Related Articles |
Does Reducing Inpatient Length of Stay Have Upstream Effects on the Emergency Room: Exploring the Impact of the General Internal Medicine Care Transformation Initiative.
Acad Emerg Med. 2016 06;23(6):711-7
Authors: McAlister FA, Bakal JA, Rosychuk RJ, Rowe BH
Abstract
OBJECTIVE: The General Internal Medicine (GIM) Care Transformation Initiative implemented at one of four teaching hospitals in the same city resulted in improved efficiency of in-hospital care. Whether it had beneficial effects upstream in the emergency department (ED) is unclear.
METHODS: Controlled before-after study of ED length of stay (LOS) and crowding metrics for the intervention site (n = 108,951 visits) compared to the three other teaching hospitals (controls, n = 300,930 visits). Our primary outcome was ED LOS for GIM patients but secondary outcomes included ED LOS for all adults and ED crowding metrics.
RESULTS: The GIM Care Transformation was associated with an additional 2.8-hour reduction in median ED LOS (from 25.6 hours to 13.5 hours) over and above the 9.3-hour decline (from 30.6 hours to 21.3 hours) seen in the three control EDs for GIM patients who were hospitalized (p < 0.001). As less than one in 30 ED visits resulted in a GIM ward admission, the median ED LOS for all patients declined by 15 minutes (from 4.6 hours to 4.3 hours, p < 0.001) in the control hospitals and by 30 minutes (from 5.7 hours to 5.1 hours, p < 0.001) at the intervention hospital pre versus post (p = 0.04 for the 15-minute additional reduction, p < 0.001 for level change on interrupted time series). Other metrics of ED crowding improved by similar amounts at the intervention and control hospitals with no statistically significant differences.
CONCLUSION: Although the GIM Care Transformation Initiative was associated with substantial reductions in ED LOS for patients admitted to GIM wards at the intervention hospital, it resulted in only minor changes in overall ED LOS and no appreciable changes in ED crowding metrics.
PMID: 26850577 [PubMed - indexed for MEDLINE]