Multicenter implementation of a consensus-developed, evidence-based, spontaneous breathing trial protocol.

Objective: Evidence-based practice recommendations abound, but implementation is often unstructured and poorly audited. We assessed the ability of a peer network to implement an evidence-based best practice protocol and to measure patient outcomes. Design: Consensus definition of spontaneous breathing trial followed by implementation in eight academic medical centers. Setting: Six medical, two surgical, and two combined medical/ surgical adult intensive care units among eight academic medical centers. Study Population: Patients initiating mechanical ventilation through an endotracheal tube during a 12-wk interval formed the study population. Interventions: Adoption and implementation of a common spontaneous breathing trial protocol across multiple intensive care units. Measurements and Main Results: Seven hundred five patients had 3,486 safety screens for conducting a spontaneous breathing trial; 2072 (59%) patients failed the safety screen. Another 379 (11%) patients failed a 2-min tolerance screen and 1,122 (34%) patients had a full 30‐120 min spontaneous breathing trial performed. Seventy percent of eligible patients were enrolled. Only 55% of passing spontaneous breathing trials resulted in liberation from mechanical ventilatory support before another spontaneous breathing trial was performed. Conclusions: Peer networks can be effective in promoting and implementing evidence-based best practices. Implementation of a best practice (spontaneous breathing trial) may be necessary for, but by itself insufficient to achieve, consistent and timely liberation from ventilator support. (Crit Care Med 2008; 36:2753‐2762)