The authors’ calculations were based on the Integrated Public Use Microdata Series Current Population Survey basic monthly data from January 2019 to December 2020. P values < .10 for July and October 2020. P values < .05 for March, April, August, and September 2020. P values < .01 for May, June, November, and December 2020. Error bars denote 95% CIs, which accounted for the use of sampling weights and for clustering at the individual level. See eAppendix 2 in the Supplement for additional notes.
eAppendix 1. IPUMS CPS and CPS Response Rates
eAppendix 2. Additional Notes for Figure
eAppendix 3. Additional Notes for Table
Customize your JAMA Network experience by selecting one or more topics from the list below.
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.
Hu X, Dill MJ. Changes in Physician Work Hours and Patterns During the COVID-19 Pandemic. JAMA Netw Open. 2021;4(6):e2114386. doi:10.1001/jamanetworkopen.2021.14386
The COVID-19 pandemic has been associated with loss of revenue, reduced work hours, and reduced earnings for physicians in the United States.1-3 Furthermore, pandemic restrictions and related regulatory changes allowing physicians greater flexibilities potentially altered physicians’ work activities and environments.4,5 We analyzed a longitudinal data set to examine changes in US physician work hours and activities before and after the COVID-19 pandemic emerged.
We analyzed the Current Population Survey (CPS) basic monthly data from January 2019 to December 2020, obtained through the Integrated Public Use Microdata Series (IPUMS). Administered by the US Bureau of the Census, the CPS contains demographic and employment information. The response rate ranged between 65% and 83%. (See eAppendix 1 in the Supplement for detailed information about the data.) Participants were interviewed 8 times during a 16-month period, allowing for longitudinal analyses.
This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline. This study was deemed exempt from review by the American Institute for Research institutional review board because it used publicly available survey data and was not human subjects research in accordance with 45 CFR §46. Participants gave consent prior to each interview.
First, we estimated panel regression models with individual- and time- fixed effects (year and month) using linked person-level data. This controlled for time-invariant factors (eg, sex, race/ethnicity, medical specialty) and minimized potential bias due to a decrease in survey response rate. Time-invariant factors, whether available in the survey (eg, sex, race/ethnicity) or not (medical specialty), were not included in the statistical model as independent variables. Next, we used pooled data to compare pre–COVID-19 (January 2019 to March 2020) and during COVID-19 (April to December 2020) mean percentages of physicians working full-time, performing the same activities, and laid off. Additionally, we compared mean percentages of parents of preschool-aged children (younger than 5 years of age) among both male and female physicians. Survey weights were applied in all analyses, and regression models were clustered at the person level. Comparisons were tested with simple linear regression, and P < .10 was considered statistically significant. Analyses were performed using Stata version 14.1 (StataCorp) in February 2021.
Our data set included 8853 observations of 2563 unique physicians (weighted demographic data were 5311 male physicians [60%] and 6230 White physicians [70.4%]; weighted mean [SD] age was 46.7 [13.4] years). In January 2019, mean weekly hours worked per week by physicians was 50.8 (95% CI, 47.7-51.8 hours per week). In March 2020, this began to decrease (49.2 hours per week; 95% CI, 48.0-50.3 hours per week; P = .05) and reached the lowest point in May 2020 (47.5 hours per week; 95% CI, 46.1-48.8 hours per week; P < .001) (Figure). Work hours stabilized in the summer before reaching another low in November. Mean weekly work hours in December 2020 were 47.8 (95% CI, 46.6-49.1 hours per week; P < .001), an approximate 6% decrease from January 2019.
The percentage of physicians reporting full-time work status declined from 84.17% (95% CI, 83.13% to 85.22%) before COVID-19 to 80.65% (95% CI, 79.03%-82.27%) (P < .001) during the pandemic, and those who claimed to “still have the same activities” at work declined from 83.90% (95% CI, 82.65% to 85.14%) to 78.00% (95% CI, 76.01% to 79.98%) (P < .001) (Table). Although rare, the percentage of physicians reporting being laid off increased from 0.05% (95% CI, −0.01% to 0.11%) to 0.45% (95% CI, 0.21% to 0.70%) during COVID-19 (P < .001). The percentage of parents of preschool-aged children among full-time, female physicians decreased from 17.98% (95% CI, 16.05% to 19.91%) to 14.10% (11.59% to 16.60%) (P = .009) and did not significantly change among male physicians.
This study found that physicians’ work hours have significantly decreased since the start of the COVID-19 pandemic in the US, accelerating an existing, gradual decline.6 There was also a decreased percentage of physicians working full-time, a rise in the percentage who were laid off, and increased changes in physicians’ usual activities. These observed changes may reflect the decrease in health care utilization1 and the increased flexibilities instigated by COVID-19–driven regulations.4 The decline in the percentage of parents with preschool-aged children among only female physicians may suggest a disproportionate uptake of childcare responsibilities among female physicians.
This study leveraged the timely, longitudinal design of CPS data to provide early evidence of changes in physician work hours and patterns. However, this study also had limitations: the CPS sample size of physicians was small and lacked detailed information. Future studies analyzing physician-specific data are needed to fully examine the impact of the pandemic on the physician workforce.
Accepted for Publication: April 19, 2021.
Published: June 23, 2021. doi:10.1001/jamanetworkopen.2021.14386
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2021 Hu X et al. JAMA Network Open.
Corresponding Author: Xiaochu Hu, PhD, Association of American Medical Colleges, 655 K St, NW, Washington, DC 20001 (firstname.lastname@example.org).
Author Contributions: Dr Hu and Mr Dill had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Hu.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: All authors.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Hu.
Administrative, technical, or material support: All authors.
Supervision: All authors.
Conflict of Interest Disclosures: None reported.
Additional Contributions: The authors would like to express our appreciation to numerous colleagues at the Association of American Medical Colleges: Sarah Conrad, MS, for her excellent support in editing; and Merle Haberman, MHA, and Janis M. Orlowski, MD, MACP, for their insightful input in reviewing the drafts. The authors would also like to acknowledge Douglas O. Staiger, PhD, Dartmouth College, for his valuable suggestions on methods and data. They received no compensation for their roles in this study.
Additional Information: Data used in this study are free of charge.