No signouts? No problem

A hospitalist program uses an innovative scheduling system to reduce handoffs.

Improving the handoff process is a priority in hospital medicine. But while many facilities attack the problem by focusing on standardization, Decatur Memorial Hospital, a 350-bed facility in Decatur, Ill., has gone a different route. Under the direction of David J. Yu, FACP, the hospitalist program has adapted its week-on, week-off schedule to decrease signouts to just once a week. Dr. Yu recently spoke to ACP Hospitalist about how he came up with this model, why he thinks it works, and what he'd recommend to other program directors considering scheduling changes.

Q: How did this scheduling system come into being?

This model has not hindered our recruitment even with younger physicians
This model has not hindered our recruitment, even with younger physicians.

A: When I started the program, roughly three-and-a-half years ago, I had a clean slate when it came to looking at the scheduling issue. There are so many variations to scheduling, but most programs usually choose one of two models. With a nocturnist model, there are multiple signouts, up to 14 potential signouts in a week. With the other model, there are signouts over the weekend and one or two hospitalists take the burden of the rest of the practice. There's no good way to do a signout; it's basically a bad process. What I decided to do was to eliminate it as much as possible.

Q: How does your model work?

A: We work in two teams of three hospitalists on a one-week-on, one-week-off schedule. During the one week on, each hospitalist takes a turn being on call and doing admissions, like shiftwork. For the entire week, though, 24/7, they take their own pager call for the patients they admit. There's no signout of patients for the entire week until Monday morning, when the new service comes on. All three hospitalists working a particular week come in on the weekend to round on, write notes for and discharge their own patients. Each hospitalist consistently signs out to the same hospitalist on the other team each Monday.

Q: Has the model been successful?

A: It's worked very well. Since there's only one signout a week, our signout process is very long and very involved—a signout for 10 to 15 patients might take half an hour to an hour—and very interactive. You're not in a rush to get through the signout since this is the only signout for the whole week, so you try to glean as much information from the other service as you can. It really gives the hospitalist ownership of his patients.

Q: What are the benefits to increasing that sense of ownership?

A: In my view, it can help decrease length of stay. A lot of hospitalist programs have so many different hospitalists involved during the week that it's almost medicine by committee. For example, if you're cross-covering on Sunday for a very complicated patient with multiple issues that need to be set up for discharge, I think the strong temptation is to just wait until Monday, until the other hospitalist who knows the patient better comes back.

When I was an attending at a teaching hospital, they had a night float system with the residents and there were multiple signouts of patients. I could never get a clear answer of who was actually in charge of my patients from a residency point of view. If I rounded early, they would say, “It's still a night float, we haven't signed that patient out.” If I rounded later in the day, they'd say, “I signed that out to night float, you'll have to call them.”

That's the same issue that our hospitalist movement is going to face with multiple signouts and using a nocturnist or cross-coverage model on the weekends. Cross-coverage is great when you're the one signing out, but it's horrendous if you're the person receiving the signout. I think no one will dispute the fact that signouts and cross-coverage are where the vast majority of the mistakes will happen. And not just mistakes; think of all the unnecessary laboratory tests and precautionary ICU transfers that occur because you don't have intimate knowledge of patients during the cross-coverage time.

Q: How did the hospitalists respond to this scheduling model?

A: We were very fortunate. Three of the original four hospitalists in our group came from private practice, and before I became a hospitalist I was in private practice for nine years, so we had a strong sense of what it meant to be on call for our patients. But to my surprise, this model really has not hindered our recruitment, even with younger physicians. Once they understand the model, they've come to accept it and it's worked pretty smoothly.

Q: How about the other staff?

A: We did have to kind of break the culture a bit in some cases. On the weekends, our operators and nurses sometimes do still page the physician who's on ED coverage insisting that they be told about the patient. That just tells me how ingrained cross-coverage has become in our medical system.

But overall, the nurses love it. Even in the middle of the night, when they page a physician about a patient, they're reassured that the doctor knows exactly what's going on. They're not wasting time educating that doctor, who might be cross-covering, about the patient. It makes an incredible difference to the hospitalist as well. It's so simple when you get called at three in the morning about a patient you know, versus trying to figure out what's going on with a patient who was recently signed out to you.

Q: How do the patients feel about the scheduling system?

A: As hospitalists, it's critical that we bond instantly with our patients, especially with the sicker patients, so that when there is a problem they see a familiar face. With this system, you get to know the patient really well and the family gets to know you really well. I've seen families' frustration when they see even two or three different physicians in a 24-hour period, wondering who is in charge of their loved ones' care. When they see the same doctor, they get a consistent story. At our hospital, if a patient has been there for six days and it's between Mondays, they're consistently dealing with the same attending every single time. That increases our bonding with the family and with the patient, which I think significantly reduces our medical and legal liability, too.

Q: Do you think this model improves career satisfaction overall?

A: Yes. You feel that you're truly the patient's attending physician throughout the length of the hospitalization, that whether good or bad you're going to have full responsibility, that you can't shift the responsibility to a nocturnalist or to the next physician coming on in the next few days. Whether medical directors choose to believe it or not, the way you set up the system really affects your hospitalists' behavior. I don't think you can underestimate how hospitalists' behavior is affected by scheduling.

Q: Do you foresee any potential problems with this scheduling scheme? Do you think it would work for everyone?

A: I'm not saying this is the perfect model for every hospital system. The town of Decatur is about 100,000 people, and there's no rush-hour traffic. All of the physicians live about 10 or 15 minutes away, so cross-coverage and coming in to see your own patients is not a big issue. This system might not work well in a major metropolitan area, where the hospitalists live in the suburbs or traffic is heavier.

But I do think that all medical directors can develop their own unique systems by thinking outside the box, looking at their own cultures, and developing their own ways of increasing patient ownership. Why do we always have to say, ‘This is the way it's always been done’? That's not a good enough answer. The latest studies show that length of stay doesn't differ much between hospitalists and primary care physicians, and a part of that might be just the way these systems are set up. Eroding the ownership of our patients really can hinder how effectively we can deliver health care as hospitalists.