Propofol is under evaluation as a sedative for endoscopic procedures. Eighty outpatients (ASA Class I or II) undergoing colonoscopy were randomized to receive either propofol or midazolam plus meperidine, administered by a nurse and supervised by an endoscopist. Endpoints were patient satisfaction, procedure and recovery times, neuropsychological function, and complications. The mean dose of propofol administered was 218 mg; mean doses of midazolam and meperidine were, respectively, 4.7 mg and 89.7 mg. Mean time to sedation was faster in the propofol patients (2.1 min vs. 7.0 min; p < 0.0001), and depth of sedation was greater (p < 0.0001). On average, after the procedure, the propofol patients could stand at the bedside sooner (14.2 vs. 30.2 min), reached full recovery faster (14.4 vs. 33.0 min), and were discharged sooner (40.5 vs. 71.1 min) (all p < 0.0001). Patients who received propofol also expressed greater overall mean satisfaction on a 10-point visual analog scale (9.3 vs. 8.6; p < 0.05). At discharge, the propofol group had better scores on tests reflective of learning, memory, working memory span, and mental speed. Four patients in the midazolam/meperidine group developed minor complications (1 hypotension and bradycardia, 2 hypotension alone, and 1 tachycardia) and 1 patient in the propofol group had oxygen desaturation develop during an episode of epistaxis. For outpatient colonoscopy, propofol administered by nurses and supervised by endoscopists has several advantages over midazolam plus meperidine and deserves additional investigation.