Predictors for daily interruption of sedation therapy by nurses: A prospective, multicenter study

Abstract

Purpose The aim of the study was to identify the nurse and patient-related factors predicting daily interruption of sedation (DIS) performance by nurses in the intensive care unit (ICU). Methods Nurses, caring for a mechanically ventilated patient receiving 24 hours or more of a continuously infused sedative, were interviewed at the bedside to determine their willingness to perform DIS on this patient and to determine the influence of 20 nurse- and 47 patient-related factors on DIS completion. Results The 57 (44%) of 130 of nurses willing to perform DIS had performed DIS at least once in the past ( P < .0001) and were not targeting deep sedation (ie, Sedation Agitation Scale [SAS] ≤ 2 [ P = .03]). The DIS performance was less likely with use of higher-dose continuous midazolam ( P = .006), a fraction of inspired oxygen (F io 2) greater than 50% ( P = .03), or positive end-expiratory pressure greater than 5 mm Hg ( P = .006) and in patients either deeply sedated (SAS ≤ 2) ( P = .05) or agitated (SAS ≥ 5) in the past 24 hours ( P = .003). Prior DIS experience (odds ratio [OR], 2.54; P = .004), hours of sedation-related continuing education (OR, 1.13; P = .02), and a target of deep sedation (OR, 0.49; P = .02) were independent nurse-related factors for DIS performance. Nurse's willingness to conduct DIS ranged from 45% to 80% based on the interaction between patient sex, current F io 2, and agitation in past 24 hours. Conclusions Educational strategies and institutional protocols focused on improving use of DIS need to consider the various nurse- and patient-related factors that affect DIS performance by nurses in the ICU.