S45. EFFICACY OF A 4-SESSION METACOGNITIVE TRAINING FOR SCHIZOPHRENIA, DEPRESSION, AND BELIEF FLEXIBILITY

BACKGROUND: Metacognitive training (MCT) was developed to increase awareness of cognitive biases (Moritz & Woodward, 2007). Each of the 8 MCT modules targets a specific reasoning bias, such as bias against disconfirmatory evidence, jumping to conclusions, and attributional biases. MCT has been shown to be effective in improving delusions. However, it remains unclear to what extent specific MCT modules are effective in ameliorating the reasoning biases that they target, and whether they may also be effective for other disorders characterized by similar cognitive biases. This study aimed to compare the efficacy of a 4-week MCT on belief flexibility among patients with schizophrenia and patients with major depressive disorder (MDD). METHODS: This study adopted a single-blind randomized controlled design. Adult patients with a schizophrenia spectrum disorder (N = 56) and MDD (N = 57) were respectively randomized into MCT or treatment as usual (TAU, i.e. standard psychiatric care). The MCT intervention consisted of the following modules: ‘attributions’, ‘changing beliefs’, ‘to empathise’, and ‘self-esteem and mood’. Patients were assessed at pre-treatment, post-treatment, 1-month and 6-month follow-ups. Belief flexibility was measured using the Maudsley Assessment of Delusions Scale (MADS) and the Bias Against Disconfirmatory Evidence (BADE) task (Wessely et al, 1993; Woodward et al, 2006). RESULTS: Among the 113 participants, 27 patients with schizophrenia and 29 patients with MDD attended the 4-week MCT. For the schizophrenia arm, repeated-measures ANOVA revealed significant improvements in PANSS total score (p < .001, d = 0.87) and PSYRATS delusions score (p = .001, d = 0.69) after MCT. These treatment effects sustained at 1-month follow-up (ps < .01), and improvement in delusions sustained at 6 months (ps < .05). Mixed-design ANOVAs revealed that improvements in PANSS total score (ps < .05) and PSYRATS delusions score (ps < .01) on the MCT condition were significantly greater than TAU over the corresponding timeframes. McNemar tests revealed that one of the MADS measures, reaction to hypothetical contradiction (RTHC), improved after MCT (p = .004), and sustained at 1 month (p = 0.016) and 6 months (p = 0.002). There was no change in belief flexibility across timepoints on the TAU condition. Change in RTHC following MCT was not significantly greater than TAU. Symptom changes were not predicted by any of the belief flexibility variables. For the MDD arm, repeated-measures ANOVA revealed significant improvement in Beck Depression Inventory (BDI-II) (p < .001, d = 1.45) after MCT, which sustained at 1-month and 6-month follow-up (ps < .01). Mixed-design ANOVAs revealed that improvements in BDI-II on the MCT condition were significantly greater than TAU over the corresponding timeframes (ps < .001). Repeated-measures ANOVA revealed that evidence integration (EI) improved after MCT (d = 0.57) and sustained at 1 month (p = 0.042) and 6 months (p = 0.041). There was no change in belief flexibility across timepoints on the TAU condition. Change in EI following MCT was not significantly greater than TAU. Symptom change was not predicted by any of the belief flexibility variables. DISCUSSION: We found large and persistent effects of reduction in delusions and depression, following a 4-week MCT, in two clinical groups. There is preliminary evidence that belief flexibility improved following MCT, although the effects were less stable and were of smaller sizes. Further research on the relationship between belief flexibility and various psychopathologies is warranted.