Self-isolation during the COVID-19 pandemic is associated with increased risk of burnout among physician trainees: A cross sectional study

Abstract

The COVID-19 pandemic emerged in late 2019. Previous research has shown a significant prevalence of burnout among physician trainees, with concern that the pandemic will increase burnout. We aimed to assess this risk among trainees at a large academic hospital. We performed a cross-sectional study during the pandemic using a survey that included the Maslach Burnout Inventory. The response rate was 94.7%. Among trainees, 58.5% changed their living arrangements to protect family. Psychological well-being was negatively affected in 81.7% and clinical performance in 64.3%; 13.8% were at high risk of burnout. Emotional exhaustion (EE) scores were high in 50% and depersonalization (DP) scores in 28.8%; a sense of personal accomplishment was low in 41.9%. Increased risk of burnout was associated with male gender and increased exposure to suspected COVID-19 cases. Risk of high EE correlated with an increased number of children and risk of high DP with male gender. High EE and DP score correlated with increased exposure to suspected and confirmed COVID-19 patients. Trainees who self-isolated to protect family were more likely to experience high DP and burnout. Trainees in surgical specialties were more likely to feel their clinical performance was negatively affected. The results suggest that a significant percentage of trainees are at high risk of burnout during the pandemic especially those that attempted self-isolation. Training programs should incorporate methods to maintain well-being and coping, including adequate time off between shifts. Future research should evaluate other aspects of trainee well-being in relation to self-isolation and/or changed living arrangements. Key words: Burnout, Medical education, pandemic, coronavirus, psychological wellbeing, medical residents

Burnout in residents during the first wave of the COVID-19 pandemic: a systematic review and meta-analysis

“…All studies were cross-sectional with an anonymous online survey. In total, eight of 23 (34.8%) studies were conducted in North America ( 23 – 29 , 45 ), 6 (26.1%) in Europe ( 30 – 35 ), 5 (21.7%) in Asia ( 36 , 37 , 39 – 41 ), 2 (8.7%) in South America ( 42 , 43 ), and 1 in Africa (4.4%) ( 38 ), but 1 (4.4%) was an international study with countries from different continents ( 44 ). Twelve studies (52.2%) were national studies ( 23 , 25 , 27 – 31 , 34 – 37 , 39 ).…”

“…The response rate ranged from 94.8 to 7.9%. Ten studies included residents from all specialties (30.3%) ( 26 , 32 , 37 – 40 , 42 – 45 ), and eight were doing surgery (34.8%) ( 23 , 25 , 27 – 29 , 34 , 35 , 41 ). Our pooled studies included ≥ 35,230 residents, and 71% of them were in direct contact with COVID-19 patients.…”

“…Overall, 15 of 23 (65.2%) studies used the original MBI or its validated modifications as a method for burnout diagnosis or measurement ( 11 , 12 ) ( Table 1 and Supplementary Table 1 ). Fourteen studies (60.9%) provided the prevalence of burnout syndrome in residents ( 23 – 26 , 29 – 31 , 35 – 37 , 39 , 42 – 44 ), six studies (26.1%) provided the prevalence of high/low dimension (EE, DP, PA) ( 23 , 30 , 31 , 33 , 36 , 37 ), and finally, three studies (13%) presented the results as mean (SD) or median (range) of each dimension ( 25 , 30 , 37 ). Some of the studies presented the burnout results in more than one way ( Supplementary Table 2 ).…”

“…Eleven articles (47.8%) studied point-prevalence of burnout during the first wave of the pandemic without any comparator group ( 23 – 26 , 30 , 31 , 36 , 37 , 42 , 44 ) and eight studies (34.7%) compared the prevalence of burnout between residents and other physicians, students, or other health-workers during the period studied ( 27 , 28 , 32 , 33 , 38 , 40 , 43 , 45 ). Four studies (17.4%) showed burnout prevalence in comparison with previously collected data ( 29 , 34 , 35 , 41 ).…”

“…Figure 2 shows the funnel plot of the overall prevalence of burnout syndrome. From the 15 studies of burnout syndrome (categorical definition), we included, in the analysis, 11 studies that made the diagnosis of burnout syndrome with the original validated MBI ( 11 , 12 ) or posterior validated versions ( 23 , 25 – 27 , 29 , 30 , 35 – 37 , 43 , 44 ). The estimated overall pooled prevalence (95% CI) was 0.40 (0.26–0.57).…”