We prospectively investigated the large-scale implementation of a respiratory-therapist-driven protocol (TDP) that included 117 respiratory care practitioners (RCPs) managing 1,067 patients with respiratory failure over 9,048 patient days of mechanical ventilation. During a 12-mo period, we reintroduced a previously validated protocol that included a daily screen (DS) coupled with spontaneous breathing trials (SBTs) and physician prompt, as a TDP without daily input from a physician or "weaning team." With graded, staged educational interventions at 2-mo intervals, RCPs had a 97% completion rate and a 95% correct interpretation rate for the DS. The frequency with which patients who passed the DS underwent SBTs increased throughout the implementation process (p < 0.001). As the year progressed, RCPs more often considered SBTs once patients had passed a DS (p < 0.001), and physicians ordered more SBTs (46 versus 65%, p = 0.004). Overall, SBTs were ordered more often on the medicine than on the surgical services (81 versus 63%, p = 0.001), likely reflecting medical intensivists' prior use of this protocol. Important barriers to protocol compliance were identified through a questionnaire (89 respondents, 76%), and included: Physician unfamiliarity with the protocol, RCP inconsistency in seeking an order for an SBT from the physician, specific reasons cited by the physician for not advancing the patient to a SBT, and lack of stationary unit assignments by RCPs performing the protocol. We conclude that implementation of a validated weaning strategy is feasible as a TDP without daily supervision from a weaning physician or team. RCPs can appropriately perform and interpret DS data more than 95% of the time, but significant barriers to SBTs exist. Through a staged implementation process, using periodic reinforcement of all participants in ventilator management, improved compliance with this large-scale weaning protocol can be achieved.