Objectives: Spontaneous awakening trials (SATs) improve outcomes in mechanically ventilated patients, but implementation remains erratic. We examined variation in reported practice, prevalence of attitudes and fears regarding spontaneous awakening trials, and organizational practices associated with routine implementation of spontaneous awakening trials in an ICU quality improvement collaborative. Design: Written survey. Setting: Michigan Health and Hospital Association's Keystone ICU, a quality improvement collaborative of 73 hospitals. Subjects: Attendees of the yearly Keystone ICU meeting, January 2011, including nurses, physicians, hospital administrators, and other healthcare professionals. Intervention: Respondents were asked about institutional characteristics, spontaneous awakening trial practice, attitudes and barriers regarding spontaneous awakening trials, and organizational cultural characteristics that might influence SAT practice. The association of organizational cultural characteristics and attitudes with reported spontaneous awakening trial use was evaluated using logistic regression. Measurements and Main Results: Three hundred nineteen participants attended the meeting. The survey response rate was 83.4%. Respondents reported wide variation in approach to spontaneous awakening trial performance and patient selection. 48.6% of respondents reported regular spontaneous awakening trial use, defined as greater than 75% of mechanically ventilated patients undergoing spontaneous awakening trials each day. In bivariable analysis, addressing sedation goals routinely in rounds and having spontaneous awakening trials as part of unit culture were positively associated with regular spontaneous awakening trial use, whereas the perception that spontaneous awakening trials increased short-term adverse effects, staff fears of spontaneous awakening trials, and the perception that spontaneous awakening trials are hard work were negatively associated with regular spontaneous awakening trial use. In multivariable analysis, only addressing sedation in rounds (odds ratio, 2.85 [95% CI, 1.55-5.23]), incorporation of spontaneous awakening trials into unit culture (odds ratio, 3.36 [95% CI, 1.75-6.43]), and the perception that spontaneous awakening trials are hard work (odds ratio, 0.53 [95% CI, 0.30-0.96]) remained statistically significantly associated with regular spontaneous awakening trial use. Respondents in managerial positions were less likely to perceive spontaneous awakening trials as hard work (odds ratio, 0.44 [95% CI, 0.22-0.85]). Conclusions: Even in a motivated statewide quality improvement collaborative, spontaneous awakening trial practice varies widely and concerns persist regarding spontaneous awakening trials. Cultural practices may counteract the effect of concerns regarding spontaneous awakening trials and are associated with increased performance of this beneficial intervention. Patient selection should be a focus for continuing medical education. Differences in perception of work between management and staff may also be a focus for improved communication.