physicians

The Human and Professional Costs of Medical Care in the United States ADMINISTRATIVE WORK CONSUMES ONE-SIXTH OF U.S. PHYSICIANS’ WORKING HOURS AND LOWERS THEIR CAREER SATISFACTION Steffie Woolhandler and David U. Himmelstein

Doctors often complain about the burden of administrative work, but few studies have quantified how much time clinicians devote to administrative tasks. We quantified the time U.S. physicians spent on administrative tasks, and its relationship to their career satisfaction, based on a nationally representative survey of 4,720 U.S. physicians working 20 or more hours per week in direct patient care. The average doctor spent 8.7 hours per week (16.6% of working hours) on administration. Psychiatrists spent the highest proportion of their time on administration (20.3%), followed by internists (17.3%) and family/general practitioners (17.3%). Pediatricians spent the least amount of time, 6.7 hours per week or 14.1 percent of professional time. Doctors in large practices, those in practices owned by a hospital, and those with financial incentives to reduce services spent more time on administration. More extensive use of electronic medical records was associated with a greater administrative burden. Doctors spending more time on administration had lower career satisfaction, even after controlling for income and other factors. Current trends in U.S. health policy—a shift to employment in large practices, the implementation of electronic medical records, and the increasing prevalence of financial risk sharing—are likely to increase doctors’ paperwork burdens and may decrease their career satisfaction.

Doctors often complain about the administrative burden they face in practicing medicine, and research has documented the high administrative costs of U.S. medical care (1–3). But few studies have quantified how much time clinicians devote to administrative tasks, factors that influence the time spent on such International Journal of Health Services, Volume 44, Number 4, Pages 635–642, 2014 © 2014, Baywood Publishing Co., Inc. doi: http://dx.doi.org/10.2190/HS.44.4.a http://baywood.com

635

636

/ Woolhandler and Himmelstein

matters, or whether time spent on administration affects career satisfaction. We assessed these questions using data from a national survey of physicians. METHODS We analyzed restricted-use data from the 2008 Health Tracking Physician Survey conducted by the Center for Studying Health System Change. This mail-based survey collected data from a random sample of 4,720 U.S. physicians working at least 20 hours per week in direct patient care. Trainees and federal government employees were excluded, as were radiologists, anesthesiologists, and pathologists. The response rate was 62 percent. Additional details on the survey methodology are available at www.hschange.org. The survey included questions about physician demographics; number of weeks worked; income; satisfaction; practice characteristics such as specialty, group size, practice ownership, and type (e.g., group or staff model HMO, private hospital, solo practice, etc.); physician compensation model; and use of electronic medical records. Our measure of administrative time is based on physicians’ responses to the following questions: “During your LAST COMPLETE WEEK OF WORK, approximately how many hours did you spend in all medically-related activities? . . . Direct patient care includes seeing patients, performing surgery, and time spent on patient record-keeping, patient-related office work and travel time. Your best estimate is fine. a. Hours in direct patient care b. Hours in administrative tasks and professional activities c. Total hours in medically-related activities” We assessed physicians’ career satisfaction using the following question: “Thinking very generally about your satisfaction with your overall career in medicine, would you say that you are currently . . .Very satisfied, Somewhat satisfied, Neither satisfied nor dissatisfied, Somewhat dissatisfied, Very dissatisfied?” We classified both physicians’ incomes and years since graduation from medical school into three categories each (< $150,000, $150,000–$250,000, and > $250,000; and < 10 years, 10–20 years, and > 20 years, respectively). Because clinicians who work for medical schools often spend time administering educational and research programs, we performed sensitivity analyses that exclude medical school faculty. All data were analyzed using SAS Survey procedures that account for stratification and unequal weighting. Multiple linear regression was used to assess

Physicians’ Administrative Work

/ 637

predictors of career satisfaction after control for factors previously reported to predict satisfaction (race, income, specialty). Human subject protections were reviewed and approved by the Hunter College Institutional Review Board. RESULTS Doctors devoted, on average, 8.7 hours each week to administrative work, accounting for 16.6 percent of their total work week. These figures exclude all patient-related record keeping and patient-related office work. In total, patient care physicians spent 168.4 million hours on administration in 2008. Table 1 displays the number of hours and percent of professional time devoted to administration according to physician and practice characteristics. Administration accounted for a larger share of professional time for female physicians (17.5%), those in the Northeast (17.9%), and those with incomes below $150,000 (17.8%). The number of years in practice had no apparent effect on administrative work. Among the various specialties, psychiatrists spent the highest proportion of their time on administration (20.3%), followed by internists (17.3%) and family/ general practitioners (17.3%). Pediatricians spent the least amount of time, 6.7 hours per week or 14.1 percent of professional time. Practice ownership by a hospital was associated with more time spent on administration. Although insurance company ownership did affect the amount of administrative work, financial incentives to reduce services were associated with more time spent on administration. Solo practice was associated with more administrative work, but small group practice was not. Indeed, doctors practicing in large medical groups (> 100 physicians) spent more of their time on administration (19.7%) than those in smaller groups (16.3%). Interestingly, physicians who reported that their practice made extensive use of information technology actually spent more time on administration. While physicians using entirely paper records spent 15.5 percent of their time on administrative matters, administration accounted for 17.2 percent of time for those in practices using entirely electronic medical records and 18.0 percent of time for those using a mix of electronic and paper records. Physicians who spent more time on administration were markedly less satisfied with their careers. Those who were very satisfied spent, on average, 16.1 percent of their time on administration, versus 20.6 percent for those who were very dissatisfied. After controlling for several other factors reported to affect physicians’ career satisfaction (4), the proportion of time spent on administration remained a significant (p = 0.01) predictor of dissatisfaction, along with lower income and specialty other than pediatrics. Being a psychiatrist was modestly

638

/ Woolhandler and Himmelstein Table 1 U.S. physicians’ administrative work: number of hours per week and percent of total professional time Administrative hours per week

Percent of total work week

All physicians

8.7

16.6

Gender Female Male

8.4 8.8

* 17.5 16.3

Region Northeast Midwest South West

9.1 8.5 8.6 8.5

* 17.9 16.0 16.1 16.7

Specialty category Internal medicine Pediatrics Family/general practice Medical specialties Surgical specialties Psychiatry Obstetrics and gynecology Primary care specialty Not primary care

** 9.1 6.7 8.6 9.1 8.5 9.6 8.2 8.4 8.8

** 17.3 14.1 17.3 16.7 15.6 20.3 15.1 16.7 16.6

Main work site Solo practice 2-physician practice 3+-physician practice Group/staff HMO Community health center Government hospital Private hospital Medical school

** 9.3 8.3 7.0 8.3 7.7 9.9 9.1 14.4

** 18.1 15.1 13.4 16.8 17.0 19.7 17.1 27.0

* 9.3 8.2

** 18.1 15.5

Hospital ownership interesta Hospital is full or part owner of practice No hospital ownership


/ 639

Table 1 (Cont’d.) Administrative hours per week

Percent of total work week

8.8 8.5

17.9 16.2

Number of physicians in main practice 1–100 More than 100

** 8.5 10.5

** 16.3 19.7

Personal financial incentives in practice favor . . . Reducing services Expanding services Neither incentive

** 10.4 7.9 8.8

** 19.3 15.2 17.1

** 7.9 9.5 9.2

** 15.5 18.0 17.2

Career satisfaction Very dissatisfied Somewhat dissatisfied Neither satisfied nor dissatisfied Somewhat satisfied Very satisfied

** 12.0 9.2 8.9 8.6 8.1

* 20.6 17.5 16.9 16.4 16.1

Net incomefrom medical practice < $150,000 $150,000–$250,000 > $250,000

8.5 8.8 8.6

** 17.8 16.7 15.1

Years in practice < 10 10–20 > 20

8.7 8.9 8.4

16.6 16.7 16.5

Insurance company ownership interest a Insurance company owns part of practice No insurer ownership

Use of electronic medical record in practice Practice all paper Part electronic, part paper All electronic

*p < 0.01 for differences among sub-categories. **p < 0.0001 for differences among sub-categories. aExcludes physicians who are sole owners of their practice, mostly solo practitioners. Source: Authors’ analysis.

640


associated (p = 0.03) with higher satisfaction in multivariate analysis (after adjustment for their very high administrative burden) but not in bivariate analysis. Not unexpectedly, employment by a medical school was associated with greater amounts of administrative work (presumably some of it teaching- or research-related). Sensitivity analyses excluding such medical school employees were virtually identical to the main analyses (data not shown). DISCUSSION U.S. physicians spend about 1.7 hours each day on non-patient-related administrative work—one-sixth of their total working hours. Administrative tasks may take more than time; they also seem to drain the satisfaction that doctors gain from their work. Many doctors have been fleeing private practice, in part to escape the administrative burden of running an increasingly complex small business. Yet our findings indicate that they face even greater administrative work as hospital employees or when practicing in a large group. Although proponents of electronic medical records have long promised a reduction in doctors’ paperwork (5), we found the reverse is true. Doctors with fully electronic medical records spent more time on administration than those using all paper records. Although documenting care often takes longer with the use of an electronic medical record, survey respondents were instructed that time on clinical documentation should be reported as “patient-related.” But perhaps physicians who entered data into electronic medical records perceived that much of this documentation was for billing purposes. If so, they may have designated some of this time as “administrative.” The only previous representative national survey that used definitions of patient-related and administrative work similar to ours was carried out in 1995 (6). That study estimated that administration and insurance-related matters accounted for 13.5 percent of physicians’ total work time. Our 2008 figure suggests a 23 percent increase in the administrative portion of physicians’ work day. This increase is somewhat smaller—10 percent—when expressed as hours per week, reflecting the shorter total work week (52.6 hours) reported in our survey as compared to the earlier one (58.3 hours). A few studies have examined the amount of time physicians spend on billing and insurance-related paperwork—a narrower definition of administrative work than we used. A 2000 California study estimated billing and insurance-related work consumed 4.9 percent of physician time (7). In a 2006 survey, physicians reported spending 3 hours per week interacting with private insurance plans, with primary care doctors and solo practitioners reporting slightly higher figures; 81.5 percent perceived that this work was increasing (8). A companion 2006 survey of office-based private practitioners in Ontario found they spent 2.2 hours per week interacting with insurers (vs. 3.4 hours in the United States when


/ 641

Medicare and Medicaid were included along with private insurers) (9). Differences in the time spent on these tasks by non-physician office staff were even larger; 20.6 hours of nurse time per physician in the United States versus 2.5 hours in Canada; 53.1 hours per week of clerical time in the United States versus 15.9 hours in Canada; and 3.1 hours per week of senior administrators’ time in the United States versus 0.5 hours in Canada. Some less reliable surveys provide interesting perspective. In an online survey with a low (2.3%) response rate, doctors reported spending 22.6 percent of their time on administrative work (10). As in our study, employed physicians reported spending more time on administration than practice owners. A pair of studies based on similar convenience samples found an increase in the proportion of doctors who reported spending 10 or more hours per week on paperwork and administration, from 33 percent in 2012 to 52 percent in 2013 (11). Several caveats apply to our findings. Although the data we analyzed are the most recent available to assess physicians’ administrative work in a nationally representative sample, they were collected in 2008 and may not fully reflect recent changes in the practice environment. In addition, data on total hours worked were imputed for 4.8 percent of the sample. The survey questionnaire asked physicians to divide their medically related work into only two categories: (1) direct patient care and (2) administrative tasks and professional activities. Hence, our measure may overstate administrative time since it includes some other non-direct patient care activities (e.g., attendance at Grand Rounds). On the other hand, the survey specifically instructed physicians to count patient documentation and patient-related office work as “direct patient care.” To the extent that documentation and office work are performed for reimbursement purposes, our measure understates doctors’ administrative time. Finally, some physicians whose main role is administrative see patients fewer than 20 hours per week and would be excluded from the survey. This might cause us to underestimate average administrative time. Taken as a whole, our analysis and previous studies suggest that U.S. doctors are devoting an increasing share of their work hours to administrative matters. If doctors’ administrative time is valued at the same rate as their clinical hours, our findings imply that $102 billion was spent on physician administration in 2014. Our findings also have implications regarding three major trends in health care delivery: the implementation of electronic medical records; a shift in practice settings from small practices to employment in large medical groups; and the increasing prevalence of risk-sharing arrangements. All three were associated with greater administrative burdens in our data—and may exacerbate physicians’ growing dissatisfaction. Much time and money is currently spent on medical billing and paperwork. A simpler system could realize substantial savings, freeing up more resources to expand and improve coverage.

642


Acknowledgments — This study used restricted-use data from the Health Tracking Physician Survey 2008 carried out by the Center for Studying Health System Change. Data were obtained through the University of Michigan’s InterUniversity Consortium for Political and Social Research (ICPSR File No. 27202). This study was performed without external funding. The authors have no financial conflicts of interest regarding this work. REFERENCES 1. U.S. Congress, Office of Technology Assessment. International Comparisons of Administrative Costs in Health Care. Publication # BP-H-135. U.S. Government Printing Office, Washington, DC, 1994. 2. Woolhandler, S., Campbell, T., and Himmelstein, D. U. Health care administration costs in the U.S. and Canada. N. Engl. J. Med. 349:768–775, 2003. 3. Ly, D. P., and Cutler, D. M. The (paper) work of medicine: Understanding international medical costs. J. Econ. Perspectives 25(2):3–25, 2011. 4. Boukas, E., Cassil, A., and O’Malley, A. S. A Snapshot of U.S. Physicians: Key Findings from the 2008 Health Tracking Physician Survey. Data Bulletin #35. Center for Studying Health System Change, September, 2009. http://www.hschange.com/ CONTENT/1078/1078.pdf (accessed February 11, 2014). 5. Anonymous. 1961 electronic medical records. Video. https://www.youtube.com/ watch?v=t-aiKlIc6uk (accessed February 13, 2014). 6. Remle, D. K., Gray, B. M., and Newhouse, J. P. Does managed care mean more hassle for physicians? Inquiry 37:304–316, 2000. 7. Kahn, J. G., et al. The cost of health insurance administration in California: Estimates for insurers, physicians, and hospitals. Health Aff. 24:1629–1639, 2005. 8. Casalino, L. P., et al. What does it cost physicians’ practices to interact with health insurance plans? Health Aff. 28:W533–W543, 2009. 9. Morra, D., et al. US physician practices versus Canadians: Spending nearly four times as much money interacting with payers. Health Aff. 30:1443–1450, 2011. 10. Merritt Hawkins. A survey of America’s physicians: Practice patterns and perspectives. The Physicians Foundation, 2012. http://www.physiciansfoundation.org/ uploads/default/Physicians_Foundation_2012_Biennial_Survey.pdf (accessed February 11, 2014). 11. Medscape. Physicians’ compensation report 2012 and 2013. http://www.medscape. com/features/slideshow/compensation/2013/public and http://www.medscape.com/ features/slideshow/compensation/2012/public (accessed February 11, 2014).

Direct reprint requests to: David U. Himmelstein, M.D. 255 West 90th Street New York, NY 10024 [email protected]