Background
The American Society of Hematology and researchers at the Fitzhugh Mullan Institute for Health Workforce Equity at the George Washington University are conducting a 3-year study of the hematology workforce to understand factors that influence the supply of hematology services in the U.S. The Survey of Practicing Hematologists and Oncologists focuses on practicing hematology/oncology physicians' practice activities and experiences, compensation, job satisfaction and burnout. While the prevalence and predictors of burnout in the oncology workforce have been studied in detail previously, less is known about factors associated with burnout in the hematology or combined hematology/oncology workforce--or whether these differ across academic and community practice settings. This study seeks to examine factors associated with severe burnout for hematology/oncology physicians in academic and community practice settings using data from a large-scale, comprehensive survey of hematology/oncology physicians.
Methods
We collected survey data via mail and online survey (using Qualtrics, an online survey tool) in April through June 2019. The survey included questions about hematology/oncology physicians' work hours, practice activities, compensation, job satisfaction and burnout.
This analysis uses data from a single validated question examining respondents' level of burnout: "Overall, based on your definition of burnout, how would you rate your level of burnout?" The question asks respondents to rank their burnout on a 5-point Likert scale. We collapsed responses into a dichotomous variable indicating severe burnout (=4 or 5 on the Likert scale). We used two weighted multiple logistic regression models to examine associations between severe burnout and work hours, practice characteristics and activities, and type of compensation for respondents in academic and community practice in Stata 15, controlling for demographics and type of practice (p<0.05=statistically significant).
Results
A total of 675 hematologists/oncologists completed the survey (27% response rate). Of these, 427 respondents reported working in academic or community practice and had complete data to be included in the analysis: 162 (38%) in academic practice settings and 265 (62%) in community practice settings. Respondents in academic practice settings were slightly less likely to report experiencing severe burnout than those in community practice settings (9% [15/162] vs. 12% [34/265], p=0.26). In the logistic regression models, we found statistically significant and positive association between severe burnout and Relative Value Unit or RVU-based compensation (vs. salaried or other compensation models) for both academic (OR=18.42, p<0.01) and community practice respondents (OR=3.05, p<0.01). We also found a significant and positive association between severe burnout and being female for respondents in academic practice only (OR=6.07, p<0.01). We found a significant and negative association between severe burnout and often working with advanced practice providers (nurse practitioners and/or physician assistants) for respondents in community practice only (OR=0.32, p<0.01).
Conclusions
Study findings suggest that severe burnout rates are similar for hematology/oncology physicians in academic and community practice settings. Severe burnout appears to be related to use of RVU-based compensation systems in both academic and community practices, suggesting that these models may require major revision to reduce burnout and support the health and longevity in practice of hematology/oncology physicians. Improving access to advanced practice providers may mitigate severe burnout, especially in community practice settings. Higher levels of burnout among women in academic hematology/oncology practices suggest an area for further research into possible explanations and solutions. These findings merit further exploration, particularly given the increased pressures on physicians in the era of COVID-19.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal