Figure. Adjusted mean differences in annual work hours for physician specialties vs family practice in community tracking survey (n=6381). Note: Estimates from a multiple linear regression that statistically adjusted for physician, practice, and market characteristics (see text for details). Error bars represent 95% confidence intervals.
Leigh JP, Tancredi D, Jerant A, Kravitz RL. Annual Work Hours Across Physician Specialties. Arch Intern Med. 2011;171(13):1211-1213. doi:10.1001/archinternmed.2011.294
Author Affiliations: Center for Healthcare Policy and Research, Departments of Public Health Sciences (Dr Leigh), Pediatrics (Dr Tancredi), Family and Community Medicine (Dr Jerant), and Internal Medicine, Division of General Medicine (Dr Kravitz), University of California Davis School of Medicine, Davis.
Debate surrounds the relative pay of physicians in various specialties. Several studies have addressed income differences.1 Few, by comparison, have addressed work hours, and we are not aware of any that consider annual work hours. Yet, work hours—independent of income—figure prominently in discussions of physician lifestyles,2 student choices,3 and patient safety.4
Data were drawn from a nationally representative sample of physicians in the 2004-2005 Community Tracking Survey (CTS). Our subsample contained 6381 physicians self-reporting 20 to 100 weekly work hours and at least 26 weeks worked annually.1 Work hours included all medically related activities (direct patient care, administrative tasks, and professional duties).1 Annual work hours were calculated as (weekly work hours) × (weeks worked per year).
We analyzed 41 specialties with at least 20 respondents, as well as 4 broad-specialty categories: primary care, surgery, internal medicine, and pediatric subspecialties, and other specialties.1 Control variables were age, sex, race, whether board certified, whether graduated from foreign medical school, residence in areas with less than 200 000 population, region of residence, practice ownership, academic employment, and revenue from managed care.
Analyses were conducted using Stata version 11 (StataCorp, College Station, Texas), accounting for the survey design. Two regression analyses compared work hours among 41 specialties, with and without control variables, with family medicine as reference. Two additional regressions compared the 4 broad categories, with and without control variables, with primary care as reference.
Mean annual hours worked was 2524 (median, 2420; interquartile range, 1960-2940). The Figure presents selected results of a linear regression of annual hours on the 41 specialties adjusted for control variables. Compared with family practice, the 3 highest specialties were vascular surgery, critical care internal medicine, and neonatal and perinatal medicine; the 3 lowest were pediatric emergency medicine, occupational medicine, and dermatology.
In the regression for the 4 broad-specialty categories with control variables, surgery (+303 hours; 95% confidence interval [CI], 219 to 387 hours) and internal medicine and pediatrics subspecialties (+208 hours; 95% CI, 132 to 284 hours) had significantly higher hours, and other medical (−228 hours; 95% CI, −295 to −161 hours) had significantly lower hours than primary care. Control variables did not materially alter any of the rankings in any regression. Excluding physicians classified as working part time (mean, <32 h/wk) did not affect relative rankings in any regression.
We found that specialists caring for more acutely ill patients or those requiring intensive monitoring (usually in hospital settings) work longer hours than physicians focused on more stable, chronically ill patients (mostly in ambulatory settings). The exceptions were physicians practicing emergency medicine or hospital medicine. Both of these specialties are characterized by fixed, hourly shifts; although patient acuity may be high, the number of work hours per day and days per month are limited.
Our rankings are somewhat similar to studies of annual income, with procedural specialties being paid more than cognitive specialties.1,2 But there are differences. Neurological surgery receives the highest statistically significant wage, yet is not significant in the hours rankings.1 Dermatologists have significantly higher wages and significantly lower hours.1 Family practice receives one of the lowest wages but is near the middle of the hours rankings.1 Finally, among broad categories, primary care receives the lowest wage but is neither the highest nor lowest for hours.
Specialties with more (less) work hours tend to have relatively low (high) physician job satisfaction ratings. For example, pediatricians, dermatologists, and child and adolescent psychiatrists reported relatively low hours and have relatively high career satisfaction.5 Similarly obstetrician and gynecologists reported relatively high hours and have relatively low career satisfaction. However, this relationship does not always hold; for example, neonatologists and perinatologists reported high average hours yet have high career satisfaction.5
Our study had limitations. The CTS excluded radiologists, anesthesiologists, and pathologists. Self-reported work hours did not capture variability across day, swing, or night shifts or for weekends or weekdays, nor were hours-on-call included. Finally, the CTS have data from 2008, the only wave after 2004-2005. The CTS administrators, however, warn against comparing the 2008 data to studies using 2004-2005 data.1,5
We ranked 41 specialties and 4 broad categories by annual work hours. We believe this ranking will likely be useful to medical students, residency directors, hospital administrators, physicians contemplating switching specialties, and policy makers.
Correspondence: Dr Leigh, Department of Public Health Sciences, UC Davis Medical School, 1 Shields Ave, MS1C, Davis, CA 95616-8638 (email@example.com).
Author Contributions: Dr Leigh had full access to all of 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: Leigh, Tancredi, and Kravitz. Acquisition of data: Leigh, Tancredi, and Kravitz. Analysis and interpretation of data: Leigh, Tancredi, and Jerant. Drafting of the manuscript: Leigh. Critical revision of the manuscript for important intellectual content: Leigh, Tancredi, Jerant, and Kravitz. Statistical analysis: Leigh and Tancredi. Obtained funding: Leigh. Administrative, technical, and material support: Leigh, Jerant, and Kravitz. Study supervision: Leigh.
Financial Disclosure: None reported.
Funding/Support: This study was funded in part by grants to Dr Leigh from the National Institute for Occupational Safety and Health (grant 1 R01 H008248-01) and Drs Tancredi and Kravitz from the UC Davis Office of the Vice Chancellor for Research.
Role of the Sponsors: The funding sources had no involvement in the design and conduct of the study; collection, management, analysis, and interpretation of the data; or preparation, review, or approval of the manuscript.