Dyrbye LN, West CP, Satele D, Sloan JA, Shanafelt TD. Work/Home Conflict and Burnout Among Academic Internal Medicine Physicians. Arch Intern Med. 2011;171(13):1207-1209. doi:10.1001/archinternmed.2011.289
Author Affiliations: Departments of Medicine (Drs Dyrbye, West, and Shanafelt) and Health Sciences Research (Mr Satele and Dr Sloan), Mayo Clinic, Rochester, Minnesota.
Studies suggest that work/home conflict may have a central role in physicians developing burnout.1- 5 In a recent comprehensive evaluation of a wide variety of personal and professional factors hypothesized to contribute to burnout in 7905 American surgeons, 3 factors—hours worked per week, experiencing a work/home conflict within the last 3 weeks, and how the most recent work/home conflict was resolved—were independently associated with burnout.5 To validate the importance of these factors to physician burnout and to explore whether they are relevant to physicians in specialties other than surgery, we assessed their importance in a large sample of internal medicine physicians at an academic center.
In the fall of 2009, all faculty physicians in the Mayo Clinic Department of Medicine received a survey with items pertaining to demographics, work characteristics, and experience of work/home conflict, including whether this was resolved in favor of work or home responsibilities or in a manner meeting both responsibilities. These questions derived from the prior study of American surgeons.5 Burnout was measured using 2 items derived from the Maslach Burnout Inventory (MBI) shown to stratify risk of burnout in multiple independent samples of physicians and medical students including more than 10 000 participants.6 Report of feeling “burned out from my work” at least weekly has a positive likelihood ratio of 14.9 for a high emotional exhaustion score on the full MBI and an area under the curve of 0.94 relative to the full MBI.6 Reporting that “I’ve become more callous toward people since I took this job” at least weekly has a positive likelihood ratio of 23.4 for a high depersonalization score on the full MBI and an area under the curve of 0.93 relative to the full MBI. Consistent with prior literature,7 participants indicating they experienced symptoms in either domain at least weekly were considered to have at least 1 symptom of burnout.
Multivariate logistic regression models were used to identify and evaluate the relative strength of independent associations of demographics, work-related characteristics, experience of a work/home conflict, and how the last such conflict was resolved with burnout. All statistical analyses were done using SAS version 9 (SAS Institute Inc, Cary, North Carolina). The Mayo Clinic Institutional Review Board approved this study.
Of 566 eligible physicians, 465 (82.2%) returned surveys. Demographics, work characteristics, burnout, and work/home conflict results are given in the Table. Physicians who worked more hours reported higher rates of a recent work/home conflict (68.0%, 69.1%, and 82.1% for ≤49 hours, 50-59 hours, and ≥60 hours per week, respectively; overall, P = .01). Frequency of work/home conflicts in the last 3 weeks did not differ by sex (women, 79.0%, vs men, 76.7%; P = .62). Physicians with a work/home conflict in the prior 3 weeks were 50% (P < .001) and 30% (P = .02) more likely to report symptoms of emotional exhaustion or depersonalization at least weekly, respectively, than those who did not have a work/home conflict.
In multivariable analyses, each additional hour worked per week (odds ratio, 1.02 [95% confidence interval, 1.00-1.03] per hour worked), experience of a work/home conflict in the last 3 weeks (odds ratio, 2.09 [95% confidence interval, 1.10-3.97]), and resolving the most recent work/home conflict in favor of work (odds ratio, 1.88 [95% confidence interval, 1.13-3.12]) were independently associated with increased odds burnout. Sex, weeks of hospital service, weeks of consult service, in house overnight duty, and number of times called in to the hospital in the last year were not associated with burnout in these models.
This study of 465 academic general and subspecialty internists demonstrates that work hours, work/home conflicts, and how such conflicts are resolved are strongly related to physician burnout. These results validate findings from our previous comprehensive study of factors associated with burnout among 7905 US surgeons that relied on the full MBI and a comprehensive inventory of personal and professional characteristics.5 Although the groups in these 2 studies were disparate, the same 3 factors (hours worked per week, work/home conflict in the last 3 weeks, and resolving the last work/home conflict in favor of work) remained independent factors associated with burnout in multivariable models in both samples with strikingly similar odds ratios.5 These findings suggest that work/home conflict and how that conflict is managed may be central factors for physician burnout in a variety of practice settings.
Limitations of this study include its cross-sectional design and inclusion of a limited number of personal and professional variables. Because the study was conducted at a single academic institution where all physicians are salaried, its generalizability is unknown. However, the results are consistent with our prior large study of American surgeons.5 Additional strengths include a high response rate and involvement of both generalists and subspecialists.
Given the high prevalence of burnout among physicians7 and well-established negative personal and professional consequences,8,9 the observed associations of work hours and work/home conflicts with burnout suggest possible targets for action.
Correspondence: Dr Dyrbye, Department of Medicine, Mayo Clinic, 200 First St SW, Rochester, MN 55905 (email@example.com).
Author Contributions: Dr Sloan had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Dyrbye, West, Sloan, and Shanafelt. Acquisition of data: Dyrbye, West, and Shanafelt. Analysis and interpretation of data: Dyrbye, West, Satele, Sloan, and Shanafelt. Drafting of the manuscript: Dyrbye, Satele, and Sloan. Critical revision of the manuscript for important intellectual content: Dyrbye, West, Satele, Sloan, and Shanafelt. Statistical analysis: West, Satele, and Sloan. Obtained funding: Shanafelt. Administrative, technical, and material support: Dyrbye, West, and Shanafelt. Study supervision: Dyrbye and Shanafelt.
Financial Disclosure: None reported.
Funding/Support: This work was supported by the Mayo Clinic Department of Medicine Program on Physician Well-being.
Role of the Sponsors: The sponsors had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; or preparation, review, or approval of the manuscript.