He suspects that hospitalists frequently help steering committees address certain issues, “but it may not be the individual hospitalist group that is managing or driving the quality issue for an entire hospital or health system.” The HELPS physicians knew that if they could surmount the scheduling challenge, however, they would have a terrific opportunity to learn from each other and move their individual institutional patient safety initiatives forward.
They also knew that the majority of hospitalized patients receive care in community hospitals and that the culture of a community hospital differed dramatically from a tertiary-care academic medical center. Interventions would need to be tried at varied settings, and HELPS would allow that to happen. In addition, few clinically active hospitalists have developed the research expertise to evaluate an intervention rigorously at their own institution, yet decision makers often require such an evaluation when they are considering adopting a new intervention. Using the resources of an academic center such as the University of Michigan, where evaluative and methodologic expertise is available, allowed them to facilitate a resolution for that problem. “As physicians, we are helping [smaller hospitals] by facilitating data collection and data analysis for their ongoing projects,” says Dr. Flanders.
Another challenge the hospitalists faced pertained to the relative instability of hospital medicine compared with other specialties. For instance, it is probably far easier for the members of the consortium of cardiology groups working on best practices for managing acute myocardial infarction to find time to work together. In contrast, because hospital medicine is a relatively new specialty, “we are seeing a certain amount of instability,” explains Dr. Flanders. “Of our nine involved health systems, at least two had dramatic organizational and structural changes within their hospital medicine group over the last year and a half.” One hospital medicine director left, with repercussions in several areas, and, in another group, the practice was sold to a large physician management company, which meant that the HELPS data that had been collected was no longer owned by the [original] group, making their activities more complex.
What They’ve Done So Far
Dr. Flanders is not aware of any other groups of physicians from another state who have undertaken such a project. The Institute for Healthcare Improvement’s “100,000 Lives Campaign” may be similar in some ways, focusing on a few key topic areas. The Michigan Keystone Project, which concentrated on various ICU patient safety topics, is another example of a regional consortium. “But the difference is that they basically drive the agenda,” he says. “They say, ‘This is what we want you all to do, and let’s figure out how we can all do it.’ In our case, we focused on some key areas, but we did not dictate what the project was.”
The following brief recaps represent some of the projects undertaken by consortium member sites.
Falls
The University of Michigan-affiliated Ann Arbor Veterans Affairs (VA) Hospital, where Sanjay Saint, MD, MPH, is a hospitalist, has many elderly and cognitively impaired patients, so preventing falls is a big issue. “My responsibility isn’t to prevent falls but to share these best practices with others in my organization,” he says. “In fact, each project team points out where to look with a certain problem but not necessarily what to do.”
Dr. Saint brought up the issue of mattresses with lipped/raised edges to prevent patients from rolling out of bed. Following that, at least one other consortium hospital suggested to its fall prevention group that they explore the use of these lipped mattresses.
Bed alarms were another problem. Participants shared a common problem: Often, nurses wouldn’t hear a patient’s alarm go off. Other providers might reset them, but if they neglected to let the assigned nurse know the patient was calling, the patient might get tired of waiting, attempt to get out of bed, and sustain a fall. Consequently, providers in most institutions thought the alarms were not useful. But one institution identified a novel way to use bed alarms: attaching them to the nurse’s pager. Now, when the alarm goes off, it alerts the appropriate nurse directly, and because that nurse is responsible for resetting the alarm, she knows that the patient has called and needs help.