To assess the efficacy of Metacognitive Training (MCT) in symptoms and metacognitive variables in people with a recent onset of psychosis.

A multicenter, randomized and controlled clinical trial was performed. One hundred and twenty-six patients were randomized to MCT or a psycho-educational intervention. Patients with a recent onset of psychosis were recruited from 9 centers of Spain. The treatment consisted in 8 weekly sessions in both groups. Patients were assessed at baseline, post-treatment, and 6 months of follow-up. Symptoms were assessed by the PANSS. Metacognition was assessed by a battery of questionnaires of cognitive biases and social cognition: BCIS, IPSAQ, TCI, Hinting task and Emotional Recognition Test.

PANSS positive symptoms significant declined between baseline and post-treatment in psycho-educational (P=0.04) and MCT group (P=0.01), while general PANSS and total PANSS were significant between baseline and post-treatment in the MCT group only (P=0.008; P=0.005). Across time, the MCT group was superior to psycho-educational on the BCIS total and self-certainty subscale (P=0.042). Regarding irrational beliefs, the intolerance to frustration subscale declined more strongly in the MCT in relation to psycho-educational group (P=0.016). ToM, Personalizing Bias and JTC improved more strongly in the MCT group compared to psycho-educational group (P<0.001–0.032). Most results remained significant at the follow-up.

MCT could be an effective psychological intervention for people with a recent onset of psychosis in order to improve symptoms, insight, tolerance to frustration and personalizing bias.