A clinical recovery score (CRS) assessing recovery after general anesthesia was compared with the Digit-Symbol Substitution Test (DSST), Trieger Test (TT), a patient-completed visual analogue scale for alertness (VAS), and an independent observer's evaluation of recovery. The CRS included ratings of the following parameters: activity, respiration, circulation, consciousness, ambulation, color, and nausea and vomiting. Forty patients requiring the removal of three or four third molars participated in the study. All patients received the same general anesthetic technique. Each patient was evaluated by the five methods preoperatively, on admission to the recovery room, and at 15-min intervals until discharge. The four recovery tests (CRS, DSST, TT, VAS) were evaluated using chi 2 analysis to determine if there was any overall difference among the tests using the observer's determination of home readiness as the standard for discharge. The CRS was significantly more in agreement with the observer's determination than were the paper and pencil tests. The recovery tests were also evaluated with regard to instances of early dismissal or prolonged retention of the patient, again using the observer's determination as the "gold standard." The CRS was the only recovery test devoid of early dismissals. We conclude that the CRS provides a valid, simple measure of recovery that can be readily used in offices providing outpatient anesthesia and in studies measuring clinical recovery from anesthesia or sedation.