In an effort to examine the placebo, psychological deterrent, and pharmacological deterrent effects associated with implanted disulfiram, subjects were given either disulfiram implants or sham operations. Ethanol challenges elicited no disulfiram-ethanol reactions (DERs), indicating that at the time of the challenge neither a pharmacological deterrent nor a placebo effect was operating. Of the patients who resumed drinking, only those with disulfiram implants experienced DERs. Sham operation subjects continued to drink after their first post-challenge drink; four of five disulfiram implant recidivists remained abstinent following their experience of a DER. It is concluded that the pharmacological deterrent effect of the disulfiram implant may have been underestimated in previous reports.