Exploring metacognition: a systematic review of measures of metacognition and a systematic review and meta-analysis of functioning and symptom correlates of metacognition across adult mental health

Abstract

Background: The last decade has seen significant growth in research that explores metacognition and its facets within the field of psychiatric disorders. A substantive body of research supports a link between reduced metacognitive capacity and psychological distress across the diagnostic spectrum. As our understanding of the role of metacognition in the maintenance of psychological distress grows, the assessment of metacognitive capacity is increasingly important within research and clinical settings. Emerging psychotherapeutic approaches theorise increased metacognitive capacity to be central to the mechanism of change. This creates additional need for reliable, valid and robust measurement tools to establish metacognitive capacity and track change. Despite this need we see variance in the conceptualisation of metacogntion which is mimicked in the instruments of measurement, covering a spectrum view from discreetly defined skills to integrated processes and sub-domains. Current research explores the associations between these conceptualisations of metacognition and functioning and symptom correlates. These associations were initially reported to cluster around populations with experience of psychosis but have latterly expanded to those with non-psychotic disorders. This suggests that metacognitive capacity can be viewed as an underlying mechanism of psychological distress. Objectives: A systematic review of measures of metacognition synthesised data on published measures of metacognition applicable for use in an adult mental health population. The methodological quality of each included study and the psychometric properties reported were critically appraised with comment on measure accessibility and generalisability. A systematic review and meta-analysis synthesised extensive data from studies on metacognition and its correlates. Using sub-group analyses to explore the associations between functioning, symptom and demographic variables and metacognitive capacity across the diagnostic spectrum drawing samples from clinical and non-clinical populations. Methods: The systematic review protocols for both studies were registered on the International Prospective Register of Systematic Reviews (PROSPERO) prior to formal screening. Eligibility criteria was establish a priori, and both systematic database searches were performed in accordance with the Preferred Reported Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. For the first systematic review the methodological quality of each included study was assessed using the COnsensus-based Standards for the Selection of health Measurement Instruments (COSMIN) checklist. For the second systematic review and meta-analysis Pearson’s correlation was selected as the common effect size metric and data were analysed using the Metafor package in R. A random effects model was applied with separate meta-analyses for each correlate and sub-group analysis by clinical population. Results: Database searching produced 254 units of identified data, 11 papers met criteria for inclusion and these reported on seven different measures of metacogntion. The psychometric properties of the measures reviewed showed inconsistency in the properties reported and acceptability judgements. Concept inconsistency was also evident. The Metacognitive Self-Assessment Scale (MSAS; Pedone et al., 2017) was rated to have the highest methodological quality. Database searching for the second study produced 626 units of identified data, 30 papers met criteria for inclusion. The pooled sample across all included studies represented 2,360 participants. Meta-analytic findings showed general functioning to be significantly negatively associated with metacognition transdiagnostically. Sub-group analysis revealed this effect to be strongest amongst samples with experience of psychosis and weakest in non-clinical samples. Social functioning also showed a moderate effect size, transdiagnostically. Sub-group analysis revealed this effect to be strongest amongst clinical samples without experience of psychosis. All symptom variables were shown to be significantly negatively associated with metacognition. Conclusions: The systematic review on measures of metacognition highlights the need for more consistent and robust evaluation of the psychometric properties of metacognitive measures. This synthesis provides those selecting a measure of metacognition with comparative data on methodological quality and psychometric properties as well as guidance on accessibility and generalisability. We highlight the importance of considering the purpose of the measure (discriminative, predictive, evaluative) and the characteristics of the instrument (number of items, accessibility, interpretation options) specific to the scenario of use, in addition to psychometric properties, when selecting a measure of metacognition. Study limitations were presented. The systematic review and meta-analysis of functioning and symptom correlates brings new depth and clarity to our understanding of metacognitive capacity and its associations across different adult mental health populations. This work supports previous literature that suggests metacognition should be considered transdiagnostically, and therefore designates it as underlying symptom phenomenology. Study limitations were presented.