How to Integrate Team, Night Float, and Apprenticeship Models into a Surgery Residency Program

How to Combine Teaching Models to Optimize Resident Education and Comply with Work Hour Mandates

Richard H. Bell, Jr., MD, FACS,
Loyal and Edith Davis, Professor and Chair and
Residency Program Director,
Department of Surgery,
Feinberg School of Medicine,
Northwestern University,
Chicago, IL

Step 1. Development of the rotation models

In September 2002, in response to the then-new ACGME work hour mandates for surgery residents, we convened a think tank in Chicago. We wished to develop some practical models for resident education that would allow programs to comply with work hour restrictions while at the same time preserving and hopefully enhancing the educational experience of residents. The conference described four models of training, which were detailed in an earlier article for this section of the American College of Surgeons Web site (November 11, 2003) and in a manuscript published in the journal Surgery in January 2003.¹

In brief, the four models described were: first, the traditional team model, with night call spread out to accommodate the 80-hour work limit. This model was called the stretch model. Secondly, we discussed a night float model, which had already been adapted by some programs around the country by that time. In the night float system, all residents who work during the day leave in the evening and are replaced by a team that works from evening until the following morning, in other words a shift system. The third model described was called the apprenticeship model, in which a resident is assigned to one or two faculty mentors for an extended period of time and shadows the attending(s) in the office and operating room. In the pure apprentice model, the resident is responsible only for the call of his/her mentor's patients and takes no in-house call. Thus, the resident's experience closely mimics the lifestyle of the attending physician to which he or she is assigned. Finally, we envisioned a model called the competency model, in which the resident is not assigned to a specific attending or team of attendings, but rather participates in those cases where the resident has a need for further experience. The resident is then responsible for following the patients in whose operation he/she has participated. Again, in the pure form of this model, the resident does not take night call.

Regardless of the model employed, each rotation was assigned an Education Coordinator (EC). Each EC drafted a rotation summary using a provided template that outlines contact infor- mation, learning objectives (written by ACGME competency category), required and elective conferences, reference materials, types of operations performed by the attending(s), and the formative and summative evaluation plan. Those using 360-degree evaluation systems include a copy of the form that the patients and/or nurses will be completing.

In trying to change our residency program from a traditional program in which residents often worked over 100 hours per week to a program compliant with ACGME standards, we assembled a working team of faculty and residents who met once a week for about nine months. A few times during the year, we scheduled extended half-day meetings. Ultimately, we decided that the most effective way of reaching our goal of reducing working hours, while preserving/enhancing the learning environment, was to combine the models we had discussed at the 2002 think tank.

Step 2. Adopting the night float rotation

We realized that trying to use the stretch version of traditional hospital-based resident teams would result in requiring that residents cross-cover a very significant number of patients at night to whom the resident had no prior exposure. Under the old system, our residents were already frustrated about the number of services which had to be cross-covered at night and we did not wish to exacerbate the problem. So we decided early on that we needed to institute a night float system.

We decided to ask the night float team to cover from 6 pm to 7 am and the day teams from 6 am to 7 pm so that there would be an hour available from 6 am to 7 am and from 6 pm to 7 pm for the transfer of care. We decided to ask the night float team to work 13 hours, six nights per week, or a total of 78 hours per week. We decided to have the night team work Sunday to Friday. This avoided the problem in traditional systems of having residents miss an operating day because of taking call the night before. We also decided that the night float team would scrub on any operations commenced between 6 pm and 7 am as well as help finish some cases that were still going from the daytime as of 7 pm.

Thus, with the creation of the night float team, we started with the structure below:

Figure 1. Day and night rotations. There is an hour of overlap from 6 am to 7 am and from 6 pm to 7 pm.

Once we had agreed to implement a night float system, the stretch model was abandoned, but we retained some traditional multi-resident teams who worked during the day from Monday to Friday. We simply called these teams.

Step 3. Weekend Coverage

We were then faced with the problem of filling in the 36 hours of weekend coverage that was not provided by the night float team. To accomplish this, we took all of the residents working on the daytime Monday to Friday services and divided them into three groups. These three groups then were assigned to weekend coverage as follows:

	WEEK 1	WEEK 2	WEEK 3
SATURDAY 6 AM – SUNDAY 7 AM	Group 1	Group 2	Group 3
SUNDAY 6 AM – SUNDAY 7 PM	Group 3	Group 1	Group 2
OFF	Group 2	Group 3	Group 1

With the assumption that the Monday to Friday work week was 65 hours (13 hours x 5 days), the result of this scheme was that resident work hours for the non-night team residents averaged 78 hours over a 3 week period, as shown below:

	Monday-Friday Hours	Weekend Hours	Total Hours
Week 1	65	25	90
Week 2	65	13	78
Week 3	65	0	65
AVG OVER 3 WEEKS			78 hours

Step 4. Maximizing the educational potential of the Monday to Friday experience

With a system in place that created an environment in which residents working during the day Monday to Friday would be well rested and hopefully well prepared for learning experiences, we wanted to create rotations that would maximize their educational opportunities. When we looked at our residency as it had been, we felt that there was not enough operative experience early in the residency. We also felt that residents were too tied down to the inpatient service to be able to see their patients in the outpatient environment before and after surgery. Even though we required our residents to attend two half days of clinic per week, the fact that they operated with multiple attendings made it unlikely that they would see their operative patients in the ambulatory setting pre- or postoperatively. Finally, we felt that the large inpatient services with multiple attendings, multiple residents, and many patients diluted the teacher-student relationship between any given attending and resident.

For all of the above reasons, we created several rotations at all levels of the residency that placed one resident with one or at the most two attendings. In these apprenticeship experiences, the resident was expected to attend all or virtually all office hours, scrub on every case done by the attending, and make inpatient rounds with the attending. Thus the rotation mimicked the life of an attending surgeon.

Naturally, not all attending surgeons were assigned residents. Those attendings who were no longer assigned residents were chosen for non-teaching roles because the size or nature of their practice was not optimal for resident education or because they had not received strong teaching evaluations. Conversely, attendings who were strong teachers and had practices that provided important case material for residents were chosen as mentors for the apprenticeship rotations. For the first year residents, two months of general surgery apprenticeships were created with surgeons who had busy practices with many intern-appropriate cases such as hernia repair. For second year residents, we created two apprentice rotations—one in general surgery and one in breast surgery, each for two months. In the third year, a two-month apprentice rotation was offered in endocrine surgery. In the fourth year, apprentice rotations were created in colorectal and thoracic surgery. In the chief resident year, an apprenticeship rotation was created in surgical oncology, focusing on an attending with a high volume of high complexity cases. On the upper level apprenticeship rotations, no interns were assigned, so nurse practitioners and/or physician assistants were recruited to assist with daily inpatient care.

We preserved some of our traditional team rotations in which a group of residents work with a group of attendings. Part of the reason for preserving the team rotations was the residents' desire to continue to have opportunities for resident-to-resident teaching, which does not occur on the apprenticeship rotations. In addition, certain services, like trauma/critical care and the VA service, lent themselves better to a team model because the attending staff functioned as a group practice with time on and off service for any given attending. On most of these services, the night responsibility was covered by the night float team, but at the VA, which is geographically separated from the rest of our hospitals, the team functioned traditionally using a stretch model.

Finally, we took some baby steps toward the competency model by allowing residents to take elective rotations. An elective was offered in the first year. Since we have had to pull residents back from surgical specialty services like orthopedics and neurosurgery, we wished to offer residents electives in these areas should they choose to accept them. We also offered one elective at the third year level. We would like to expand the opportunity for self-directed rotations in the future, but the scheduling issues are quite complex. We would particularly like to offer electives to senior residents who could use these opportunities to enhance their skills in areas they feel will be important to their practice or areas where they feel they need further experience to achieve competency.

The figure below indicates by resident year the percentage of time spent in the team, night float, apprenticeship, and elective rotations. We believe that the blending of these models has created an improved educational environment while, at the same time, allowing us to comply with the currently mandated limitation on resident working hours.

R5

R4

R3

R2

R1

Key:		Night Float
		Day Team
		Apprentice
		Elective

Figure 2. Portion of time spent in various training models by resident level at Northwestern in 2003-4 academic year.

Reference:

1. DaRosa DA, Bell, RH Jr., Dunnington, GL: Residency program models, implications, and evaluation: results of a think tank consortium on resident work hours. Surgery, 133 (1):13-23, 2003.