Tulane Internal Medicine Residency, Program Information

Tulane Internal Medicine: Fully Compliant with the Institute of Medicine’s Duty Hours Regulations since 2008.

In 2008, the Institute of Medicine released its recommendations for duty-hour reform in residency training, sparking a controversy across the country. While most programs resorted to further fragmenting their system by introducing multiple shifts and handoffs, the Tulane Team took the harder road of completely retro-fitting the training program to enable duty hour compliance, while still maintaining those principles essential to becoming a great internist.  Here’s how we did it…

If you want to read more about the rationale that went into our innovative solution (and the problems that are inherent in the systems used by most other residency programs) click here...



The secret is in team-based care: the Tulane firms have increased the on-call team size to provide the team with the necessary resources to ensure continuity of care, but also to insure compliance with the duty hours… and to ensure that each team member gets appropriate rest. The Tulane On-Call Team is composed of five team members: The Attending, The Ward Resident, The Moondog, and two Interns. In parallel to this team is the Orion Team. It’s easiest to understand this by looking at the anatomy of the On-call Day. Follow along with the figure below:

3:00 AM: The ER "Lockout." The call day technically begins at 3 AM. Any patient admitted prior to 3 AM goes to the team on call the day before. Patients who are admitted from 3 AM to 6 AM are held in the emergency department, awaiting the arrival of the "Moondog" (see below) who will begin admitting and caring for these patients, who will then fall under the stewardship of the on-call team.

6:00 AM:
The call day begins at 6 AM with THE MOONDOG. The moondog is a resident member of your firm… linked to your ward team for the month. The moondog kicks off the start of the call day at 6 AM… taking admissions by herself from 6 AM to 8 AM.

7:00 AM: At 7 AM the TWO INTERNS on the admitting team arrive and join their Moondog resident in taking admissions. They will learn about any admissions that the Moondog has admitted in the previous hours, but their primary task will be in taking new admissions. This overlap ensures that management of active issues in the patients admitted by the Moondog will be appropriately managed, and that each new admission becomes fully integrated into the ward team that will be responsible for their on-going care.

10 AM: At 10 AM, the WARD RESIDENT will arrive. On-call rounds begin at 11 AM, and include the attending, the Moondog resident, the ward resident and the two interns. This is a critical transition point, and as illustrated in the diagram below, the safety of the transition is ensured by having ALL team members present, and by couching the transition in formal, educational rounds. During this time, the ward resident will learn about the patients the Moondog resident and the two interns have admitted in the first 8 hours of the call day (i.e., from 3 AM to 11 am). The beauty is that the team is present for all of the active patient care issues that are inherent in the first few hours following an admission. The Moondog will remain with the team until two or three PM… or however long it takes to ensure that the care has been carefully transitioned to the team. After that, he goes home. His day is finished. The ward resident and the two interns continue with their call day, fully aware of the patients that have been admitted thus far, and thus prepared to continue their care through the night.

6 PM: The teaching hospitalist (i.e, the attending) returns at 6 PM to do evening rounds. During this time of the day, the pressures from pagers, nurses calling, etc are less, and the attending has the protected time to teach in a relaxed setting. And with seventy percent of the call day now elapsed, some of the early admissions may be ready for discharge… thereby decompressing the post-call day to follow. And of course this early discussion of patients accelerates care, enabling the team to make early interventions (tests, consults, discharges) that would ordinarily have been rolled over to the post-call rounds. This greatly decompresses post-call rounds, because 16 hours of the call day will already have elapsed by this point. This means that post-call rounds can have no more than 8 hours of new admissions to discuss.

10 PM: At 10 PM, the interns' 16 hour shift expires, and they go home. (See the diagram below). The admitting resident remains on call to ensure that, even though the interns have left, there remains continuity within the team. Whatever happens over night, the resident is perfectly positioned to pass along those events to the interns such that they learn from the interventions they made at the time the patient was admitted. To help the admitting resident, the one of the four Orion residents will arrive at 10 PM. (see below)

10 PM to 3 AM: The ACGME allows for a 24-hour call day for upper-level residents, and though it is not required, Tulane believes that any resident working more than 16 hours should have protected sleep time. To make this happen, the Tulane Team invented the Orion Service (see below). The Orion Team is a four-resident team (no interns) that works directly with a supervising hospitalist. The four residents share a q 4 call cycle, each night from 10 PM to 3 AM. The Orion on-call resident admits patients from 10 PM to 3 AM, and continues to manage those patients (just as a hospitalist would) the following day and through the patient’s hospital course. With this system, there is support for the on-call resident, but not at the expense of a “night admitting” team (as is standard in other residency programs) dumping their overnight admissions on to the post-call or other ward teams.

The Orion support allows the on-call resident to continue to manage the patients admitted up until 10 PM (i.e., continuity of care for the newly admitted patients), but since the admits stop at 10 PM, it also permits her to get some sleep even while on call.  The resident will awake at 8 AM to start post-call rounds at 9 AM, joined by the two interns who will return (10 hours after leaving their shift). Overlapping all of this, of course, is the Knightrider resident, fielding all of the calls for the other three teams’ services not on call.

6 AM: The next-team’s Moondog will arrive at 6 AM to start the next call day. (The pink shaded box in the diagram below illustrates when the admission clock begins and ends for the on-call team).

9 AM: The attending will join the two interns and the resident to start their post-call rounds at 9 AM.

1 PM: The resident goes home.

4 PM: The "post-call" interns leave no later than 4 PM…. Enabling them to catch the educational components of the day (i.e., "Afternoon Delight" or "Friday School") even on post-call days. No educational time is wasted in the Tulane system!

Click on the image for a full view of the On-Call Ward Schedule.


There are four big advantages to the Tulane System.

1. First, the Moondog is paired to her firm’s team… she will be with the team each and every call day. The team’s personnel remain the same throughout the four-week ward block: the ward resident, the Moondog, the two interns, the attending… all the same team throughout the month. And this builds a team-relationship that enables better communication, especially during the transitions, and ensures that she receives teaching as a part of the team. What errors she made in the admission of patients will be addressed as part of the team’s teaching. It also absolves the “loneliness” that is inherent in most programs “night float” systems.

2. Second, the system is built to ensure that there is overlap during transition of care points, (i.e., the overlap of the moondog and the interns from 6 AM to 11 AM, and the moondog with the ward resident/interns from 10 AM to 2 PM)… and these overlaps are couched in educational discussions (i.e., attending rounds). There are no hard-stop sign-outs. For the intern/resident coming on, there is time even after the transition-of-care has occurred to come back and ask question of the resident who has transferred the care. For the resident/intern going off, there is no pressure to rush away from the transition-of-care… adequate time and attention can be given to a detailed and thoughtful signout. Further, the resident/intern going off has the luxury of seeing how his or her interventions played out… an essential component to learning medicine.

3. Third, the “cross cover admission” role of the Moondog starts the call day instead of ending it. Most other systems bring in the “nightfloat admission resident” late at night to finish the call day. The downside to that model is that patients admitted late at night (say, 6 am) do not have a team around to care for them during the critical 12 hours after their admission … they are left in the lurch because post-call team goes home after post-call rounds. In the Tulane System, any patient admitted during the call day has the on-call team available to them for at least 12 hours after their admission, insuring strong continuity or care during this critical time. Have a look at the arrows at the right of the diagram. The start of each arrow is when the last patient was admitted; the end of the arrow is the closure of the critical 12 hours after admission. Note how the on-call team in this system is available through these 12 hours, regardless of when the patient is admitted. And finally, because the team does not rely upon random other residents to come in for nightfloat admissions, the other residents on the team who are doing their “diastole months” (i.e., their electives) are protected from having these electives disrupted by being called in to do so.

4. And of course, the system ensures absolute compliance with the ACGME AND the IOM requirements of a maximum of 16 hours prior to a 5 hour protected sleep time… .and compliance with the ACGME admission caps (The Moondog takes some of the pressure off of the Ward Resident’s 10 patient cap by doing the first 8 hours of the call… functionally reducing the Ward Resident’s exposure to admissions by 33%).

5. And finally, the system enables providing cap coverage, cross coverage, and work hours fidelity without resorting to pulling in the other firm’s residents who are enjoying their protected “diastole” time… catching up on their sleep, spending time with their families on weekends and evenings, and engaging in their tailored curriculum (electives) to launch their careers. At Tulane, when you are on Systole (ward months), you are on Systole. When you are on Diastole (elective months), you are on Diastole… not the “high-output heart failure” intrinsic in most programs that have residents on some sort of systole (call months, nightfloat, night admissions, short call, etc) all of the time.


The ORION service is so named because it rises with the night sky. It is comprised of four upper-level (PGY 2 and 3) residents, and staffed by a teaching hospitalist. While the remainder of the Tulane inpatient rotations are all teaching services, the Orion service allows residents to experience direct, hands-on management of their own patients, very similar to what they might experience after residency. It is a high-autonomy service, with the residents calling all of the shots.

The ORION rotation is comprised of four elements, each occurring on successive days of the q 4 call schedule. On the first day, the resident assumes the duties of the admitting Orion resident (see above), taking admissions from 10 PM to 3 AM. On Day 2, the ORION resident manages her admitted patients post call. All patients that the ORION residents admits will remain with her throughout their hospital stay; this ensures that there is no fragmentation of admitting patients and then handing them off to someone else. On day three, the ORION resident plays the role of “Moondog.” The “Moondog” is supervises and teaches the admitting interns (see above) until their on-call resident arrives to take over. Thereafter, the Moondog plays the role of “ER Throughput Director,” making decisions as to where admitted patients should go (i.e., ICU, CCU, wards, home, etc). This is a valuable skill that the Tulane residents acquire, preparing them for real-world practice upon graduation. The fourth day of their cycle is a day off.


Does Tulane have a Nightfloat System? KNIGHTRIDER!!!

We don’t have “nightfloat” but we do have “Knightrider.” (David Hasselhoff not included). The difference is mostly in the name…. which is important, actually. Because why would you settle for being called “the nightfloat” (yawn, boring!)… when you could be “THE KNIGHTRIDER!” Much more satisfying.
The other difference is that the Knightrider does not take admissions. His job is solely to provide cross-coverage support from the other firms’ patients. The Knightrider begins at 3 PM…. Starting the shift with an educational hour of "Afternoon Delight"…. And then takes sign-out from the three non-call teams. The Knightrider answers calls for these patients until 7 AM the following morning (i.e., a 16 hour shift) and then hands back the patients to their respective teams (i.e., as the interns from the other teams arrive). There are three Knightriders in each week… on night one, you'll work at Charity Hospital. On night two, you'll work at Tulane/VA Hospital. The third day/night is off. Every third day during the Knightrider week is a day off, enabling you to catch up on your sleep, reading, and life!

There is a formal curriculum built for the Knightrider to ensure that this rotation is an educational experience. And remember, Tulane took a bold step forward in ensuring patient safety and quality by removing the Knightrider duties from the interns…. Only experienced PGY-2 and PGY-3 residents are allowed to staff the Knightrider role, providing both additional experienced coverage in house over night, but also instruction to the interns (signing out to/signing in from) as to best practices in transitions of care.



In any systems change, there are two outcomes: the measures (i.e., the intended results of the change; i.e., compliance with the duty hours), and countermeasures (i.e., the consequences or cost of the change. That is, as a program becomes compliant with the duty hours, what was lost?). To make an effective change, both measures must be carefully considered. For The Tulane Team, we knew what we wanted in the form of the “measures:” We wanted to be IOM compliant. But we had to spend some extra time thinking about what we didn’t want to loose (i.e., the countermeasures) as we made that change.

1. We didn’t want to introduce more handoffs. One of the core philosophical principles of The Tulane Team is that we do not believe in frequent handoffs. With each handoff comes the risk for mis-communication, and the resulting risk of compromising patients safety. Besides, Internal Medicine is about continuity, and a patient in his time of greatest need (i.e., his admission to the hospital) must have the assurance that the admitting team will be with him through his hospital stay. When a patient is admitted to the hospital, he doesn’t want his doctor to whisk into the room, write some orders, and then look at the clock and say “Oh, sorry I have to leave you in the time of your greatest need… but it’s 3 pm, and my “short call” is up…. Hope things work out for you. I’ll see you tomorrow!” Nor does he want his doctors saying, “Yeah, I’ll be your doctor for the next four hours during the night… but tomorrow, I’ll be handing you off to a completely different team.”

2. We didn’t want to introduce more inefficiency into the system. Past the patient’s concerns (which should be paramount), the short-call/long-call system never worked anyway. Short-call looked good on paper- “I’ll be on call from 8 am to 3 pm,” but in reality short-call quickly turned into “long call.” Why? Because any internist worth her salt is not going to do a 10 minute admission for the patient that comes in at 2:50 PM. She’ll stay with that patient for as long as it takes, and that usually means that short call becomes “long call” with the resident being there until 10 PM (and “long call becomes “longest call”). Further, the receiving team (if they are great internists) are not going to settle with just taking the previous shift’s word on it… they are going to re-do the history, exam, and clinical decision-making… and this doubles the work.
In short, multiple shifts (especially when the shift works are not a part of the team) introduces great “fragmentation” into the system. FRAGMENTATION refers to the energy lost from a system every time there is a break in the circuit, and having multiple shifts is just that.

3. We didn’t want to wreck the resident’s “diastole” (elective) experiences by bringing them in for nightfloat/short call all of the time. "Fragmentation” extends to the night-float as well. Again, it looks really good on paper…. At some point in the night, someone will come in to do the admissions. Wow, fantastic! Until you realize that it is you coming in on your elective months (you know, that month where you really want to figure out if cardiology was for you… or that month that you really wanted to do some research or impress the subspecialist to get the greater letter of recommendation. Yeah, functionally, you won’t be doing those months in most systems, because you will be perpetually called in to do nightfloat or short-call). Add in the fact that nightfloat in most programs is also ultimately non-educational… you are on your own with no attending assigned to teach you; the result is that you end up making the same mistakes again and again. For nightfloat systems that take admissions overnight and then sign them out the following morning, you also introduce the pain of multiple handoffs, the pain of trying to re-do an admission in the morning (the nightfloat work doesn’t really save you much time), and the loss of quality of care that goes with each handoff. The Tulane Team does not believe in calling in someone from their elective (diastole months) to do five or six hours of admissions just to sign them out to a day team (Who will have to re-do the admissions in order to properly care for the patient) and then leave without receiving any teaching at all.

So as we made the change to become IOM compliant, this was our first countermeasure: WE DIDN’T WANT TO FRAGMENT CARE WITH MUTLIPLE HANDOFFS, because it is not good for patient care and it’s wasteful and inefficient. . In the Tulane system, if a team is on-call, then that team is on-call. If a team is not on-call, then they are not on-call. Further, all team members are part of a “team,” and that fulfills our mission to team-based care.