The Effect of the ABCDEF Bundle on Incidence of Delirium in Critically Ill Patients

摘要

Objectives: The purpose of this project was to study the effect of multiple components of the ABCDEF bundle on the incidence of delirium, number of mechanical ventilator days, and intensive care unit length of stay on critically ill patients. The ABCDEF bundle includes assessing and management of pain (A), both spontaneous awakening and breathing trials (B), choice of sedation (C), delirium monitoring and management (D), early mobility (E), and family engagement (F).;Design: A quasi-experimental, pre and post-bundle, quality improvement project in the Intensive Care Unit (ICU).;Setting: A 20-bed intensive care unit in southern California.;Methods: The healthcare team was educated on their role in implementing the appropriate components of the ABCDEF bundle into everyday ICU care. After staff education on the ABCDEF bundle the post bundle group was evaluated. The main outcome variables compared between the two groups included delirium daytime and nighttime events, number of mechanical ventilator days, and ICU stay days. Demographic and clinic outcomes data were also compared between pre and post bundle groups.;Results: Patients in the post-bundle group had a lower incidence of delirium in both the day and night than those in the pre-bundle group (Day 11%, n = 16, 51%, n = 77; p <.001) (Night-18%, n = 27, 59%, n = 89; p <.001). Patients in the post-bundle group also had lower median ICU length of stay days compared to the pre-bundle group (Mdn = 3.0 days, IQR = 2-5, Mdn = 4.0 days, IQR = 3-8; p <.001). There was no statistically significant difference in patient's age, gender, primary diagnosis, use of mechanical ventilation, and ventilator days.;Conclusion: The ABCDEF bundle was successfully implemented into everyday ICU care. Post-bundle patients had a statistically significant reduction in the incidence of delirium in the day and night as well as ICU length of stay. Average number of mechanical ventilation was reduced but not statistically significant.