BACKGROUND: Endotracheal tube (ETT) occlusion is a potentially life-threatening event. This study describes a quality improvement project to prevent ETT occlusion in critically ill patients. METHODS: After a cluster of clinically significant ETT occlusion incidents at a tertiary-care intensive care unit (ICU), the root cause analysis suggested that the universal use of heat moisture exchangers (HMEs) was a major cause. Then, we prospectively audited new ETT occlusion incidents after changing our practices to evidence-based active and passive humidification during mechanical ventilation (MV). We also compared the outcomes of affected patients with matched controls. RESULTS: During 100 weeks, 18 incidents of clinically significant ETT occlusion occurred on a median of 7 days after intubation (interquartile range, 4.8–9.5): 8 in the 10 weeks before and 10 in the 90 weeks after changing humidification practices (8.1 vs. 1.0 incidents per 1000 ventilator days, respectively). The incidents were not suspected in 94.4%, the peak airway pressure was >30 cm H 2 O in only 25%, and 55.6% were being treated for pneumonia when ETT occlusion occurred. Compared with 51 matched controls, ETT occlusion cases had significantly longer MV duration (median of 13.5 vs. 4.0 days; P = 0.002) and ICU stay (median of 26.5 vs. 11.0 days; P = 0.006) and more tracheostomy (55.6% vs. 9.8%; P < 0.001). The hospital mortality was similar in cases and controls. CONCLUSIONS: The rate of ETT occlusion decreased after changing humidification practices from universal HME use to evidence-based active and passive humidification. ETT occlusion was associated with more tracheostomy and a longer duration of MV and ICU stay.