It is assumed that dysfunctional meta-cognitive beliefs about one's thoughts increase problematic appraisals and coping behaviors, which further contribute to the development of mental disorders (Wells and Matthews, 1994; Wells, 2000). Although this research interest originated around generalized anxiety disorder (GAD), recent studies have begun to examine similar meta-cognitive processes in other disorders. The majority of studies using Meta-cognitions Questionnaire (MCQ; Cartwright-Hatton & Wells, 1997) and its variants to assess meta-cognitive beliefs.

We conducted a meta-analysis to integrate empirical findings on group differences in meta-cognitive beliefs between healthy individuals and patients with various psychiatric disorders.

We followed the PRISMA guideline (Liberati et al., 2009). A systematic literature search was conducted. We included studies that involved a diagnosed psychiatric group and healthy controls (aged 18 or above), reported group comparisons of metacognition, and were published during the period of 1990–27 August 2015. Effect sizes were computed.

A final set of 43 studies was included. Large combined effect sizes were found on each subdomain of the MCQ, indicating increased levels of dysfunctional meta-cognitive beliefs in patients. Subgroup analyses were carried out based on psychiatric diagnosis (i.e. psychosis, n=10; GAD, n=7; obsessive-compulsive disorder, OCD, n=15; anorexia nervosa, n=5). All patient groups were more dysfunctional on each subtype of meta-cognitive beliefs than controls. Effect size of U/D was particularly large for GAD, and that of CSC was particularly large for OCD.

Dysfunctional meta-cognitive beliefs are evident across several psychiatric disorders, with specific types of beliefs being more marked in certain diagnoses.