How Do Areas of Work Life Drive Burnout in Orthopaedic Attending Surgeons, Fellows, and Residents?

Abstract

Background: Concerning levels of burnout have been reported among orthopaedic surgeons and residents. Defined as emotional exhaustion and depersonalization, physician burnout is associated with decreased productivity, increased medical errors, and increased risk of suicidal ideation. At the center of burnout research, person-centered approaches focusing on individual characteristics and coping strategies have largely been ineffective in solving this critical issue. They have failed to capture and address important institutional and organizational factors contributing to physician burnout. Similarly, little is known about the relationship between burnout and the working environments in which orthopaedic physicians practice, and on how orthopaedic surgeons at different career stages experience and perceive factors relevant to burnout.

Questions/purposes: (1) How does burnout differ among orthopaedic attending surgeons, fellows, and residents? (2) What specific areas of work life are problematic at each of these career stages? (3) What specific areas of work life correlate most strongly with burnout at each of these career stages?

Methods: Two hundred orthopaedic surgeons (residents, fellows, and attending physicians) at a single institution were invited to complete an electronic survey. Seventy-four percent (148 of 200) of them responded; specifically, 43 of 46 residents evenly distributed among training years, 18 of 36 fellows, and 87 of 118 attending physicians. Eighty-three percent (123 of 148) were men and 17% (25 of 148) were women. Two validated questionnaires were used. The Maslach Burnout Inventory was used to assess burnout, measuring emotional exhaustion and depersonalization. The Areas of Worklife Survey was used to measure congruency between participants and their work environment in six domains: workload, control, reward, community, fairness, and values. Participants were invited to openly share their experiences and suggest ways to improve burnout and specific work life domains. The main outcome measures were Maslach Burnout Inventory subdomains of emotional exhaustion and depersonalization, and Areas of Worklife Survey subdomains of workload, control, reward, community, fairness and values. We compared outcome measures of burnout and work life between groups. Simple linear regression models were used to report correlations between subscales. Stratified analyses were used to identify which group demonstrated higher correlations. All open comments were analyzed and coded to fully understand which areas of work life were problematic and how they were perceived in our population.

Results: Nine percent (7 of 80) of attending surgeons, 6% (1 of 16) of fellows, and 34% (14 of 41) of residents reported high levels of depersonalization on the Maslach Burnout Inventory (p < 0.001). Mean depersonalization scores were higher (worse) in residents followed by attending surgeons, then fellows (10 ± 6, 5 ± 5, 4 ± 4 respectively; p < 0.001). Sixteen percent (13 of 80) of attending surgeons, 31% (5 of 16) of fellows, and 34% (14 of 41) of residents reported high levels of emotional exhaustion (p = 0.07). Mean emotional exhaustion scores were highest (worse) in residents followed by attending surgeons then fellows (21 ± 12, 17 ± 10, 16 ± 14 respectively; p = 0.11). Workload was the most problematic work life area across all stages of orthopaedic career. Scores in the Areas of Worklife Survey were the lowest (worse) in the workload domain for all subgroups: residents (2.6 ± 0.4), fellows (3.0 ± 0.6), and attending surgeons (2.8 ± 0.7); p = 0.08. Five problematic work life categories were found through open comment analysis: workload, resources, interactions, environment, and self-care. Workload was similarly the most concerning to participants. Specific workload issues identified included administrative load (limited job control, excessive tasks and expectations), technology (electronic medical platform, email overload), workflow (operating room time, patient load distribution), and conflicts between personal, clinical, and academic roles. Overall, worsening emotional exhaustion and depersonalization were most strongly associated with increasing workload (r = - 0.50; p < 0.001; and r = - 0.32; p < 0.001, respectively) and decreasing job control (r = - 0.50; p < 0.001, and r = - 0.41; p < 0.001, respectively). Specifically, in residents, worsening emotional exhaustion and depersonalization most strongly correlated with increasing workload (r = - 0.65; p < 0.001; and r = - 0.53; p < 0.001, respectively) and decreasing job control (r = - 0.49; p = 0.001; and r = - 0.51; p = 0.001, respectively). In attending surgeons, worsening emotional exhaustion was most strongly correlated with increasing workload (r = - 0.50; p < 0.001), and decreasing job control (r = - 0.44; p < 0.001). Among attending surgeons, worsening depersonalization was only correlated with increasing workload (r = - 0.23; p = 0.04). Among orthopaedic fellows, worsening emotional exhaustion and depersonalization were most strongly correlated with decreasing sense of fairness (r = - 0.76; p = 0.001; and r = - 0.87; p < 0.001, respectively), and poorer sense of community (r = - 0.72; p = 0.002; and r = - 0.65; p = 0.01, respectively).

Conclusions: We found higher levels of burnout among orthopaedic residents compared to attending surgeons and fellows. We detected strong distinct correlations between emotional exhaustion, depersonalization, and areas of work life across stages of orthopaedic career. Burnout was most strongly associated with workload and job control in orthopaedic residents and attending surgeons and with fairness and community in orthopaedic fellows.

Clinical relevance: Institutions wishing to better understand burnout may use this approach to identify specific work life drivers of burnout, and determine possible interventions targeted to orthopaedic surgeons at each stage of career. Based on our institutional experience, leadership should investigate strategies to decrease workload by increasing administrative support and improving workflow; improve sense of autonomy by consulting physicians in decision-making; and seek to improve the sense of control in residents and sense of community in fellows.

Fig. 1

This graph shows the differences in Maslach Burnout Inventory domain scores among the medical roles. ^ap ≤ 0.05 indicates burnout domains with differences among medical roles.

Fig. 2

This graph shows the differences in Areas of Worklife Survey domain scores among the medical roles. ^ap ≤ 0.05 indicates categories with differences among medical roles.

Fig. 3 (A-B)

These radar graphs demonstrate the spread of the correlation of each Areas of Worklife Survey score with (A) emotional exhaustion, and (B) depersonalization. Points that are closer to the center demonstrate a stronger relationship between the Maslach Burnout Inventory domains (emotional exhaustion, depersonalization) and Areas of Worklife Survey domains (workload, control, reward, community, fairness, values).