National Survey on Burnout in Pediatric Residents

1. Burnout in Pediatric Residents A National Survey to Inform Future Interventions Maneesh Batra, John Mahan, Charles Schubert, Paria Wilson, Betty Staples, Janet Serwint, Hilary McClafferty, Alan Schwartz, Kathi J. Kemper, For the Pediatric Resident Burnout – Resilience Study Consortium

2. All authors have documented that we have no financial relationships to disclose or Conflicts of Interest (COIs) to resolve

3. Background Resident burnout has been associated with: • Poor resident health • Impaired resident performance • Suboptimal patient care • Decreased patient satisfaction Prins, Med Teach 2007; Wallace, Lancet 2009; Dyrbye, JAMA 2010; Baer, Pediatrics 2017

4. Background • Prevalence of burnout among peds residents: 46-74% • Limited by small sample sizes, low response rates, and minimal assessment of demographic characteristics and risk factors/impacts • Less is known about combined (e.g., Med-Peds) • Knowledge gaps remain: • National prevalence, distribution and natural history • Understanding individual and institutional protective/risk factors to target future interventions McLafferty Pediatrics 2014; Pantaleoni, AcadPediatr2014; Dyrbye, Acad Med 2014, Baer Pediatrics 2017

5. Objectives • To describe the prevalence of burnout among pediatric and medicine-pediatrics residents in a national sample • To characterize potential personal and programmatic risk/protective factors for burnout

6. Pediatric Resident Burnout-Resilience Study Consortium (PRB-RSC) • Established in 2015 • Goals: • To improve pediatric resident resilienceand attributes of wellness • Describe the epidemiology and relationships between burnout, resilience, and attributes of wellness • Test interventions that positively impact burnout, resilience, and attributes of wellness

7. Participating Institutions • 34 institutions • 22 of which included peds and med-peds programs • Representative of region and size

8. Methods • Anonymous, online survey of all PRB-RSC residents via APPD LEARN from April-June of 2016 • No exclusion criteria • IRB approval at each individual institution • Programs encouraged to employ their own recruitment strategies (including incentives) • Engagement of program leadership and commitment to address wellness • Program level incentive offered by PRB-RSC to top 3 recruiting programs

9. Survey Tool • Total of 141 items, 12-15 min to complete • Demographics & Residency Characteristics Age, Race Gender Marital status Debt • Year of training • Proximity to weekend and vacation • Type of rotation • Recent experiences (e.g., night call, patient death) • Attributes of Wellness • Burnout • Resilience • Stress • Mindfulness • Self-compassion • Empathy • Sleepiness • Career satisfaction

10. Maslach Burnout Inventory • Considered criterion standard for measuring burnout • Validated 22-item questionnaire • 3 subscales: Depersonalization (DP), Emotional Exhaustion (EE), Personal Accomplishment (PA) • By convention, high score in DP (≥ 27) or EE (≥10) is considered ‘burned out’

11. Analyses • Rates of burnout were calculated by: • Residency type (peds or med/peds) • Residency year • Institution • To account for clustering of learners in programs, mixed-effects logistic regression models were used to predict factors associated with burnout

12. Response Rates • Overall response rate: 62% (1693/2723)

13. Demographics • No differences between residents who were vs. were not burned out

14. Prevalence of Burnout • Overall prevalence of burnout: 56% • Did not differ by program size

15. Burnout by Year of Training

16. Resident Experiences and Burnout

17. Resident Personal Attributes and Burnout

18. Associations with Burnout

19. Limitations • Program participation in PRB-RSC voluntary • Dependent on self-reporting • Cross-sectional study design

20. Conclusions • Overall rates of burnout in this nationalsample are high • Time for action • Rate of burnout varied markedly by program, program type, and year of training • Unclear if this is due to cohort effect, or year-of training effect • Several personal and programmatic characteristics are attractive targets for interventions to prevent and/or reduce burnout • PRB-RSC is a platform to address key issues of burnout and wellness

21. Participating Institutions 2016 and 2017 U of California Los Angeles/Mattel Children’s U of California San Diego/Rady Children’s U of Chicago/ Corner Children’s Hospital U of Cincinnati/Cincinnati Children’s U of Colorado/Denver Children's U of Florida/Shands Hospital U of Illinois- Chicago U of Kansas U of Louisville U of Michigan U of Minnesota U of New Mexico U of Oklahoma- Oklahoma City U of Oklahoma- Tulsa U of Pennsylvania/Children's Philadelphia U of Pittsburgh U of Rochester U of South Alabama U of Washington/Seattle Children's U of Wisconsin Virginia Tech Carilion School of Medicine Weill Cornell Medical Center Wright State U Albert Einstein College of Medicine Baylor College of Medicine/Texas Children’s Beth Israel Medical Center- Newark Boston Children’s Hospital Carolinas Medical Center Case Western/Rainbow Babies and Children's Crozer-Chester Medical Center Duke U Inova Fairfax Medical Campus Jefferson Medical College/duPont Hospital for Children Johns Hopkins U Louisiana State U Maine Medical Center Mayo Clinic College of Medicine Medical College of Wisconsin Northwestern U/Lurie Children's Ohio State U/Nationwide Children's Rush U Tufts Medical Center Tulane University U of Alabama U of Arizona U of California Davis

22. http://pedsresresilience.com

24. Resident Experiences and Burnout

25. Other PRB-RSC Projects • Longitudinal Study of Resident Burnout • For 2017 survey, expect 46 institutions • Resident burnout and competency (milestones) • Other outcomes within PRB-RSC dataset • Advocacy

26. Future directions • Longitudinal Study of Resident Burnout • For 2017 survey, expect 46 institutions • Faculty Survey • Pediatric Experience Collaborative • International Comparisons • Karolinska Institute, Sweden • Interventional Trials

27. PRB-RSC program reports

28. PRB-RSC program reports

29. Prevalence of burnout by type