Clients in a community-based methadone treatment program earned stars for attending counseling sessions as scheduled and for providing clean urines. These stars were later redeemed for contingent rewards (food or gas coupons or bus tokens) according to one of three randomly assigned reward schedules, including high reward (four stars per prize), low reward (eight stars per prize), or delayed reward (those who had to wait 3 months to earn a prize). Clients in the high-reward condition showed a pattern of increasing the number of stars earned for group sessions and clean urines across the 3-month intervention. All clients, independent of reward condition, attended significantly more group counseling sessions during the months that contingent reinforcers were available than in the months prior to, and after, the intervention. Finally, urinalysis data indicated that, in the postintervention period, high-reward clients had fewer dirty urines than did low-reward or delayed-reward clients. This study suggests that a simple system of recognizing client progress with stars and modest prizes for performing specific behaviors can be an effective tool in increasing clinic attendance rates and reducing positive urines.