Wilper AP, Weppner WG, Smith CS. Changes in Idaho Primary Care Physician Clinical Work Hours, 1996-2009. JAMA. 2010;304(13):1443-1445. doi:10.1001/jama.2010.1403
To the Editor: A study by Staiger et al1 described a 5.7% decrease in US nonresident physician work hours between 1996 and 2008. Because the data precluded analysis by specialty and identification of a source work hour loss, more specific data on Idaho primary care physicians (PCPs) were examined.
Survey data were analyzed for all Idaho PCPs (defined as general practitioners and family physicians [combined for these analyses], general internists, and pediatricians) between 1996 and 2009. In Idaho, all PCPs are interviewed every 3 to 5 years, with state workers identifying physicians using the Idaho Board of Medicine and other publicly available sources. Physicians or office managers completed a questionnaire on practice characteristics. Residency programs identified current trainees. Veterans Affairs–based physicians were excluded from the analysis. Population statistics were obtained from the US Census Bureau.2,3
The survey questions were, “How many hours per week is the practitioner engaged in outpatient (office visits) care activities at this location? How many hours a week does the practitioner work providing hospital inpatient care?” and included practice location. Changes in work hours for inpatient and outpatient care as well as patient waits for nonurgent clinic appointments were analyzed. Only respondents with complete data for the variable of interest were included in each analysis.
Linear regression was used to measure the relationship between wave of interview and change in work hours or appointment waits. Because of data skewing, log transformations were used to normalize data, which were tested for proportional trends. For weeks worked yearly, tests were performed directly on weeks and linear trends reported in hours. Time was analyzed as an independent continuous variable. Significance was defined as a 2-sided P < .05. SAS version 9.2 (SAS Institute, Cary, North Carolina) was used for analyses. The University of Washington institutional review board approved this study, and oral participant consent was provided at the time of the survey.
State workers attempted to interview 2084 PCPs in 3 waves between 1996 and 2009 and completed 2046 surveys, for a response rate of 98.2%. Physician characteristics are shown in Table 1. The number of PCPs increased from 602 between 1996 and 2000 to 752 between 2005 and 2009. Because of inconsistent work hour reporting, resident physician data were excluded from subsequent analyses.
Overall weekly clinical work hours decreased from a mean of 47.7 hours (95% confidence interval [CI], 46.4-49.1) in 1996-2000 to 38.9 hours (95% CI, 38.0-39.9) in 2005-2009 (P < .001) (Table 2). There were significant decreases for family medicine/general practice and internal medicine, but not for pediatrics. Number of weeks worked yearly remained stable, from 48.0 weeks (95% CI, 47.7-48.4) in 1996-2000 to 47.8 weeks (95% CI, 47.6-48.1) in 2005-2009 (P = .38). Patient waits for nonurgent appointments increased (Table 2). Over the study period, the population of Idaho increased 30.2%, from 1 187 706 to 1 545 801.
Clinical hours worked by PCPs in Idaho decreased between 1996 and 2009 despite an increase in the state population, while patient waits increased. The total and percentage decrease in work hours was larger than that documented by Staiger et al.1 Primary care physicians may be performing more nonclinical functions, such as administration or management duties, or may be actually working less overall. If national trends are similar, it is plausible that PCP changes were largely responsible for the observed decrease in physician work hours, and that access to these physicians' services declined as a result.
These results should be considered with the study limitations. Data are based on self-report, although there is no reason to believe that the accuracy would have varied systematically over the study period. Idaho is rural, with relatively few PCPs per capita and little managed care, and may not be representative of other states.6,7 There may have been too few pediatricians in the study to detect a significant change in work hours. Research with nationally representative data is needed to confirm these findings.
Author Contributions: Dr Wilper 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: Wilper, Smith.
Acquisition of data: Wilper.
Analysis and interpretation of data: Wilper, Weppner.
Drafting of the manuscript: Wilper.
Critical revision of the manuscript for important intellectual content: Weppner, Smith.
Statistical analysis: Wilper, Weppner, Smith.
Obtained funding: Wilper.
Administrative, technical, or material support: Wilper.
Financial Disclosures: None reported.
Funding/Support: The Mountain States Tumor and Medical Research Institute and the Idaho Office of Rural Health and Primary Care provided financial support. The Boise Veterans Affairs Medical Center provided resources and the use of facilities.
Role of the Sponsor: The funding organizations had no role in the design and conduct of the study; in the collection, analysis, and interpretation of the data; or in the preparation, review, or approval of the manuscript.
Additional Contributions: Laura Rowen, MPH (Idaho Office of Rural Health and Primary Care), provided assistance with survey analysis, and Rick Tivis, MPH (Statistically Speaking), provided help with programming and statistical analysis. Mr Tivis was compensated for his assistance.