With increased economic pressures on hospitals, limitations on resident physician hours, and payment reductions for preventable harms, hospitals seek to increase productivity while improving the quality of patient care. Frequently, relative value units and patient encounters are used to track physician productivity and establish national benchmarks.1,2 However, productivity varies based on a range of characteristics that are not generally reported, limiting the accuracy of comparisons across institutions. Also, comprehensive process and outcome measures from different stakeholders' perspectives need to be established to align diverse health care interests, ensure widespread acceptability, and provide comprehensive goals.3 In the present study, we (1) identify an actionable measure of attending physician workload; (2) characterize factors accounting for differences in workload; and (3) identify a congruent set of measures that would be valued by disparate stakeholders.
We performed in-depth semistructured interviews with 8 hospitalist program directors in the Maryland/District of Columbia region exploring measures of workload, factors causing its variation, and potential safety and quality concerns. We then used a modified Delphi technique with small groups of hospitalists, nonphysician practitioners, house staff, and hospital administrators from private, academic, and community hospitals. Participants identified, critiqued, organized, and operationalized characteristics affecting attending workload. The authors also performed a stakeholder analysis and reviewed the nurse staffing and resident physician work hours literature to identify pertinent patient safety and quality outcomes for attending physicians.4,5
We found that workload was frequently tracked both as number of patient encounters and relative value units. Hospitalist directors reported difficulty in predetermining the relative value units of a particular service or shift, but the number of encounters could be tracked easily and controlled through systematic mechanisms.
Factors identified as affecting workload centered around physician, hospital, team, and patient characteristics (Figure). Physician characteristics included demographics, practice environment, work day activities, and compensation; hospital factors were primarily location and services related. Team characteristics included assistance, delegation of tasks, geographical localization of patients, and system controls for patient volume. Important characteristics of the patient group served included age, complexity of care, and access to health care.
Our literature review and stakeholder analysis revealed outcomes considered attributable to attending physicians: tests, radiographs, procedures and consultations ordered; critical value response time; medication errors; incident reports; morbidity; mortality; completeness of treatment discussions; patient satisfaction; and overall quality of care. Other suggested measures included procedure, test, admission or discharge delays; number of patients cross-covered; handoffs; transfers to higher levels of care; medication reconciliation; communication with the primary care provider; and readmission.
This is the first study, to our knowledge, to explore attending physician workload, develop a model for factors that may affect it, and identify generalizable process and outcome measures for attending physicians. Both relative value units and patient encounters are typically reported as hospitalist benchmarks,1 but participants believed that number of patient encounters could be more easily tracked and intervened on in real time. They also identified factors affecting workload as physician, hospital, team, and patient characteristics. Of these categories, team structure is likely the most modifiable. By using different team structures, such as with house staff or midlevel care providers, workload and efficiency may be improved.6
From our review of the literature and stakeholder analysis, we found that pertinent process and outcome measures centered around the hospitalization and transitions of care. Hospitalization outcomes focused on reducing unnecessary testing and consultation, increasing patient flow, addressing safety concerns, and providing high-quality patient-centered care. Transition of care measures focused on potential patient care delays and preventing clinical decompensation and readmission. Readmissions are currently being used as a benchmark; these other measures may also be used in the future.
Limitations of the study include potential lack of generalizability because our participants were from the Maryland/District of Columbia area. However, we had broad representation of the private, community, and academic settings, and many workload and quality concerns transcend geographic and service-level boundaries. Second, some of the categories of factors affecting workload may not be readily modifiable. However, we also present factors that can be intervened on, such as assistance (house staff and midlevel care providers)6 or geographic localization of patients.7 Understanding of both the modifiable and fixed factors is important for the overall assessment of how to influence workload.
This study has significant research implications. It recognizes that systems adapt differently to handle workload, and therefore it is important to understand the contextual factors within which patients are treated. This is essential to ensure accurate comparisons are being made among institutions. Second, the potentially modifiable factors we identified may be used to improve workload efficiency and should be further studied to assess their impact on both workload and outcomes. The Centers for Medicare & Medicaid Services Physician Quality Reporting System and value-based purchasing models make the study of attending physician outcomes increasingly important.8 Future research can use this workload model; identify the association of physician, hospital, team, and patient factors with outcomes; and determine targeted interventions to improve both the efficiency and quality of care.
Correspondence: Dr Michtalik, Division of General Internal Medicine, Hospitalist Program, 1830 E Monument St, Ste 8017, Baltimore, MD 21287 (hmichta1@jhmi.edu).
Published Online: April 22, 2013. doi:10.1001/jamainternmed.2013.405
Author Contributions: Drs Michtalik and Brotman had full access to all of the data in the study and take responsibility for the integrity of the data and accuracy of the data analysis. Study concept and design: Michtalik, Pronovost, Spetz, and Brotman. Acquisition of data: Michtalik. Analysis and interpretation of data: Michtalik, Pronovost, and Marsteller. Drafting of the manuscript: Michtalik and Pronovost. Critical revision of the manuscript for important intellectual content: Michtalik, Pronovost, Marsteller, Spetz, and Brotman. Statistical analysis: Michtalik. Obtained funding: Michtalik. Administrative, technical, and material support: Michtalik and Pronovost. Study supervision: Michtalik, Spetz, and Brotman.
Conflict of Interest Disclosures: None reported.
Funding/Support: This study was supported by grant T32 HP10025-17-00 from the National Institutes of Health (NIH) and NIH/Johns Hopkins Institute for Clinical and Translational Research KL2 Award 5KL2RR025006.
Role of the Sponsors: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript.
Additional Contributions: We thank the Hospitalist Directors for the Maryland/District of Columbia region for sharing their models of care and the Johns Hopkins Clinical Research Network Hospitalists and General Internal Medicine Research in Progress Physicians for their comments on the model and quality measures.
1.Landis WA, Ahlstrom TW, Blue SM,
et al. State of Hospital Medicine: 2011 Report Based on 2010 data: Medical Group Management Association and Society of Hospital Medicine. Philadelphia, PA: Glacier Publishing Services; 2011
2.Ward C, Reckewey K, Silbaugh B, Lewis T, Burnham S, Rogers T. Financial benchmarks for hospitalist programs.
Physician Exec. 2002;28(6):43-4712448143
PubMedGoogle Scholar 3.Jha AK, Li Z, Orav EJ, Epstein AM. Care in US hospitals—the Hospital Quality Alliance program.
N Engl J Med. 2005;353(3):265-27416034012
PubMedGoogle ScholarCrossref 4.Fletcher KE, Davis SQ, Underwood W, Mangrulkar RS, McMahon LF Jr, Saint S. Systematic review: effects of resident work hours on patient safety.
Ann Intern Med. 2004;141(11):851-85715583227
PubMedGoogle ScholarCrossref 5.Kane RL, Shamliyan T, Mueller C, Duval S, Wilt TJ. Nurse staffing and quality of patient care.
Evid Rep Technol Assess (Full Rep). 2007;(151):1-11517764206
PubMedGoogle Scholar 6.Singh S, Fletcher KE, Schapira MM,
et al. A comparison of outcomes of general medical inpatient care provided by a hospitalist-physician assistant model vs a traditional resident-based model.
J Hosp Med. 2011;6(3):122-13021387547
PubMedGoogle ScholarCrossref 7.Singh S, Tarima S, Rana V,
et al. Impact of localizing general medical teams to a single nursing unit.
J Hosp Med. 2012;7(7):551-55622791661
PubMedGoogle ScholarCrossref 8.Centers for Medicare & Medicaid Services, Department of Health and Human Services. Medicare program; Medicare shared savings program: accountable care organizations: final rule.
Fed Regist. 2011;76(212):67802-6799022046633
PubMedGoogle Scholar