the first sign of burnout.
Instead of waiting until you're sick, because you've been so decompensated for so long, because you've run yourself ragged and then you get sick and you have to call in sick … it'd be nice just to have a pre-emptive mental health day once in a while to just regain your composure. (P1) Residents also requested additional support services throughout training, including access to counselors, psychologists or group sessions.
If we could all have a personal counselor... or something in that nature... a venue for us to come together and support each other, learning from our mistakes or moving forward so that we don't have to sit with that for the rest of our lives. (P7) Program support and training for physician mentors was also suggested along with establishing an ongoing process for eliciting resident feedback and implementing changes.

Discussion
This qualitative study on resident burnout contributes to the existing, mostly quantitative, literature by offering a rich, in-depth description of the experience of burnout in family medicine residency and identifying contributory and protective factors, as described by residents. All participating residents described experiencing burnout at some point during their training. Similar to the participants in the Satterfield et. al. study, our participants expressed feelings of guilt, anxiety and frustration. 21 Our results also support the literature reporting lack of control over scheduling, lack of time for leisure, family and friends and time demands as contributing to burnout. 3,16,19,20 This study further elaborates on these factors and identifies the role of heavy workload in addition to long work hours, and residents feeling isolated from, as well as insufficient time for, family and friends as contributing to burnout. Unlike other studies, neither financial concerns were cited as contributing to burnout nor was increased financial support suggested as a means to reduce burnout.
Our results indicate that burnout is a complex and e21 multifactorial phenomenon. Our participants identified and emphasized more systemic and program level factors as contributing to burnout than individual level ones. This is consistent with Maslach and Leiter's work on developing interventions to combat burnout, which emphasizes the importance of the worker-workplace interface and of situational and organizational interventions in addition to individual ones 1,26 . Some of the contributory factors identified, such as an unsupportive medical culture and heavy workloads, will require collaboration between many agencies to resolve. Others however, such as frequent rotation changes and responsiveness to resident feedback, could be addressed by individual residency programs. Notably, protective factors elicited did not include the current resident wellness curriculum. Although training sites have different curricula, all offer between one and three lectures focused on self-care or resident wellness and provide residents with a list of resources for resident resilience, including the BC Physician Health Program and 24 Hour Suicide Crisis Line. Residents' suggestions for decreasing burnout rates further emphasized the importance of system and program level changes. These include allowing residents more control over their schedules, offering access to counseling services, support for training physician mentors, work hour limits and a medical culture that places higher value on work-life balance.
Our study identifies a number of aspects of burnout that may be unique to family medicine training, and provides some insight into the high burnout rates found in UBC family medicine residents. Key risk factors for burnout considered specific to family medicine training were: 1) frequent off-service training, 2) devaluing of family medicine in the medical culture, and 3) discrepancy between goals of learning and actual learning on non-family medicine rotations. Compared to other specialty training programs, family medicine residents spend more of their first year on off-service rotations. During these rotations, they report hearing demoralizing and negative comments towards family medicine and experiencing a mismatch between their learning objectives and the learning opportunities available. Tailoring the structure of off-service rotations to meet the future practice needs of family physicians is likely to improve residents' sense of personal accomplishment. Shifting the image of family physicians towards valued members of the medical team is also important.
In contrast to what they experienced off-service, participants in this study found their family medicine rotations helped them build confidence. They noted the protective nature of the longitudinal family medicine clinic, where they developed ongoing relationships with staff, preceptors and patients. These results are consistent with research showing that the development of professional confidence is protective against burnout. 21 The importance of continuity of care in family medicine was also highlighted in a recent qualitative study of family physicians and residents. Participants indicated that continuity of care and the establishment of longterm relationships with patients enhanced family physicians' feelings of professional competence and fostered personal growth. 27 The inability to form long-term patient relationships, a core value of family medicine, could help to explain the distress residents experience during frequent rotation changes and constantly shifting patient lists while on off-service rotations.
This study has some important limitations. Participant recruitment and focus group scheduling posed a challenge; only ten residents were able to attend the focus group, despite more expressing interest in participating. It is thus possible that thematic saturation was not achieved. For similar reasons, although attempts were made to maximize the diversity of our sample, the results likely over represent an urban training experience and underrepresent male residents. However, the relative underrepresentation of male residents in our sample must be viewed in the larger context of a lower representation of men in family practice residency training (e.g. men made up only 25% of residents at the two UBC urban program sites in 2012). Finally, additional research is needed to confirm if these findings persist in family medicine training programs that are more predominantly rural, and those in other provinces.
Our results support a growing interest in advocating for improving medical residents' work-life balance, both as a means to reduce burnout and to improve patient safety. 28 To achieve this, more research is also needed to determine how best to implement e22 the suggestions elicited by this study and whether targeted interventions at the program level will translate into lower rates of burnout. Further insight could be gained with a follow-up study after residents' first few years in practice to see how the experience of burnout changes following transition to practice and whether burnout in residency predicts physician burnout in future. Unfortunately, burnout may not be limited to residency; Lee et. al show that practicing family physicians in Canada also experience high rates of burnout. 29

Conclusion
Resident burnout poses a significant problem: it affects over two thirds of family medicine residents in BC and has wide ranging consequences for patient care, the well-being of residents and their educational experience. This study highlights a number of potentially unique factors relating to burnout in family medicine residents. Program directors, rotation directors and all physicians with teaching responsibilities should be aware of resident burnout and take steps to implement the suggestions outlined above. While systemic and program level changes may be more difficult to implement than individual level interventions, they may also have more potential to decrease rates of resident burnout.