For a medical resident, the first day on a new service is never smooth.
You’re on a new floor with new nurses, other new residents, new supervisors and, of course, new patients — sometimes two dozen of them, seriously ill, whom you’re meeting for the first time and for whom you’re expected to make important, sometimes life-altering decisions.
You’ve read about them, of course, and perhaps had a conversation with a departing colleague about their recent clinical course. But it always feels as though you’re starting in the middle of a long movie after reading a plot summary of the first half. You have a general sense of what’s going on — who the characters are and what has happened to them — but your understanding is long on overview and short on nuance.
Sometimes a previous resident’s description of a patient is spot on. Other times, the patient isn’t at all the person you imagined. (I thought she’d look older, I muse. And wouldn’t have a facial droop.)
It takes a few days before the new medical team is functioning smoothly. We need to learn each other’s quirks, strengths and shortcomings. As a senior resident, I’m asking questions such as “How much supervision does our medical student need? How strong is the intern’s understanding of heart failure? Is our attending a micromanager or a ghost?”
Care transitions — when one doctor or medical team takes over for another — have increasingly been recognized as a vital but challenging aspect of medical care. Typically, at the end of a two- or four-week block, a team of doctors — interns, residents and attending physicians — passes responsibility for a group of patients to a new team. As someone leading these transitions, I find the process varied and haphazard, a time when important details and intimate clinical understanding are lost in translation.
Research has shown that handoffs between shifts can be dangerous, and efforts to standardize these transitions have led to measurable improvements in patient safety. But less attention has been paid to what happens when an entire team leaves and a new team comes on at teaching hospitals, which train new doctors and provide about half of all hospital care.
Now a new study suggests that whatever happens, it isn’t good.
Researchers studied data on more than 230,000 patients at 10 university-affiliated Veterans Affairs hospitals to see how they fared when the residents caring for them transferred responsibility to a new team. These patients had a wide range of medical conditions, including diabetes, heart failure, cancer and liver disease, and were admitted to general medicine services as well as to subspecialty floors and intensive care units.
The data showed that patients who underwent transitions were twice as likely as others to die during their hospitalization: Four percent of transition-group patients died, compared with 2 percent of those who didn’t experience a transition. And the ominous link persisted after patients left the hospital. Three months later, 23 percent of transition-group patients had died, compared with 14 percent of the others.
“I wasn’t surprised we found an association,” Joshua Denson, the study’s lead author, told me. “But I was surprised by the magnitude. These aren’t just a couple of mishaps. We may be doing something during transitions that changes the trajectory of patients’ health in the long run.”
There are many plausible explanations for these findings. The most obvious is that important clinical information is lost or overlooked when one medical team takes over for another. But more-subtle factors may also be contributing.
For example, say I’m told by departing residents that an older woman’s shortness of breath is caused by worsening heart failure. I continue the treatment they started, but she doesn’t get better. Because I’ve anchored on their diagnosis, it will almost certainly take me longer to recognize that pneumonia is actually causing her symptoms than it would have had I started fresh and reasoned through the diagnosis myself. Meanwhile, she isn’t getting the antibiotics she needs.
But it’s also not clear that care transitions are entirely to blame. Patients who remain hospitalized during transitions may simply be sicker. Before rotating out of a service, we doctors often try to finish workups and discharge patients to reduce the burden on the next team. This means that sicker patients and those with difficult social circumstances are more likely to experience transitions; the easy ones get discharged. So it may not be that the transition itself is responsible for higher death rates, but rather that patients with extended and complicated hospitalizations are more likely to die regardless.
“We all try to clean house before a service change,” said Vineet Arora, an associate professor of medicine at the University of Chicago and co-author of an editorial accompanying the study. “We know handoffs are risky, especially when you have more patients. So there’s a pressure to discharge, and the patients left are generally sicker.”
Indeed, after accounting for how old and sick patients were, researchers found that the odds that a person would die in the hospital after a transition dropped from nearly 100 percent greater than the odds of other patients to just 20 percent greater. Other intangible factors could be contributing as well. Are transition patients also more likely than others to be homeless? Are they less likely to have family members who can support them through a tough time?
Still, few doctors would argue that transitions aren’t a precarious time for patients or that transitions wouldn’t benefit from more-concerted attention.
One important step would be standardizing the handoff process, which has been shown to improve end-of-shift transitions but hasn’t rigorously been tried for end-of-rotation transitions. Hospital protocols for transition and formal education on end-of-rotation handoffs vary widely. A resident might communicate important history and tasks verbally, in person, in writing — or not at all. Hospitals should ensure that residents have dedicated time and standard templates for communicating about consultant recommendations, discharge planning and important patient conversations. Some programs are experimenting with doing handoffs at the bedside or even having a resident from the prior team participate in rounds with the new team on its first day.
Another change might involve staggering when residents rotate off service so at least one team member has a longer perspective about the unit’s patients. And greater patient and family involvement will also be important.
“Most patients aren’t even aware these transitions are happening,” Denson said. “We need to do a better job of letting them know and engaging family members who can ask questions during high-risk periods.”
Care transitions are an unfortunate but inevitable part of providing medical care. Experience and data suggest that patients are particularly vulnerable during these periods but also that there’s much more we can do to minimize harm and maximize safety.
“There’s been major investment in end-of-shift handoffs, but much less in end-of-service transitions,” Arora said. “But these are permanent handoffs. Residents are leaving and not coming back. It’s risky, and it deserves more attention.”
Khullar is a resident physician at Massachusetts General Hospital and Harvard Medical School. Follow him on Twitter: @DhruvKhullar.