Impact of a Rapid Response Team on Intensive Care Unit Readmission Rate

Background:

Rapid response teams (RRTs) assist in the assessment, treatment, and triage of hospitalized patients with physiologic instability, but their effectiveness in reducing transfers to the intensive care unit (ICU), cardiopulmonary arrests, and in‐hospital mortality has been questioned. Rounding by an RRT on patients recently discharged from an ICU is uncommon in American academic medical centers, and little is known about the effect of such a model on the rate of ICU readmission. We investigated the effect of proactive rounding by an RRT on the rate of ICU readmission of all adult patients discharged from an ICU at our institution over a 42‐month period.

Methods:

The RRT at UCSF Medical Center, composed of a critical care nurse and respiratory therapist with backup from a critical care physician, has, since its inception in June 2007, proactively rounded at least once on all patients discharged from an ICU within the previous 12 hours. We conducted a retrospective pre‐/postintervention study to investigate the effect of proactive rounding by the RRT on the rate of ICU readmission; secondary outcomes included the average length of ICU stay and in‐hospital mortality. Hospital administrative databases were queried to identify all adult patients requiring intensive care from January 2006 until June 2009, their admitting service, ICU length of stay, vital status at the time of hospital discharge, and whether billing data indicated 2 noncontiguous ICU slays during 1 hospitalization. Data were analyzed using analysis of variance and t tests to compare mean outcomes in the preintervention and postintervention periods

Results:

We analyzed 17 months of data prior to initiation of the RRT and 25 months of data following initiation of the RRT. The ICU readmission rate for all services was 6.0% (321 readmissions of 5320 total patients) prior to introduction of the RRT and 6.5% (494 readmissions of 7605 total patients, P = 0.36) following introduction of the RRT. When the data were stratified by admitting service, there remained no statistically significant differences in ICU readmission rates before and after initiation of the RRT, with the exception of the neurosurgical service, for which there was a small increase in the rate of ICU readmissions (4.2% pre‐RRT versus 6.0% post‐RRT; P = 0.02). Introduction of the RRT had no significant impact on the average length of stay in the ICU (5.2 days pre‐RRT versus 5.4 days post‐RRT, P = 0,49) and no significant impact on mortality of patients requiring ICU admission.

Conclusions:

Introduction of an RRT that proactively rounds on all patients discharged from an ICU did not decrease the rate of readmission to the ICU. the average ICU length of stay, and the mortality of patients requiring intensive care. Although RRTs have been widely implemented, their impact on patient outcomes continues to remain unclear.

Author Disclosure:

B. Butcher, none; J. Maselli, none; S. Ranji, none, A. Auerbach, none.