Impact of a Predominant Geographic Model on Outcomes for Hospitalist Patients

1UMDNJ‐Cooper University Hospital, Camden, NJ
2UMDNJ‐Cooper University Hospital, Camden, NJ
3UMDNJ‐Cooper University Hospital, Camden, NJ
4UMDNJ‐Cooper University Hospital, Camden, NJ
5UMDNJ‐Cooper University Hospital, Camden, NJ
6Hospital of the University Of Pennsylvania, Philadelphia, PA

Meeting: Hospital Medicine 2008, April 3-5, San Diego, Calif.

Abstract number: 42

Background:

Concern for patient safety and efficient throughput has been the impetus for the development of many initiatives aimed at improving hospital processes. It is our hypothesis that the traditional assignment of patients may adversely affect outcomes by interfering with teamwork between hospitalists, nurses, and consultants. Our hospital (450‐bed teaching institution) has implemented a partial geographic system (4 of 6 teams) where each admitted patient is assigned to a team whose patients reside only on that floor.

Methods:

We performed surveys before and 7 months after the change for residents/hospitalists, nurses, and consultants. Our goal was to assess whether the change would affect their perception of delivery of care. We also assessed hospital length of stay (LOS) and death. Data were analyzed with f‐test and chi‐square analyses.

Results:

After the change, consultants (77 surveyed before and 41 after) found it easier (“meets/exceed expectations”) to reach the team responsible for the patient (28.6% vs. 58.5%, P = .003), that teams more frequently read their recommendations (15% vs. 34%, P = .03), that the relation with the teams was more collaborative (26% vs. 70%, P < .001), that questions were more clearly stated (7% vs. 22%, P = .018), and that consultations were placed earlier in the course of the hospitalization (16% vs. 34%, P = .034). Nurses (97 before, 49 after) found it easier to find physicians responsible for patients (8% vs. 55%, P < .001), easier to communicate with house staff in a timely manner (13% vs. 49%, P < .001), that orders were placed in a more timely manner (15% vs. 32%, P = .02), that their input was more frequently taken into consideration (5% vs. 22%, P = .03), and that they were more informed about plans (5% vs. 22%, P = .03) and discharges (5% vs. 31 %, P < .001). However, this change did not improve their ability to actively participate in rounds. Residents (47 before, 38 after) expressed having more time to spend with patients (13% vs. 36%, P = .017) to complete the medication reconciliation form (2% vs. 19%, P = .011) and more often cited the feeling of working as a team (4% vs. 28%, P = .004).There was no change in LOS (2.9 vs. 3) or death rate (2.1%).

Conclusions:

Geographical assignment of patients improved many aspects of communication between health care providers without compromising efficiency or outcome measures. Future studies would aim at measuring the impact of these changes on the long‐term outcome of patient satisfaction and compliance with bundles of care and to learn if this system is translatable to other hospitals.

Author Disclosure:

E. Kupersmith, Ortho‐McNeal, speaker bureau; A. Chaaya, none; E. Cerceo, none; J. S. Rachoin, none; M. Kreidy, none; S. Gandhi, none.

To cite this abstract:

Kupersmith E, Gandhi S, Rachoin J, Kreidy M, Chaaya A, Cerceo E. Impact of a Predominant Geographic Model on Outcomes for Hospitalist Patients. Abstract published at Hospital Medicine 2008, April 3-5, San Diego, Calif. Abstract 42. Journal of Hospital Medicine. 2008; 3 (suppl 1). https://www.shmabstracts.com/abstract/impact-of-a-predominant-geographic-model-on-outcomes-for-hospitalist-patients/. Accessed May 26, 2019.

« Back to Hospital Medicine 2008, April 3-5, San Diego, Calif.