BACKGROUND The present study describes the results of the pilot testing of a therapy we have developed for people with first-episode psychosis. Cognitively-oriented psychotherapy for early psychosis (COPE) is aimed at facilitating the adjustment of the person, and at preventing or alleviating secondary morbidity in the wake of the first psychotic episode. METHODS Eighty people formed three groups: those who were offered and accepted COPE (COPE subjects); those who refused COPE (refusal subjects); and those who were offered neither COPE nor any other continuing treatment from our service (control subjects). The individuals were assessed prior to, and at the end of, COPE treatment (a 12-month period) on the Integration/Sealing Over, Explanatory Model, Scale for the Assessment of Negative Symptoms, Brief Psychiatric Rating Scale, Quality of Life, SCL-90-R, and Beck Depression Inventory measures. RESULTS People who received COPE obtained significantly superior scores (P < 0.05) to the control group on four of the seven measures but only significantly out-performed the refusal group on one of the seven measures (P < 0.05). The COPE group performed significantly worse on the BDI than the refusal group (P < 0.05). Effect sizes are also provided for each measure. CONCLUSIONS There seems to be a place for psychological therapy in this group of people but our results need to be replicated in a more definitive randomised controlled trial and such a study is now in progress.