A relatively recent practice catching on in many different hospitalist groups is geographic cohorting, or unit-based assignments. Traditionally, most hospitalists have had patients assigned on multiple different units. Unit-based assignments have been touted as a way of improving interdisciplinary communication and provider and patient satisfaction.1
How frequently are hospital medicine groups using unit-based assignments? SHM sought to quantify this trend in the recently published 2018 State of Hospital Medicine Report. Overall, among hospital medicine groups serving adults only, a little over one-third (36.4%) of groups reported utilizing unit-based assignments. However, there was significant variation, particularly dependent on group size. Geographic cohorting was used only in 7.6% of groups with 4 or fewer full-time equivalents, and in 68.8% of groups with 30 or more FTE. These data seem logical, as the potential gains from cohorting likely increase with group/hospital size, where physicians would otherwise round on an increasingly large number of units.
As has been shared in the hospital medicine literature, groups have experienced variable success with geographic cohorting. Improvements have been achieved in interprofessional collaboration, efficiency, nursing satisfaction,2 and, in some instances, length of stay. Unit-based assignments have allowed some groups to pilot other interventions, such as interdisciplinary rounds.
But geographic cohorting comes with its implementation challenges, too. For example, in many hospitals, some units have differing telemetry or nursing capabilities. And, in other institutions, there are units providing specialized care, such as care for neurology or oncology patients. The workload for hospitalists caring for particular types of patients may vary, and with specialty units, it may be more difficult to keep a similar census assigned to each hospitalist.
While some groups have noted increased professional satisfaction, others have noted decreases in satisfaction. One reason is that, while the frequency of paging may decrease, this is replaced by an increase in face-to-face interruptions. Also, unit-based assignments in some groups have resulted in hospitalists perceiving they are working in silos because of a decrease in interactions and camaraderie among providers in the same hospital medicine group.
At my home institution, University of California, San Diego, geographic cohorting has largely been a successful and positively perceived change. Our efforts have been particularly successful at one of our two campuses where most units have telemetry capabilities and where we have a dedicated daytime admitter (there are data on this in the Report as well, and a dedicated daytime admitter is the topic of a future Survey Insights column). Unit-based assignments have allowed the implementation of what we’ve termed focused interdisciplinary rounds.
Our unit-based assignments are not perfect – we re-cohort each week when new hospitalists come on service, and some hospitalists are assigned a small number of patients off their home unit. Our internal data have shown a significant increase in patient satisfaction scores, but we have not realized a decrease in length of stay. Despite an overall positive perception, hospitalists have sometimes noted an imbalanced workload – we have a particularly challenging oncology/palliative unit and a daytime admitter that is at times very busy. Our system also requires the use of physician time to assign patients each morning and each week.
In contrast, while we’ve aimed to achieve the same success with unit-based assignments at our other campus, we’ve faced more challenges there. Our other facility is older, and fewer units have telemetry capabilities. A more traditional teaching structure also means that teams take turns with on-call admitting days, as opposed to a daytime admitter structure, and there may not be beds available in the unit assigned to the admitting team of the day.
Overall, geographic cohorting is likely to be considered or implemented in many hospital medicine groups, and efforts have met with varying success. There are certainly pros and cons to every model, and if your group is looking at redesigning services to include unit-based assignments, it’s worth examining the intended outcomes. While unit-based assignments are not for every group, there’s no doubt that this trend has been driven by our specialty’s commitment to outcome-driven process improvement.
Dr. Huang is physician advisor for care management and associate clinical professor in the division of hospital medicine at the University of California, San Diego. He is a member of SHM’s practice analysis subcommittee.
1. O’Leary KJ et al. Interdisciplinary teamwork in hospitals: A review and practical recommendations for improvement..
2. Kara A et al. Hospital-based clinicians’ perceptions of geographic cohorting: Identifying opportunities for improvement..