INTRODUCTION: Several methods, including loss of eye lash (eyelash) reflex and loss of verbal contact, have heen used as criteria for estimating the optimal hypnosis for anesthesia induction. However, these methods are too objective. We examined the hypnotic dose response of etomidate, using a bispectral index as a more subjective criterion for anesthesia induction. METHODS: Fourty-Five ASA I or II adult patients scheduled for elective orthopedic surgery were randomly allocated to three groups according to induction dose of etomidate. They were Group 1: 0.1 mg/kg ; Group 2: 0.15 mg/kg; Group 3: 0.2 mg/kg (n = 15 for each group), respectively. Etomidate diluted as 10 ml in a syringe injected through an 18G forearm intravenous catheter, using a syringe pump at the rate of 20 ml/min. Observer's Assessment of Alertness/sedation (OAA/S) scale of 1 was considered optimal for hypnosis. Bispectral index, OAA/S scale, and vital signs were checked every minute until spontaneous eye opening after end of drug infusion. Also, the correlation coefficient between BIS and OAA/S scale was checked for evaluating the bispectral index; this was a useful tool for estimating the degree of hypnosis. RESULTS: BIS and OAA/S showed their lowest scores around 60 sec. after the etomidate injection, which was very different from time to peak effect known to be the 2 min. The correlation coefficient (r) between BIS and OAA/S was 0.84 on average, suggesting the BIS as an good subjective indicator of optimal hypnosis for anesthesia induction. Vital signs were stable in all groups. Hypnotic ED50 and ED95 were 0.12 and 0.19 mg/kg, respectively. CONCLUSIONS: Bispectral index can be a useful tool for estimating the optimal hypnosis for anesthesia induction. Hypnotic ED50 of etomidate was 0.12 mg/kg.