This paper describes the rationale for a randomized controlled trial, comparing cognitive behavior therapy in addition to treatment as usual with treatment as usual alone, for borderline personality disorder. Previous pioneering randomized controlled trials of psychotherapies have suffered from methodological weaknesses and have not always been reported clearly to allow adequate evaluation of either the individual study or comparisons across studies to be undertaken. We report on the recruitment and randomization, design, and conduct of an ongoing randomized controlled trial of one hundred and six patients with borderline personality disorder. Primary and secondary hypotheses and their planned analyses are stated. The baseline characteristics of 106 patients meeting diagnostic criteria for borderline personality disorder are described.