149

Abstract

Introduction: The utility of in-house night intensivist coverage remains controversial with conflicting data on its ability to affect ICU metrics. We investigated the effect of a dedicated night critical care service on ICU length of stay (LOS) and mortality before and after the addition of an in-house night critical care attending at a tertiary referral, university hospital with a Level 1 Trauma Center. Methods: After an IRB exemption, we queried the hospital’s database to evaluate adult ICU LOS and mortality data before and after implementation. In-house attending night critical care coverage of at least 15 nights per month began in July 2011, and complete night coverage began in September 2011. Monthly data were counted as prior to the change (October 2009 until June 2011) and post-change (July 2011 to February 2013). Night critical care involves one attending acting as critical care backup on all 72 adult ICU beds. This attending is available and actively involved in resident critical care consultations, procedural supervision and resuscitative care management. No other new ICU initiatives were begun during this interval. Patient demographics, mortality index based on UHC 2012 risk adjustment, and ICU LOS were evaluated using t-tests and Wilcoxon’s rank sum respectively. Results: Total patients included in the study were 10,876, 50% prior to the June 2011 implementation date. Mean expected mortality using 2012 UHC Risk Model was 0.207 prior to June 2011 and 0.227 post implementation for our ICU populations. Mean (±SD) ICU LOS was 4.8 ± 9.3 days prior to July 2011, and 3.9 ± 4.5 thereafter. The overall LOS improvement was 0.8 days [95% CI 0.6–1.1]. LOS decrease per unit: Medical ICU 0.75 days (p<0.01); Neurosurgical ICU 0.22 days (p=0.33); Trauma- Surgical ICU 1.03 days (p<0.01). Overall our ICU UHC mortality index decreased 0.3 from 1.35 to 1.05 (95% CI:0.07–0.53; p<0.01). Conclusions: At our institution implementation of a single, actively involved, night intensivist significantly shortened ICU LOS and reduced mortality.