Background: Psychotic disorders are a serious mental health problem. Intervention before the onset of psychosis might result in delaying the onset, reducing the impact or even preventing the first episode of psychosis. This study explores the effectiveness of cognitive behavioural therapy (CBT) in targeting cognitive biases that are involved in the formation of delusions in persons with an ultra-high risk for developing psychosis. A single blind randomised controlled trial compares CBT with treatment as usual in preventing or delaying the onset of psychosis. Method/design: All help seeking patients aged 14 to 35 years referred to the mental health services in three regions in the Netherlands are pre-screened with the Prodromal Questionnaire during a period of two years. Patients with a score of 18 or more on the sub-clinical positive symptoms items (45 items in total) will be assessed with the Comprehensive Assessment of At Risk Mental State (CAARMS). In a different pathway to care model all referrals from the mental health services in Amsterdam to the specialized psychosis clinic of the Academic Medical Centre in Amsterdam are also assessed with the CAARMS. The primary outcome is the transition rate to psychosis according to the CAARMS-criteria. Group differences will be analysed with chi-square tests and survival analyses. Discussion: CBT is a highly tolerated treatment. The psycho-educational CBT approach may prove to be a successful strategy since most people with an At Risk Mental State (ARMS) are distressed by odd disturbing experiences. Giving explanations for and normalising these experiences may reduce the arousal (distress) and therefore may prevent people from developing a catastrophic delusional explanation for their odd experiences and thus prevent them from developing psychosis. Screening the entire help-seeking population referred to community mental health services with a two-stage strategy, as compared with traditional referral to a specialist clinical psychosis centre, might detect more ultra-high-risk (UHR) patients. This type of screening could be implemented in mental health care as routine screening. The trial is registered at Current Controlled trials as trial number ISRCTN21353122.