Development, implementation, and evaluation of an institutional daily awakening and spontaneous breathing trial protocol: a quality improvement project.

Introduction: While one controlled trial found that a daily awakening and spontaneous breathing trial (DA-SBT) decreases time on mechanical ventilation (MV), there is a paucity of real-world data surrounding the development, implementation, and impact of DA-SBT protocols. We describe a multidisciplinary process improvement effort in 2, 10-bed medical intensive care units (MICUs) at a 330-bed academic medical center that focused on the development, implementation, and evaluation of a new DA-SBT protocol.

Methods: A DA-SBT protocol, developed using results from a nursing survey literature and available institutional resources, was implemented after extensive clinician education and institution of quality reminders to boost use. Postprotocol compliance was evaluated. Use of sedation, DA and SBT practices, and clinical outcomes were retrospectively compared between the before and after DA-SBT protocol groups (ie, consecutive MICU patients requiring a continuously infused sedative [CIS] ≥24 hours).

Results: In the after group (n = 32), the DA and SBT compliances were 44% and 84%, respectively. Compared with the before group (n = 33), after group patients received CIS on fewer days of MV (100% vs 67%, P = .003) and had their CIS down-titrated by ≥25% on more days of CIS (40% vs 71%, P = .006). Neither total CIS dose (P = .49), total MV days (P = .75), days of MV where a SBT occurred (P = .38), nor episodes of self-extubation (15% vs 6%, P = .43) differed between the 2 groups.

Conclusion: Despite the implementation of a DA-SBT protocol that was individualized to clinician preferences and institutional resources and accompanied by substantial education and reminders for use, compliance to the DA component of this protocol was low and duration of MV remained unchanged. Additional quality improvement strategies are needed to overcome barriers to DA-SBT protocol use that may not exist in controlled clinical trials.