One might say that a good handoff is like pornography. Just in the sense that you know it when you see it, of course!
But given the increasing frequency and importance of handoff communications, some leaders in medicine have been searching for more precise ways to assess the quality of these exchanges.
Vineet M. Arora, MD, FACP, associate professor of medicine at the University of Chicago, is a leader in this area. Along with colleagues, she authored studies about handoff evaluation that were published in the March and April issues of the Journal of Hospital Medicine. In the first study, a tool to assist peer clinicians or outside observers in assessing handoffs was developed and tested. In the second, the results of peer handoff evaluations among residents were examined.
Dr. Arora recently spoke with ACP Hospitalist about these studies and how they could change practice for medical educators and hospitalist program leaders.
Q: What motivated you and your colleagues to study evaluations of handoffs?
A: One of the things that motivated us was this really big push by accreditation organizations, including the ACGME [Accred-itation Council for Graduate Medical Education] and Joint Commission, to be monitoring handoff quality. That sounds great, but we don't really have good tools to evaluate handoff quality that have been tested. So we applied for a grant from the Agency for Healthcare Research and Quality to actually develop and test tools to measure handoff quality.
Q: What are some of the most significant conclusions you can draw from your research?
A: There's a lot of interest in having peers evaluate handoff quality. Direct observation [by a non-peer] is very difficult because the handoff often occurs at the end of a shift or late at night. [Peers] are already there, and they're the stakeholder that's most affected by the quality of the handoff. A critique of peer evaluation is that maybe you're afraid to evaluate your peers critically or you might be too easy on them.
We created this end-of-rotation evaluation. It was thinking back on your month—how did this person do on handoffs? It is possible to have peers do this, and there are some interesting trends. The interesting things that we found were more experienced interns were rated higher, and there was a correlation with parameters related to workload.
Q: How did workload affect handoff quality?
A: Those that were post-call and had the greatest numbers of new patients on their signouts were also rated more poorly. That's important, because that's usually the most active handoff—from the person who just admitted all night and is leaving in the morning. We also noted that when everybody was at a community hospital with a lower census the handoffs were rated better.
Q: How should program leaders act on these findings?
A: I would say that the main take-home piece of advice is it's possible to incorporate peer evaluation into your program and then, by monitoring the handoff quality, it's possible to target your improvements to the handoff that is most at risk. For example, now that we know that one of the handoffs that is least satisfying is from the night intern, we could target interventions to that area.
Q: Have you targeted any interventions in your hospital?
A: We've done a lot of educational programming and interventions. We do a lot of training for the handoff.
Most interventions focus on the sender: Let's have the sender use if/then statements or to-do statements, or make sure that they've got a mnemonic, or use some kind of structured format. The truth is even in those perfect situations, the receiver may be totally out to lunch and not paying attention due to distractions or interruptions or the fact that they are not using active listening behaviors.
In our newest work, we've seen that the receiver may not always have the best listening behaviors. We had a paper in BMJ Quality and Safety on listening behavior. This work was fueled by an earlier study that showed that in 60% of handoffs, the most important piece of information was not transmitted, despite the sender believing that it had been. We looked to quantify what receivers are really doing. We found that a high fraction of them are affected by these external distractions.
I would argue that more interventions need to be happening not only to teach people how to send a handoff—we've been doing that for a while—but to teach people how to listen and receive a handoff, especially in an era of technology and increasing distractions on the wards.
Q: What are other next steps in this research?
A: We'd like to link evaluation to outcomes. Are handoffs that are rated higher associated with better patient outcomes? That's a pretty tricky thing to try to do, given that patient outcomes are affected by a lot, but we're working on thinking about that.
Another next step is to develop simulation programs to teach people how to receive handoffs. How do you really bundle a set of behaviors for receivers? They say doctors are not always the best listeners. Medical education is very good at coaching people to give information, but we have not really focused on teaching them how to receive information.
It's especially important for receivers to take an active role, because not only are they primarily responsible, but we do see a lot of information overload in handoffs. So that's another next step: How do you break through all the information overload? How can the receiver tailor what they're receiving to enhance their memory and action?
Q: How does your work apply specifically to hospitalists, especially program leaders?
A: Our listening behavior work was actually done in hospitalists. The peer evaluation work was actually done in residents. Translating this to hospitalists is interesting. We created a tool called the Handoff CEX, which hospitalists used to evaluate each other at end of shift.
What we noticed was that external observers were more harsh than peers. But they did track together. And that's important because, if you're using peer evaluations, you should expect some score inflation. If somebody's getting a 7 or an 8 out of a 9-point scale, you can't assume that they're doing a great job. You need to look at the trend over time.
The work about listening behavior in hospitalists highlights that program directors of hospitalist programs can do a few things to improve the quality of their handoffs. Clinicians arriving late was a major distraction. If somebody starts at 7:00 and somebody leaves at 7:00, somebody is always late and somebody else is time-pressured to get out. Overlapping the shift and allowing space for handoff, you can ensure that a proper handoff is taking place.
A lot of times the handoffs were not interrupted by pagers or other people. They were interrupted by side conversations. Those side conversations often were taking place because people needed a place to vent, a place to socialize. One of the things program directors could do is create space for the hospitalist program to express their views about the fact that it's very difficult to order a certain radiology test or to socialize. That way, when the handoff starts, it's all hands on deck.
One of the things that we have been using here is called the sterile cockpit rule. During a certain portion of an airline flight, the pilots are supposed to talk only about the flight. Pager interruptions are bound to occur, but there are a lot of other interruptions. They are controllable, but they are not being controlled right now. Start thinking about: When does the handoff start? When does the handoff end? And what do I need to do to make sure I have an “all hands on deck” approach?