Toward better prevention of physician burnout: insights from individual participant data using the MD-specific Occupational Stressor Index and organizational interventions.

Abstract

Background: Physician burnout has become a public-health crisis. The need is dire for robust organizational solutions, focusing on reduction of specific stressors. The physician-specific Occupational Stressor Index (OSI) based on cognitive ergonomics can help. Individual-participant data (IPD) from different studies addressing physician burnout are lacking.

Aims: To perform IPD analysis regarding job stressors and their relation to physician burnout and to utilize the IPD results to inform a systematic review of the stressors that show an association with physician burnout, focusing on intervention studies.

Methods: PRISMA guidelines are followed for the IPD analysis and systematic review of intervention studies on the implicated stressors, taking the COVID-19 pandemic into consideration. The IPD analysis is performed on studies using the physician-specific OSI vis-à-vis burnout assessed by the Copenhagen Burnout Inventory (CBI). Odds ratios (OR) ± 95% confidence-intervals (CI) are reported, adjusting for age, gender and caring for patients with suspected COVID-19 infection.

Results: Three studies fulfilled the inclusion criteria, providing complete IPD data for 95 physicians. Thirty-two (33.7%) physicians had total OSI scores >88, for which intervention is urgently needed. Unit-change in the total stressor burden assessed via OSI yielded OR = 1.11 (95%CI: 1.03-1.18) (p = 0.003) for personal burnout, OR = 1.17 (95%CI: 1.08-1.26) (p = 0.0001) for work-related burnout and OR = 1.07 (95%CI: 1.01-1.15) (p = 0.03) for patient-related burnout. Caring for patients with suspected COVID-19 infection showed significant multivariable results (p = 0.04) only for personal burnout. Twenty distinct work stressors revealed multivariable associations with CBI. Systematic examination via PUBMED, CINAHL and OVID Medline yielded 33 publications mitigating those stressors among physicians. Adequate staffing was pivotal. Clerical staff off-loaded administrative burden. Information-technology staff helped diminish interruptions, enhancing workflow. Cross-coverage reduced time constraints, ensured separate periods for non-clinical tasks, and ≥1 work-free day/week. Several interventions impacted physician burnout, as did recognition of physicians' efforts/achievements. Other OSI-identified stressors were insufficiently examined in intervention studies: e.g. vacation; appropriately-timed, cross-covered restbreaks; and counter-measures for emotionally-disturbing aspects of MD's work, particularly during the pandemic.

Conclusions: Further participatory-action research is needed in well-controlled intervention trials to alleviate physician burnout.