STUDY OBJECTIVE: To determine whether a set of predetermined discharge criteria, used by nurses, shortens the length of patient stay in the postanesthesia care unit (PACU) without compromising safety. STUDY DESIGN: Prospective clinical study. SETTING: Postoperative recovery area of a large, tertiary-care, academic hospital. PATIENTS: 1,198 adult, ASA physical status I, II, and III patients, 18 years or older, requiring general anesthesia. INTERVENTIONS: Two groups of inpatients were followed. In the first group, over a 90 day period, all inpatients meeting study inclusion criteria were discharged by a physician (traditional discharge group [TDG]). This period was followed by a subsequent 90 day period in which a second group of patients were discharged by a PACU nurse using predetermined discharge criteria (discharge criteria group [DCG]). MEASUREMENTS: Demographic and discharge variables, including the time when the physician was called for an order, the time a discharge order was written, and the actual discharge time, were recorded. Other comparisons of various discharge variables included the time that discharge criteria were met, the actual discharge time, and the difference between these times. The frequency of PACU stays longer than 60 minutes, PACU discharge delays, and the time discharge criteria were met longer than 60 minutes were also compared between groups. Floor nurse satisfaction with patient status and any related morbidities noted were also measured. MAIN RESULTS: Length of PACU stay was significantly shorter (133.1 +/- 91.4 vs 101.7 +/- 53.7 min; P < 0.05) for inpatients in the DCG group. Discharge delays were higher with physician discharge. The DCG had a lower number of patients with multiple delays. There was no significant difference in the occurrence of adverse events between the two groups. CONCLUSIONS: Predetermined discharge criteria resulted in a 24% decrease in PACU time.