eTable. HCPCS Codes Used to Identify Practice Settings
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
Ryskina KL, Shultz K, Unruh MA, Jung H. Practice Trends and Characteristics of US Hospitalists From 2012 to 2018. JAMA Health Forum. 2021;2(11):e213524. doi:10.1001/jamahealthforum.2021.3524
Recent decades have observed a rapid growth in the number of hospitalists.1,2 Receiving care from hospitalists has been associated with shorter length of stay and similar outcomes3 and lower readmission and mortality rates compared with care from nonhospitalist physicians who were not the patient’s primary care physician.4 Furthermore, hospitalist experience (years in practice) has been inversely correlated with patient mortality.5 Surveys indicate high turnover among hospitalists compared with other specialties,6 but those surveys measure turnover within hospitalist groups rather than the specialty overall. Our objective was to characterize hospitalists who shift practice to settings outside of the hospital.
For this cohort study, we used Medicare’s Provider Utilization and Payment Data records from 2012 to 2018, which aggregate Medicare Part B billings by service. This database was linked to physician demographic details and practice characteristics from the Medicare Data on Provider Practice and Specialty files and to county-level characteristics from the Area Health Resource Files. Generalist physicians (internal medicine, geriatrics, general practice, pediatrics, or family medicine) who billed at least 90% of visits from the hospital were considered hospitalists.4,5 Physicians with 100 visits or fewer per year were excluded.
Hospitalists were identified in 2012, and those who continued to practice as generalists were followed up through 2018 to examine the settings where they provided care. We measured the proportion of visits in the following settings (service codes in eTable in the Supplement): (1) acute care hospital; (2) office (ie, a location where patients are treated on an ambulatory basis); (3) nursing home or skilled nursing facility; and (4) other (all other settings).
We examined physician characteristics (gender, age, time from medical school graduation, medical school rank), practice characteristics (size of group, Medicare beneficiary volume), and county characteristics (urban location, region, household income, and the proportion of the population who were White). The data sets we used do not contain information on physician race and ethnicity. The characteristics were selected a priori based on availability and relevance in physician workforce literature. We compared the characteristics between physicians who continuously practiced as hospitalists vs those who changed practice in any year by using the χ2 test for categorical and the Wilcoxon test for continuous variables. We used multivariable linear probability regression to measure the association between the characteristics and continuous hospitalist practice. We also performed a sensitivity analysis, including in our sample the hospitalists who did not bill Medicare or subspecialized in 2013 to 2018.
Data analyses were conducted from March 15 to June 28, 2021. Two-sided P values <.05 were considered significant. Analyses were conducted using SAS, version 9.4 (SAS Institute Inc). The University of Pennsylvania Institutional Review Board waived review of this study and the need for informed consent because the data used were publicly accessible. We followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cohort studies.
Of the 16 985 hospitalists identified in 2012 (mean [SD] age, 41.1 [8.1] years; 6146 women [36.2%] and 10 839 men [63.8%]), 4052 (23.9%) shifted practice to another setting in at least 1 year during the follow-up period. Those physicians tended to be in smaller practices (eg, compared with solo practice, the regression coefficient for practices with ≥300 physicians was 22.06; 95% CI, 18.60-25.52; P < .001) and/or rural practices (regression coefficient for urban vs rural, 3.40; 95% CI, 1.28-5.53; P = .002) (Table 1). Physicians who changed practice were more likely to split time between a hospital and other settings than to practice predominantly outside of the hospital; eg, after hospitalists, the second largest category (601 of 16 985 [3.5%]) was physicians who provided 50% or more but less than 90% of visits in the hospital, with the other visits provided in other settings (Table 2).
Our sensitivity analyses that included physicians who did not bill Medicare (n = 2994) or subspecialized (n = 2644) in any year during study follow-up were generally consistent with the primary analyses. In addition to smaller practice size and rural location, physicians who shifted practice were more likely to be aged 60 years or older (7.1% vs 2.4%, P < .001) and less likely to come from a top 20 medical school (3.6% vs 4.6%, P < .001) compared with physicians who continuously practiced as hospitalists, which were not statistically significant coefficients in our primary analysis.
This study found that between 2012 and 2018, nearly 1 in 4 hospitalists shifted practice to other settings, although most hospitalists continued to practice in the hospital at least part time. Our findings have important implications for hospitals, especially during infectious disease outbreaks, such as the COVID-19 pandemic, that place extra demands on hospital-based clinicians. Understanding the drivers behind these trends may help mitigate the high costs of physician turnover in health systems.
Limitations of this study include the use of Medicare fee-for-service claims but not claims for other payers. However, Medicare is the most common payer for inpatient care in the US.
Accepted for Publication: September 10, 2021.
Published: November 5, 2021. doi:10.1001/jamahealthforum.2021.3524
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2021 Ryskina KL et al. JAMA Health Forum.
Corresponding Author: Kira L. Ryskina, MD, MS, 423 Guardian Dr, 1230 Blockley Hall, Philadelphia, PA 19104 (email@example.com).
Author Contributions: Dr Ryskina 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.
Concept and design: Ryskina, Unruh, Jung.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Ryskina, Shultz, Jung.
Critical revision of the manuscript for important intellectual content: Ryskina, Unruh, Jung.
Statistical analysis: Shultz, Unruh, Jung.
Administrative, technical, or material support: Ryskina.
Conflict of Interest Disclosures: None reported.
Funding/Support: This study was supported in part by the National Institute on Aging Career Development Awards (K08AG052572 to Dr Ryskina and 5K01AG057824 to Dr Jung). Drs Jung and Unruh were supported, in part, by The Physicians Foundation Center for the Study of Physician Practice and Leadership of Weill Cornell Medical College.
Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.