One of the things that Jennifer Feighner, MD, cheerfully came away with at HM16 was how to better complete a task that is distinctly uncheerful but also important to any high-performing hospital: how to collect the data of the dead.
The quality improvement session “Reducing Inpatient Mortality: A Standardized Approach to Identify Preventable Deaths” demonstrated still evolving but, so far, well-performing projects that have been rolled out at Brigham and Women’s Hospital and the Duke University Health System.
“I was struck by the methodology for getting input from multiple providers and the nursing staff,” said Dr. Feighner, director of hospital medicine at Marcus Daly Memorial Hospital in Hamilton, Mont.
As the role of the hospitalist as agent of change and improvement continues to grow, the topic of quality improvement (QI) maintained a high profile at the annual meeting, with talks on the latest literature, sustaining motivation to complete projects, and dealing with issues such as handoffs and frequent fliers.
In the mortality review session, presenters set out to give details that could be a model to be used elsewhere. At Brigham and Women’s, all of the frontline clinicians are asked by email to fill out a report when a death occurs in any case with which they’ve been involved, with the Web-based reports to be completed within 48 to 72 hours of the death.
The number of deaths, the report completion rate, death “preventability,” and issues that arose for the patient during hospitalization are some of the data that are tracked. So far, the system has identified such themes as “alarm fatigue,” high oxygen requirements on non-intensive-care floors, handoffs, and transfers from other hospitals, said Kiran Gupta, MD, MPH, who completed her residency at Brigham and is now assistant professor of clinical medicine at the University of California San Francisco.
At Duke, where the mortality review system improvements have been led by Jonathan Bae, MD, assistant professor of medicine, self-nicknamed “Dr. Death,” inpatient deaths undergo a similarly comprehensive review, with an enhanced end-of-life section to cover issues particular to those cases and flags for cases that need independent review.
Dr. Gupta and Dr. Bae emphasized the confidentiality of the reviews and that they are non-discoverable in the event of litigation, which they hope give clinicians the freedom to fully report their observations.
Dr. Feighner said that her 23-bed hospital is far smaller than either Brigham or Duke, of course, but that the overall ideas can translate.
“I am the medical director of our hospitalist program, and our chief of staff and I’ve created a peer-review medical staff quality improvement committee,” she said. “So this obviously has a lot of interest to me.”
With only 4.2 full-time equivalents (FTEs) in her department, she said changes would be even easier to put into place.
“I think that will be really helpful for our peer-review committee and our quality and safety committee. I could see how we could take this and kind of revise it a little bit,” she said. “When you’re in charge of that few people, it’s easy to get processes implemented. We are more limited in monetary resources, but we make up for that in manpower-to-problem ratio, I guess.”
In another session, Jordan Messler, MD, SFHM, a hospitalist and former medical director of the hospitalist group at Morton Plant Hospital in Clearwater, Fla., confronted the startling statistic that 80% of initiatives in hospitals never meet their objectives. Hurdles such as burnout and disengagement, he said, often stand in the way of successful QI projects.
He emphasized the importance of intrinsic motivation (a sense of ownership and passion for the work) over extrinsic motivation (a fear of reprisal if something isn’t done). A step as simple as assigning a title (e.g., “head of readmissions”) can be a motivator, he said. But he also emphasized that project ideas need to be timed correctly and the ideas should ideally come from the physicians leading them.
Robert Clothier, RN, a practice manager for the hospitalist group at ThedaCare in Wisconsin, said he was struck by the lessons gleaned in a workshop on the I-PASS system of handoffs—a standardized system with a handoff sheet, studied prospectively, in which medical errors decreased by 23% and preventable adverse events fell by 30%.1 The system was created in pediatric departments but was deliberately made to be translatable to other settings.
“Instead of focusing on the outcome of the quality of the handoff, they focus on the quality of the feedback sessions,” Clothier said. “So it’s not the person giving the handoff or receiving the handoff that actually assesses it. It’s just the person who’s sitting there watching.”
He said the workshop underscored the importance of standardization, a concept with which he was familiar but which now seemed particularly vital.
“If you do the process and everybody does it the same, then it’s not only the person that’s giving the information who can do it in a very standardized way but the person who’s listening already knows what they’re going to be listening for so they hear it more clearly because they don’t have to discern what’s coming next,” Clothier said. “They already know what’s coming next.” TH
1. Starmer AJ, Spector ND, Srivastava R, et al. Changes in medical errors after implementation of a handoff program. N Engl J Med. 2014;371:1803-1812.