Nationwide trend analysis of burnout among resident physicians in Japan before and after the national workstyle reform: a repeated cross-sectional survey of 25 368 participants

Objectives: This study examined temporal changes in burnout and working environments among Japanese resident physicians across the COVID-19 period and the preparation and implementation phases of the national workstyle reform. Furthermore, it assesses associated changes in their working environments. Design: This study was conducted as a multicentre repeated cross-sectional study. Setting: Data were used from the General Medicine In-Training Examination surveys conducted in 2020, 2022, 2023 and 2024. Participants: Participants included postgraduate year 1 (PGY-1) and PGY-2 residents across Japan. Primary and secondary outcome measures: Data on working hours, emergency duties, inpatient load, self-study time and burnout were collected through standardised questionnaires. Multivariable logistic regression was performed to assess associations between workstyle reform phases and burnout, adjusting for demographic and institutional characteristics. Results: Altogether, 25 368 residents participated (6816 in 2020; 6063 in 2022; 6584 in 2023; 5905 in 2024). Burnout prevalence was the highest during the COVID-19 pandemic in 2020 (21.7%), which decreased to 13.3% in 2024. Weekly working hours of ≥80 hours declined from 19.9% to 2.3%, and those working <45 hours increased from 2.5% to 24.7%. Compared with the pre-reform period (2020–2022), odds of burnout were significantly lower during the preparation (2023; adjusted OR (aOR) 0.60, 95% CI 0.55 to 0.66) and implementation (2024; aOR: 0.67, 95% CI 0.61 to 0.74) phases. Conclusions: Residents’ burnout was the highest in 2020 and subsequently declined in parallel with improvements in job satisfaction, working hours and clinical workload. These findings suggest that system-level efforts to improve workload balance may be associated with better resident well-being, while causal attribution to the reform should be interpreted cautiously.