Effective management of behaviour disorders following brain injury is essential if individuals are to achieve their rehabilitation potential Best practice dictates that the intrusiveness of any operant approach used be minimal, remain in operation for the shortest time possible, and emphasize skill building. Ideally, treatment gains should maintain following its withdrawal. Reinforcement methods fulfil these criteria in that they are less intrusive, concerned with the establishment of pro-social behaviours, and encourage positive staff-patient interaction. While their efficacy has been well documented with other clinical populations, less is known regarding treatment of behaviour disorders in survivors of brain injury. Some existing studies are characterized by methodological weaknesses that limit understanding of any contribution made to observed improvement, and little is known regarding maintenance of treatment effects. In this paper the effectiveness of a variant of differential reinforcement, DRL, will be examined. Three cases will be presented which demonstrate increased behavioural control in response to the use of DRL. A strength of this paper is that the use of appropriate single-case design methodology, and follow-up data up to 18 months after treatment, permits more robust conclusions regarding the efficacy of DRL to be made. These are discussed, together with practical points regarding programme design.