The rotating hospitalist: a solution for an academic internal medicine practice.

Background: Concerns have been raised about the role that hospitalists may have in changing the educational process for medical students and residents, especially with regard to the primary care specialties.

Methods: We implemented rotating hospital and office duties within our five-physician group.

Results: Resident and student satisfaction increased, and additional faculty members could be added without expanding office space. Financial benefits then also ensued.

Conclusion: Rotating hospital with office duties among a small group of internists has resulted in most of the advantages with few of the drawbacks related to the hospitalist approach.

**********

Key Points

* Rotating physicians to function as hospitalist improved quality of life and resident training compared with the traditional practice approach.

* A physician was able to be added without requiring additional examination room space.

* The practice benefitted financially, and individual physicians were minimally impacted.

* Call and after-hours responsibilities needed to be fine-tuned to reasonably distribute the workload.

Hospitalists have recently become an increasingly popular means for providing inpatient care. (1,2) Advantages have included more efficient delivery of care as well as better time management for both the primary care physician and the hospitalist. Hospitalists have become integrated into academic programs more recently, (3,4) although concerns have been raised about the impact of these programs on medical education. (5-7) The impact has typically been a positive one on student and resident education and satisfaction, although comparative studies are lacking. (3,4,8)

Description of the Traditional System

Before August 2000, a traditional team was used for the hospital practice. This consisted of a senior resident, two junior residents, and medical students at various stages of training. There were four attendings, two associated with each team. The teams would perform work rounds with the attendings in the morning, but times would vary because of outpatient practice responsibilities. The attendings would then return to their office where they would see patients, teach ambulatory medicine, prepare lectures, and perform other duties as needed. Admissions came from two sources. The outpatient practice provided a moderate stream of admissions, and "unattached" admissions from the emergency department supplemented this and assured an adequate patient population for the services. These unattached patients are patients who do not have a local primary care physician with admitting privileges to our hospital. Unattached admissions were assigned to the attendings and their respective teams on a rotating basis.

Problems Encountered with the Traditional System

The hospital duties and office duties did not coexist well. When performing work rounds, attending rounds, or otherwise supervising the residents in the hospital, office messages would pile up and unexpected events would occur that would delay or disrupt the office schedule. When in the office, the attending was less available to the resident staff for informal discussion, guidance in decision making, and coverage of admissions. A common feeling was that wherever the attending was, he or she needed to be in the other place. This resulted in increased stress and decreased job satisfaction in the attending staff as well as missed teaching opportunities for the residents and students.

Methods

Description of the New System

Beginning August 1, 2000, we instituted a rotating hospitalist system. In essence, one attending is responsible for virtually all hospital care for both teams for a 2-week span. This frees the remaining attendings to concentrate on the ambulatory practice as well as lecture preparation, research, and other activities.

In an attempt to level the workload, the hospitalist does not take call. Admissions that come in after hours are assessed by the resident staff and presented to the attending on call. That attending may choose to see the patient, or if the problem is straightforward may defer the admission workup to the hospitalist the next day. On weekends, the on-call physician is responsible for admission workups on patients admitted until the evening. The hospitalist works up patients admitted overnight. These modifications reduced the workload of the hospitalist on evenings and weekends, reducing the stress and producing a more manageable workload for the 2-week tour of duty.

In both systems, the emergency department physicians directly contact the attending for new admissions. This enables the attending to determine appropriateness of admission to our service. The resident staff then evaluates the patient. The attending, then, receives at least two separate evaluations of the patient, one from the emergency department physician and the other from the resident.

Evolution of the New System

Initially, the on-call physician was also responsible for admissions between 4:30 PM and 9:00 PM. This resulted in common trips to the hospital that were thought not to be necessary for the management of stable admissions. Currently, the on-call physician will visit a hospital patient if the diagnosis is in doubt or the patient is critically ill.

Results

Resident and student satisfaction has increased. Having an attending in the hospital has increased their comfort level with the care of the patients. They believe the ability to run ideas past an attending is beneficial, and are quite pleased with the informal teaching opportunities that have arisen. This correlates well with the hospitalist-as-educator experiences elsewhere. (5)

Attending satisfaction has increased. No longer tom between office and hospital, the attending staff has been able to concentrate on the task at hand. This has reduced stress and improved job satisfaction, and we believe the improved attitudes have increased our ability to provide a positive role model for residents and students considering internal medicine careers.

There have also been financial benefits. We were able to add a fifth attending to the group without needing to add examination rooms. Because the number of physicians actually in the clinic at any time was not increased, we needed to add only an office. This then resulted in an increase in the overall number of outpatient encounters through the office. Although encounters by the original four physicians decreased from 8,584 to 8,147 in the first year of the program (a decrease of 437 encounters, or 5%), we think this is likely because of the addition of a new physician to the group (emergency visits would be seen by the new physician rather than squeezed into the established physician's schedule). Overall, including the new physician, the number of outpatient encounters increased from 8,584 to 9,518, an increase of 10.9%. It is expected that this will continue to rise, as the new physician further builds his practice.

Discussion

Concerns were raised that this change could adversely affect the practice in a number of ways. There was concern that the doctor-patient relationship would be adversely affected as the patient's primary care provider would likely not be the one following them in the hospital, and that patients would object to seeing the covering attendings when the patient's primary care physician was on hospitalist rotation. This was countered with the fact that coverage occurs when an attending is on vacation or otherwise unavailable, and that as an academic practice the patients have become accustomed to seeing other providers as well as the primary attending. Another potential pitfall is that patients would need to wait too long for test results when their attending went on hospitalist duty. This problem was countered by distributing test results to the office-based attending physicians who are then responsible for reporting results to the patients and ordering appropriate follow-up. (9) The chart with the test result is sent for review to the primary attending after being addressed by the covering attending to improve continuity. Also, the primary attending may be contacted if needed to establish the approach to a particular problem.

Continuity of care has required additional work. Communication is imperative between the hospitalist, residents caring for the patient in the hospital, and the primary physician. Aside from informal discussions of patients, a copy of the discharge note is faxed from the hospital floor to the office at the time of discharge. This discharge note includes a brief summary of care as well as the medication list. Residents are encouraged to dictate discharge summaries on the day of discharge, reducing the time it takes for this summary to reach the primary physician. Still, direct telephone communication between the resident and the primary physician is occasionally necessary and may interrupt the activities of one or the other.

The first 1 to 2 weeks on return from hospitalist duty tend to be quite busy in the office, dealing with a backlog of patient visits and paperwork. The first day back is typically scheduled lightly in an attempt to allow the physician to clear the backlog of paperwork and review tests and events involving outpatients.

There was also the concern about loss of hospital skills. There are five attendings currently, and each attending is hospitalist for 2 weeks followed by 8 weeks in the office. The amount of time out of the hospital does not appear to be enough to reduce those skills, and hospital skills are also reinforced when on call on weekends, when the on-call attending is responsible for an average of two hospital admission workups per day.

In all, this has been a positive experience. Attendings, resident staff, and medical students have all expressed their enthusiasm for this system. Patients have for the most part accepted the change, especially when the purposes are explained.

References

(1.) Redelmeier DA Canadian perspective on the American hospitalist movement. Arch Int Med 1999; 159:1665-1668.

(2.) Echikson AB. The hospitalist movement. Arch Intern Med 2000;160:552 (letter).

(3.) Shea JA, Wasfi YS, Kovath KJ, et al. The presence of hospitalists in medical education. Acad Med 2000;75(10 Suppl):S34-S36.

(4.) Brown MD, Halpert A, McKean S, et al. Assessing the value of hospitalists to academic health centers: Brigham and Women's Hospital and Harvard Medical School. Am J Med 1999;106:134-137.

(5.) Haftel HM, Bozynski ME. Changing teaching for changing times: The effect of a hospitalist program on the education of students. Acad Med 2000;75:521.

(6.) Goldman L. The impact of hospitalists on medical education and the academic health system. Ann Int Med 1999;130:364-367.

(7.) Whitcomb WF, Nelson JR. The role of hospitalists in medical education. Am J Med 1999;107:305-309.

(8.) Hauer K, Wachter R. Implications of the hospitalist model for medical students education. Acad Med 2001;76:324-330.

(9.) Ridgeway NA, Ginn DR, Harvill LM, et al. An efficient technique for communicating reports of laboratory and radiographic studies to patients in a primary care practice. Am J Med 2000;108:575-577.

From the Department of Internal Medicine, James H. Quillen College of Medicine, East Tennessee State University, Johnson City, TN.

Reprint requests to Jeffrey A. Summers. MD, ETSU Physicians and Associates, 146 W. Park Drive, Kingsport, TN 37660. Email: [e-mail]

Accepted July 25, 2002.