Recent research has shown increasing levels of physician burnout.1 Studies show that decreased physician satisfaction is costly, disruptive, leads to turnover, and negatively affects patient access.2–4 In contrast, satisfied doctors have happier staff and lower rates of medical errors.5 Patients of these physicians are also more satisfied6,7 and more likely to adhere to care recommendations.8
Previous studies of career satisfaction among academic faculty have cited institutional leadership factors and relationships with colleagues as key drivers of satisfaction.9,10 However, practice environment and administrative workload have also been identified as key factors. Physicians continue to report frustration with time spent on administrative tasks and paperwork,11 which has ranged from 16% and 22% of their work hours in national surveys.12,13 The proportion of time spent on administrative duties is associated with lower levels of career satisfaction.13 Electronic health records (EHRs), mandatory documentation requirements, and incentive programs including Meaningful Use are frequently cited14 as contributing to this trend, but other requirements, such as maintenance of certification, have also recently come under criticism.15
Given the current level of escalating complaints and the link between administrative workload and dissatisfaction, we designed this study to identify physician characteristics that predict levels of administrative burden, including gender, age, specialty, and years of experience. We also assessed the correlation between administrative burden, dissatisfaction, and symptoms of burnout. Finally, to identify strategies to relieve administrative burden, we assessed the relative value and burden of specific administrative tasks, and whether there were opportunities for further delegation and distribution of them.
Data for this study are from the 2014 Massachusetts General Physicians Organization (MGPO) Physician Survey, which is conducted every two years and was administered in the spring of 2014. The biennial survey targets the clinically active physician members of the MGPO, which is the multispecialty academic faculty practice of physicians who work at Massachusetts General Hospital, in Boston, Massachusetts, as part of its Quality Incentive Program, through which physicians qualify to earn a financial incentive for completion of the survey.16 The survey authors (S.K.R., S.R.L.), who are staff of the MGPO, overseen by the medical director, modify the content of the survey each time it is administered to reflect current hospital priorities. Previous surveys included questions assessing physicians’ satisfaction with their careers, compensation, hospital leadership, and aspects of clinical practice.
We designed the 2014 survey to incorporate assessment of administrative workload, physician satisfaction, burnout,17 and engagement, which were all new content areas not addressed by previous MGPO surveys. We included questions from the 2012 survey on physician satisfaction, compensation, and administrative support. We compared responses to questions that were included in both the 2012 and 2014 surveys.18 We used literature review and expert interviews to develop questions. We tested the instrument on 10 subjects using cognitive interviewing. Through this process, we modified 17 questions and excluded 13. The Partners human research committee approved the survey.
As a proxy for “percent effort” spent on administrative duties, we asked physicians, “In an average week, what percentage of the total hours you work is spent on administrative tasks related to patient care?” Physicians reported their age, gender, specialty, and years since training.
To assess the impact of administrative workload on physicians, we used a five-point Likert scale (strongly agree, agree, neither agree nor disagree, disagree, strongly disagree) to assess to what degree physicians agreed that administrative duties affect their ability to focus during clinical encounters and deliver high-quality care, as well as whether they had enough administrative support. We measured levels of physician burnout using the Maslach Burnout Inventory–General Survey, a 16-item validated instrument widely used to assess physician satisfaction.17 The Maslach Burnout Inventory groups items into three scales: exhaustion, cynicism, and professional efficacy. Per the Maslach Burnout Inventory manual, respondents with scores of greater than or equal to 3.2 on the exhaustion subscale, greater than or equal to 2.6 on the cynicism subscale, or less than or equal to 3.8 on the professional efficacy subscale were defined as having high levels of burnout in that particular scale.
We used responses on a five-point Likert scale (very satisfied, satisfied, neutral, dissatisfied, very dissatisfied) to the question “How satisfied are you with your career as a physician?” to assess overall career satisfaction. Although there are many reasons why academic physicians reduce their clinical commitment, we used responses to the question “In the next one to three years, do you plan to continue as you are, cut back on hours, cut back on patients seen, or close your practice to new patients?” to determine whether physicians who spent less time on administrative tasks were more likely to continue practicing as they had been.
To understand the impact of specific administrative tasks, the leadership at the MGPO, comprising the CEO, COO, medical director, and other senior leadership, identified 15 administrative activities that were thought to be contributing to administrative workload (Figures 1 and 2). Physicians were first asked, “How valuable is this task?” and answered from four choices: very valuable, valuable, somewhat valuable, or not valuable. We then asked, for the same 15 tasks, “How burdensome is this task?” Respondents chose between four answer choices: very burdensome, burdensome, somewhat burdensome, and not burdensome. For each of the tasks, we calculated an average response across all respondents for both value and burden. Finally, for the same set of tasks, we asked respondents to indicate what percentage of the task could be delegated. To assess potential interventions, we listed nine potential interventions to reduce administrative workload and asked physicians how helpful each would be. To assess physicians’ attitudes toward potential financial trade-offs of relieving administrative burden, we asked, “What percentage of your income would you be willing to forgo in return for not having to fulfill EHR-related documentation requirements?”
We used descriptive summary statistics for levels of administrative workload and chi-square analyses to assess differences between demographic groups and response categories. Univariate and multivariate analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, North Carolina) to identify predictors of time spent on administrative duties and perceptions that administrative duties affect the quality of care provided. The Partners human research committee approved this study.
We hypothesized that the data collected would indicate that there is variation across demographic groups in time spent on administrative duties; that the amount of time spent on administrative duties impacts physicians’ career satisfaction and burnout; that administrative burden is the result of many different tasks which vary in value and burden; and that physicians would have interest in certain solutions being developed.
Of the 1,850 clinically active physicians at the MGPO, 1,774 (96%) responded to the survey (Table 1). By gender, 747 (42%) were female, 985 (56%) practiced a medical specialty, and 768 (43%) had 10 or fewer years of experience since training (Table 1).
Amount of time spent on administrative duties
Responding physicians reported that on average, 24% of their time was spent on administrative duties related to patient care. Primary care physicians (33% of time) and women (27%) reported a higher percentage of time spent on administrative duties, whereas emergency medicine, radiology, anesthesia, and pathology (ERAP) specialties reported the lowest (Table 1). Multivariate regression analysis controlling for demographic and specialty factors confirmed that these differences are statistically significant (Table 2). We found that female physicians spent 3.3% more time on administrative duties than male physicians (P < .01). Relative to medical specialties, primary care physicians spent 8.5% more time (P < .01) while ERAP specialties spent 6.9% less time (P < .01) on administrative duties. Surgical specialties spent less time on administrative duties than medical specialties, but the difference was not statistically significant (P = .10). We also found that number of years since training was predictive of time spent on administrative duties. Compared with those with more than 30 years of experience, those with 21 to 30 years spent 3.2% less time (P = .04), those with 11 to 20 years spent 2.4% less time (P = .09), and those with 10 years or less spent 3.2% less time (P = .02) on administrative duties.
Impact of time spent on administrative duties
The majority of physicians agreed or strongly agreed that administrative duties affected their overall ability to deliver high-quality care (1,153; 65%) and their ability to focus during patient encounters (1,064; 60%). Five hundred thirty-two (30%) physicians reported that they had enough administrative support, a decrease from 50% since 2012 (1,649 respondents; response rate = 92%). ERAP physicians were less likely to report that administrative duties impacted their ability to deliver high-quality care (b = −1.34, P < .01), while primary care physicians were more likely to encounter this problem (b = 0.39, P < .01) (Table 2). Respondents who planned to close their practice to new patients (71; 4.1%) or cut back on patients seen (213; 12.0%) spent more time on administrative duties (24.4% and 30.4% of time, respectively) (P < .01) (Table 1). After adjusting for gender, race, specialty, and years of experience, a higher percentage of time spent on administrative duties was predictive of reduced career satisfaction (OR = 0.99, P < .01) and burnout as defined by the Maslach Burnout Inventory17 (OR = 1.01, P < .01).
Causes of administrative burden
Prior authorizations, ambulatory clinical documentation, and medication reconciliation were three categories of tasks most often cited as creating the greatest administrative burden. Prior authorizations, academic promotion, and mandatory training were rated as the least valuable, while care team communication, inpatient clinical documentation, and ambulatory clinical documentation were seen as the most valuable (Figure 1). Over half of the respondents felt that all of the work involved in prior authorizations could be delegated, and over two-thirds of physicians felt that at least 50% of the work related to medication reconciliation could be managed by others (Figure 2).
Physicians reported that they were generally willing to try any of the potential solutions offered in the survey; they most strongly preferred centralized preauthorization services, a single electronic credentialing repository, automated documentation technology, scribes, and more clinical support. However, only 454 (25%) physicians surveyed indicated that they would be willing to forgo part of their incomes in return for relief from EHR documentation, and 767 (43%) would not be willing to give up any income.
Our findings confirm that the amount of time physicians spend on administrative work has a serious effect on physician well-being and may be contributing to high rates of physician burnout.1 Some of the administrative activities could result in better availability of information and clinical care, both of which our physicians clearly identified. In aggregate, the accumulation of these tasks produced a substantial negative impact on our physicians. Our findings are particularly relevant for academic physicians who typically balance clinical responsibilities with multiple additional roles. Leaders of academic medical centers should be aware that administrative tasks related to clinical practice may crowd out other aspects of the academic mission, such as teaching or research.
Our study identified specific populations at more or less risk of experiencing burden from administrative tasks. Our primary care faculty, like at many academic medical centers, have been intensely engaged in system-based improvement initiatives including Patient-Centered Medical Home transformation and often face additional documentation requirements related to quality measurement, which may explain their higher levels of administrative burden. Possible explanations for the fact that ERAP physicians experienced less burden are the shift-based structure that these fields have, the nature of those particular specialties, and specific characteristics of physicians that go into those specialties. Our findings do not provide a clear answer. These specialties have many different features tying them together including absence of longitudinal responsibility for patients, relatively low Meaningful Use requirements, and, at our institution, the use of standardized documentation technology. Future surveys should compare shift- and non-shift-based clinicians within specialties, such as obstetrics–gynecology. Future studies could look into individual characteristics that are associated with reduced percentages of time spent on administrative duties, such as technologic proficiency, to determine whether physician features underlie this specialty-specific pattern. The increased burden for female physicians we identified warrants further exploration, especially as it may relate to part-time work, since it was not explained by specialty alone. Finally, although some have posited that younger physicians may adapt more readily to the electronic environment, our data suggest that the differences are not large across the generations, except for those clinicians with more than 30 years of experience.
Ambulatory clinical documentation was the only task identified as being both valuable and burdensome, suggesting it as a priority area to be explored further. Because all of our respondents use an EHR, we were not able to determine the effect of EHR adoption on physicians. It is notable that ERAP specialties, which are generally less affected by ambulatory clinical documentation requirements, and which at our institution have adopted automated documentation technology, had fewer physicians report that their ability to deliver high-quality care and focus during patient encounters was impacted by administrative tasks. Many of today’s documentation requirements originated with Medicare billing requirements and are outdated in the context of the EHR, which enables automated mechanisms for physicians to attest to the complexity of their services. Furthermore, in value-based payment models, certain aspects of clinical documentation are irrelevant, such as the number of minutes spent with a patient. Other clinical data, such as smoking status, which inform quality measurement and risk adjustment, will become more important. It will be critical for Medicare and other payers to revise documentation requirements to reflect these changes and to encourage systems to take advantage of automated technology to fulfill them.
Physicians in our survey described several additional areas, such as maintenance of certification and continuing medical education, as at least “somewhat burdensome,” confirming that a gradual accumulation of requirements in different areas is contributing to the rising sentiment of complaint. Interestingly, our physicians identified the area of “mandatory training” as burdensome, despite efforts at our institution to limit required training programs to 45 minutes of online curriculum, covering a range of topics including infection control, HIPAA compliance, and disability awareness. This suggests either that respondents considered other training requirements, such as specialty-specific requirements, when responding, or that in the context of many different administrative requirements, even a task requiring less than one hour annually is perceived as burdensome. If the latter is true, that is particularly concerning for academic physicians, as grants administration, human resources, expense reports, conflict-of-interest reporting, and trainee evaluations move to electronic systems reliant on physician entry. Although many of these tasks may seem minor, the aggregate can be overwhelming.
We suggest there are three broad strategies for reducing administrative burden for physicians. The first is to reduce the number of requirements. The organizations that develop requirements for physicians must work together to reduce the overall load by eliminating low-value mandates and aligning high-value requirements through unified activities. The second is to try to manage the flow of requirements better. For example, we have organized quality improvement projects that can count towards American Board of Medical Specialties maintenance of certification credits, continuing education credits, and fulfillment of ongoing professional practice evaluation requirements, which has consolidated multiple requirements for physicians.
Third, our data show there is clearly an opportunity for more delegation of tasks. Although many institutions, including ours, have incorporated nonphysician providers into clinical practice, our study does not enable us to determine how these team members impact physician burden. Ironically, some of the laws, regulations, and directives from the Centers for Medicare and Medicaid Services and other regulators like the Joint Commission concerning scope of practice limit the activities that other members of the medical staff can do to help physicians. For example, medical assistants cannot enter lab orders without certification that was until recently expensive and difficult to obtain in our state.
Where requirements cannot be eliminated, consolidated, or delegated, human assistance technology such as voice recognition and scribes should be developed to automate administrative tasks and off-load low-value activities from physicians. At our institution, we are piloting the use of scribes in high-volume ambulatory specialties. Although many studies have demonstrated improved satisfaction, efficiency, and revenue gain for physicians using scribes, most are small with methodological limitations.19 Further study is warranted, since for many cost-constrained clinical practices, the cost of scribes would need to be covered by increased clinical volume, which may exacerbate administrative burden.
Limitations to this study include generalizability since we conducted our survey in a single academic medical center, in a single city, with an EHR. Although we only included physicians with some level of clinical activity, many of our physicians are involved in research and other administrative roles. Without controlling for time spent on these activities and other features of academic work life, we are limited in our ability to draw actionable conclusions from the relationship we observed between time spent on patient-related administrative duties, burnout, and reported likelihood to reduce clinical practice. However, we were able to get a robust response across a large multispecialty group which increases the accuracy of the data.
The biggest challenge to solving these problems may be cost. The cost of a physician doing data entry following a clinical session is not currently accounted for in most of our systems, and finding a substitution for this labor will be expensive. Physicians, not surprisingly, report that they are not enthusiastic about shouldering these costs themselves, but some have found the tasks onerous enough that they would be willing to forgo income to reduce the burden. This finding may signal a tipping point: We are seeing an increasing number of physicians choosing to take a penalty rather than complete and submit data for incentive programs such as Meaningful Use and the Physician Quality Reporting System. Our findings suggest that for a larger portion of physicians, however, it will be important that salary levels be maintained while solutions for relieving burden are implemented. With that in mind, solutions will have to generate increased efficiency, and even productivity, for organizations to address this in a cost-neutral way, in the short term. Ideally, organizations will see that there will be indirect and long-term benefits of these interventions, such as faculty retention.
Future investigations need to examine the different aspects of clinical practice, documentation, and technology adoption that may have driven the variation we identified across groups, especially primary care, which is uniquely affected.
In conclusion, our survey of clinically active academic physicians shows that physicians spend a substantial amount of time on administrative duties, which they reported was impacting their ability to focus on patients, their overall career satisfaction, and levels of burnout. Survey responses suggest that further study of interventions targeted at multiple aspects of academic clinical practice will be necessary to address this complex issue.
Acknowledgments: The authors would like to acknowledge Kaitlyn Moran, BS, Massachusetts General Hospital, who contributed to the writing, editing, and review of this manuscript. They would also like to acknowledge Daniel Balliro, Massachusetts General Hospital, who contributed to the data analysis. Both of these individuals were compensated by Massachusetts General Hospital.
1. Shanafelt TD, Hasan O, Dyrbye LN, et al. Changes in physician burnout and the general US working population between 2011 and 2014. Mayo Clin Proc. 2015;90:16001613.
2. Pathman DE, Konrad TR, Williams ES, Scheckler WE, Linzer M, Douglas J; Career Satisfaction Study Group. Physician job satisfaction, dissatisfaction, and turnover. J Fam Pract. 2002;51:593.
3. Buchbinder SB, Wilson M, Melick CF, Powe NR. Estimates of costs of primary care physician turnover. Am J Manag Care. 1999;5:14311438.
4. Landon BE, Reschovsky JD, Pham HH, Blumenthal D. Leaving medicine: The consequences of physician dissatisfaction. Med Care. 2006;44:234242.
5. Williams ES, Manwell LB, Konrad TR, Linzer M. The relationship of organizational culture, stress, satisfaction, and burnout with physician-reported error and suboptimal patient care: Results from the MEMO study. Health Care Manage Rev. 2007;32:203212.
6. DeVoe J, Fryer GE Jr, Straub A, McCann J, Fairbrother G. Congruent satisfaction: Is there geographic correlation between patient and physician satisfaction? Med Care. 2007;45:8894.
7. Haas JS, Cook EF, Puopolo AL, Burstin HR, Cleary PD, Brennan TA. Is the professional satisfaction of general internists associated with patient satisfaction? J Gen Intern Med. 2000;15:122128.
8. DiMatteo MR, Sherbourne CD, Hays RD, et al. Physicians’ characteristics influence patients’ adherence to medical treatment: Results from the Medical Outcomes Study. Health Psychol. 1993;12:93102.
9. Bunton SA, Corrice AM, Pollart SM, et al. Predictors of workplace satisfaction for U.S. medical school faculty in an era of change and challenge. Acad Med. 2012;87:574581.
10. Bergus GR, Randall CS, Winniford MD, Mueller CW, Johnson SR. Job satisfaction and workplace characteristics of primary and specialty care physicians at a bimodal medical school. Acad Med. 2001;76:11481152.
11. Jauhar S. Why doctors are sick of their profession. Wall St J. August 29, 2014. http://online.wsj.com/articles/the-u-s-s-ailing-medical-system-a-doctors-perspective-1409325361
. Accessed August 9, 2016.
12. Physician Foundation. A survey of America’s physicians: Practice patterns and perspectives. http://www.physiciansfoundation.org/uploads/default/Physicians_Foundation_2012_Biennial_Survey.pdf
. Published 2012. Accessed August 16, 2016.
13. Woolhandler S, Himmelstein DU. Administrative work consumes one-sixth of U.S. physicians’ working hours and lowers their career satisfaction. Int J Health Serv. 2014;44:635642.
14. Friedberg MW, Chen PG, Van Busum KR, et al. Factors affecting physician professional satisfaction and their implications for patient care, health systems, and health policy. 2013.Santa Monica, CA: RAND Corporation.
15. Teirstein PS. Boarded to death—Why maintenance of certification is bad for doctors and patients. N Engl J Med. 2015;372:106108.
16. Meyer GS, Torchiana DF, Colton D, et al. Henriksen K, Battles JB, Keyes MA, et al. The use of modest incentives to boost adoption of safety practices and systems. In: Advances in Patient Safety: New Directions and Alternative Approaches (Volume 3: Performance and Tools). 2008. Rockville, MD: Agency for Healthcare Research and Quality; http://www.ncbi.nlm.nih.gov/books/NBK43652/
. Accessed August 9, 2016.
17. Maslach C, Jackson SE, Leiter MP. Maslach Burnout Inventory Manual. 1996.3rd ed. Palo Alto, CA: Consulting Psychologists Press.
18. Rao SK, Lehrhoff S, Litvak AJ, Kimball AB, Torchiana DF, Ferris TG. Osbourne R. Predictors of satisfaction with career and compensation among physicians in a multispecialty academic practice. In: Job Satisfaction: Determinants, Workplace Implications. 2015.New York, NY: Nova Science Publishers.
19. Shultz CG, Holmstrom HL. The use of medical scribes in health care settings: A systematic review and future directions. J Am Board Fam Med. 2015;28:371381.