OBJECTIVES: To evaluate the use of sedatives in critically ill adult patients with endotracheal tube (ETT) and mechanical ventilator (MV) in a hospital in Taiwan. METHODS: We conducted the study retrospectively by reviewing medical records. All patients with ICD-9, 96.71, 96.72, 96.04, from Oct 2008 to Feb 2009 were included. Patients were excluded if they were pediatric patients, without complete records, using MV after discharged, in regular wards, or using MV less than 24 hours. The primary end point is the duration of mechanical ventilation. The secondary end points were length of ICU stay, hospital stay and events of tracheostomy, reintubation, self-extubation. Chi square and t-test were performed for dichotomous and continuous variables respectively. SPSS (Version 13.0) was used. The data were log transformed to address the non-normal distribution. RESULTS: We collected 50 events of insertion of endotracheal tube with mechanical ventilator (ETT+MV) in using sedatives group and 163 in not using group. The duration of mechanical ventilation was not significantly difference between two groups. (p= 0.582) The using sedatives group had longer ICU (p= 0.04) and hospital length of stay (p= 0.001). There were much more patients were physically restricted in the using sedatives group (84% vs. 56%). The usages of sedatives were 50% of midazolam and 54% of propofol. The mean treatment duration among patients receiving midazolam was 4.6 days and 2.2 days among patients with propofol. This longer duration of length of stays were possibly related to the usage of midazolam and the prescribing habits of physicians. CONCLUSIONS: We found the physician preference by using physical restriction as first line and sedatives as second line did not benefit the critically ill patients. We will implement a sedation guideline to tailor the needs of our facility for optimal the patient care.