Cull WL, Frintner MP, Starmer A, Freed G, Byrne B

Accepted for 2020 Pediatric Academic Societies Annual Meeting

Background: Cross-sectional studies find roughly a quarter of pediatricians work part-time or reduced-hours positions. These studies are unable to examine individual changes in work hours across a pediatrician's career.

Objective: Use the unique capabilities of a longitudinal design to 1) examine pediatrician work hours across a 7-year period and 2) identify associations of changes in work hours with various aspects of pediatrician's life and career.

Design/Methods: Data from a national, longitudinal study, AAP Pediatrician Life and Career Experience Study (PLACES), were used to examine pediatrician self-reported work hours each year from 2012 to 2018. Participants graduated residency training between 2002 and 2011 (N=1804; 41% enrollment), and 1782 (99%) responded in multiple years and were included in the analyses. Pediatrician weekly work hours were categorized for analysis in 5-hour increments. Additional data included 12 measures in 4 domains: work-life balance, financial, health, and satisfaction. For each measure, we reported frequency percentages averaged across study years. We conducted mixed-effects logistic regression for longitudinal analysis for each measure to examine associations with a 5-hour change in work hours for an individual, controlling for survey year, gender, graduation year, medical school location, and subspecialty fellowship training.

Results: Over a third (38%) of early to mid-career pediatricians reported similar work hours in 2012 and 2018, while 27% reported increases and 35% reported decreases. At both the start and end of the study period, roughly half (2012: 50.9%; 2018: 49.3%) reported working more than 40 hours a week, and less than 10% reported working over 60 hours per week (2012: 7.8%; 2018: 4.2%). In the longitudinal models, a pediatrician working 5 additional hours per week was associated with greater income (aOR = 1.14; 95% CI 1.04-1.25) but also associated with lower reported values in work-life balance, health, and satisfaction measures (Table). For example, a 5-hour increase in work hours was associated with lower satisfaction with time for personal interests (aOR = .73; .69-.78), greater agreement that professional responsibilities interfere with personal responsibilities (aOR = 1.42; 1.34-1.49), increased reports of being anxious or nervous (aOR = 1.11; 1.04-1.19), and lower satisfaction with life as a whole (aOR = .81; .76-.86).

Conclusion: Over a 7-year period, a 5-hour increase in work hours was found to be associated with greater income but also with negative tradeoffs in work-life balance, health, and satisfaction.