Caring for patients in a hospitalist-run clinical decision unit is associated with decreased length of stay without increasing revisit rates.

Background: Observation medicine is a growing field with increasing involvement by hospitalists. Little has been written regarding clinical outcomes in hospitalist-run clinical decision units (CDUs).

Objective: To determine the impact of a hospitalist-run geographic CDU on length of stay (LOS) for observation patients. Secondary objectives included examining the impact on 30-day emergency department (ED) or hospital revisit rates.

Design: Retrospective cohort study with pre- and post-implementation analysis.

Setting: Urban, academic, 600-bed teaching hospital in Camden, New Jersey.

Patients: Observation patients discharged from medical-surgical units before and after CDU opening and those discharged from the CDU after CDU opening.

Intervention: Creation of a hospitalist-run, 20-bed geographic CDU, adjacent to the ED with order sets, protocols, and priority consults and testing.

Measurements: Median LOS for observation patients was calculated for 7 months pre- and post-CDU implementation. ED and hospital revisits requiring an observation or inpatient stay within 30 days of discharge were measured.

Results: CDU observation patients had a lower median LOS than medical-surgical observation patients during the same period (17.6 hours vs 26.1 hours, P < 0.001). CDU LOS was lower than medical-surgical observation LOS in the 7 months 1 year prior to CDU implementation (17.6 hours vs 27.1 hours, P < 0.001). CDU patients had a similar 30-day ED or hospital revisit rate compared with observation patients pre-CDU.

Conclusions: Implementing a hospitalist-run geographic CDU was associated with a 35% decrease in observation LOS for CDU patients compared with a 3.7% decrease for medical-surgical observation patients. CDU LOS decreased without increasing ED or hospital revisit rates.