BMC Medical Education (May 2024)

Resident worklife and wellness through the late phase of the pandemic: a mixed methods national survey study

  • Mark Linzer,
  • Sanjoyita Mallick,
  • Purva Shah,
  • Anne Becker,
  • Nancy Nankivil,
  • Sara Poplau,
  • Shivani K. Patel,
  • Caitlin Nosal,
  • Christine A. Sinsky,
  • Elizabeth Goelz,
  • Martin Stillman,
  • Michaella Alexandrou,
  • Erin E. Sullivan,
  • Roger Brown

DOI
https://doi.org/10.1186/s12909-024-05480-5
Journal volume & issue
Vol. 24, no. 1
pp. 1 – 11

Abstract

Read online

Abstract Background System contributors to resident burnout and well-being have been under-studied. We sought to determine factors associated with resident burnout and identify at risk groups. Methods We performed a US national survey between July 15 2022 and April 21, 2023 of residents in 36 specialties in 14 institutions, using the validated Mini ReZ survey with three 5 item subscales: 1) supportive workplace, 2) work pace/electronic medical record (EMR) stress, and 3) residency-specific factors (sleep, peer support, recognition by program, interruptions and staff relationships). Multilevel regressions and thematic analysis of 497 comments determined factors related to burnout. Results Of 1118 respondents (approximate median response rate 32%), 48% were female, 57% White, 21% Asian, 6% LatinX and 4% Black, with 25% PGY 1 s, 25% PGY 2 s, and 22% PGY 3 s. Programs included internal medicine (15.1%) and family medicine (11.3%) among 36 specialties. Burnout (found in 42%) was higher in females (51% vs 30% in males, p = 0.001) and PGY 2’s (48% vs 35% in PGY-1 s, p = 0.029). Challenges included chaotic environments (41%) and sleep impairment (32%); favorable aspects included teamwork (94%), peer support (93%), staff support (87%) and program recognition (68%). Worklife subscales were consistently lower in females while PGY-2’s reported the least supportive work environments. Worklife challenges relating to burnout included sleep impairment (adjusted Odds Ratio (aOR) 2.82 (95% CIs 1.94, 4.19), absolute risk difference (ARD) in burnout 15.9%), poor work control (aOR 2.25 (1.42, 3.58), ARD 12.2%) and chaos (aOR 1.73 (1.22, 2.47), ARD 7.9%); program recognition was related to lower burnout (aOR 0.520 (0.356, 0.760), ARD 9.3%). These variables explained 55% of burnout variance. Qualitative data confirmed sleep impairment, lack of schedule control, excess EMR and patient volume as stressors. Conclusions These data provide a nomenclature and systematic method for addressing well-being during residency. Work conditions for females and PGY 2’s may merit attention first.

Keywords