Ongoing Evaluation of Burnout and Graduate Medical Education-Led Learning Environments to Improve Resident Well-Being | BMC medical training


For all GME residents, scores on the emotional exhaustion, depersonalization, and personal accomplishment subscales improved over time, but the changes were not statistically significant. However, all three measures of the MBI subscale improved significantly among residents in the internal medicine residency program. We noted a high mean baseline rate of burnout (e.g. mean emotional burnout score of 28.12 rated as high) which is consistent with the literature, with recent studies reporting rates of physician burnout up to 89%. [9, 16, 19,20,21,22]. Previous research has shown that burnout rates increase significantly during residency training [23]. Rosen et al. reported in a cohort of internal medicine residents who were followed longitudinally that they had significantly more depersonalization and emotional exhaustion at the end of the residency year. may have stemmed or lessened the natural progression of resident burnout until it worsened over the study period [24, 25].

We used a time-series evaluation that helps refine the intervention and follow-up to determine evidence of effectiveness. Assessing time series using PDSA cycles offers an advantage over simple before-and-after studies in that it allows teams to draw conclusions about whether improvement has occurred, as well as whether the improvement has occurred. intervention led to the observed improvement. 19 The use of PDSA to improve medical education has been reported. Arnstead et al. used PDSA cycles to improve the frequency of competency-by-design assessment among surgical residents in Canada [26]. Dunbar et al. demonstrated that the use of PDSA cycles improved physician recognition and reporting of patient safety events [27],

The 22-item MBI considered the gold standard self-reported questionnaire was used in our study to measure burnout, but some recent studies have used other measures. For example, Choi et al. used a one-item MBI scale and reported less exhaustion when measured twice from 2011 to 2013 (2.35 to 2.33, p= 0.023) [19]. Mari et al. used the Copenhagen Burnout Inventory (CBI) and noted a clinically significant reduction in burnout after 1 year of intervention implementation [15].

The benefits of residency training may be offset by rigorous educational demands, long work hours, a lack of autonomy, an imbalance between work, home, and personal life, and a lack of a supportive and nurturing environment. These factors related to the learning environment can have adverse effects on the mental health of residents and a significant proportion of residents can experience burnout. [19]. Thus, an important component of our study focused on assessing the unique learning environment of each residency training program and developing appropriate interventions. For all GME residents, self-defined burnout scores showed a statistically significant decrease over time. Since the MBI was not significant at the GME level, this may suggest that a self-defined assessment of burnout may be more discriminating. All GME residents also had reduced perceptions of impaired social and personal relationships, with improved perceptions of burnout support by the program. Bird et al., also using a single nonexclusive burnout item and questions regarding burnout support, learner satisfaction, and learner engagement, demonstrated that a wellness program fostered a sense of unity among peers and created an additional support system. [25].

Although this was a GME-wide approach, individual programs responded differently. The internal medicine residency program achieved statistically significant improvement on all measures of the MBI and the learning environment. This variable result is not unexpected, as different program-specific conditions, commitment, academic status, leadership, and learning environments would influence residents’ perceptions. Other researchers have also noted variations in individual residency program responses to wellness interventions. Unlike our study in which we used organization-level interventions, Aggarwal et al. the use of individual intervention introduced a 12-week wellness program in five residency training programs and noted that general surgery residents never implemented the booster sessions due to lack protected time and change in program direction, while anesthesiology residents had organizational challenges [16]. It is also possible that the observed improvements in the internal medicine program are due to other factors that were not measured. For example, the significant results may have been due to a larger sample size, but it should be noted that responses from family physicians outnumbered those from internal medicine residents.

To reach residents’ perceptions of well-being, we provided organization-led and individual-led interventions. This is consistent with previous reports showing that organization-directed interventions compared to individual interventions were associated with a significant reduction in emotional exhaustion and depersonalization scores. [11, 28, 29]. DeChant et al. reviewed studies on 4 unique categories of organization-led workplace interventions regarding teamwork, time, transitions, and technology. They concluded that organization-led workplace interventions that improve processes, optimize electronic medical records (EHRs), reduce workload through the use of scribes, and implement team-based care can reduce physician burnout. [30]. In our internal medicine residency program with significantly improved results; associated major interventions were organization-led, including leadership change, faculty recruitment, resources, and service workflow improvement.

Overall, more work is needed to alleviate burnout and its sequelae such as depression and suicide among postgraduate trainees. Naji et al. in a systemic review that spanned the past 2 decades noted a clustered prevalence of burnout involving 31,210 residents from 47 countries and reported that the prevalence of burnout has remained unchanged over the past 2 decades. Burnout varied by region, with the lowest level in European countries. They concluded that current wellness efforts and policies have not changed the prevalence of burnout worldwide and that future research should focus on understanding systemic factors and building on these findings to design interventions to address burnout. [31]. Thus, more studies of interventions similar to our study using continuous quality improvement and PDSA cycles are needed worldwide.


There are many limitations to this study. The findings come from a single county hospital located in an underserved minority community; thus, the results may not be generalizable to different locations. Since this is an observational study with no control groups, the changes reported could be caused by other unobserved factors; thus, the reported associations cannot be interpreted as causal. Specifically, we did not focus on the role of ACGME surveys and citations on interventions and perceptions of resident well-being. Although we used the validated MBI survey, our learning environment surveys were locally developed. There is still an issue of incomplete data and selection bias, but the 73-82% response rate would be classified as a high response rate. Other limitations were that each time point included residents from different graduate years and that it may take longer for interventions to be effective. Another potential limitation of this study is that data pooling may have occurred since data points within training programs may be more similar than between training programs.

Source link


Comments are closed.