Differential associations of cognitive insight components with pretreatment characteristics in first-episode psychosis

Abstract

An increasing number of studies have focused on cognitive insight (i.e. awareness of one's own thinking) in psychotic disorders. However, little is known about the premorbid and pretreatment correlates of cognitive insight in the early course of psychosis. One hundred and three patients experiencing first-episode psychosis (FEP) were assessed shortly after treatment initiation for cognitive insight. Pretreatment and baseline clinical, functional and neurocognitive characteristics were examined. The self-reflectiveness dimension of cognitive insight was independently associated with clinical insight and executive functioning, whereas self-certainty was associated with premorbid IQ, premorbid academic adjustment and clinical insight. The amount of variance explained by the independent variables was small to moderate. Self-reflectiveness and self-certainty have differential pretreatment correlates in FEP and may reflect separate cognitive processes which require targeted interventions.

Introduction

Since the first descriptions of schizophrenia an extensive body of research and theorizing has been devoted to impaired insight. Although schizophrenia-related disorders are characterized by marked heterogeneity, lack of insight is common in psychosis, with virtually all patients having poor insight at some point in the course of the illness (Carpenter et al., 1973). The clinical interest in this construct stems from its relationship with poorer adherence to medication (Buchy et al., 2010, Lincoln et al., 2007) and psychological treatment (Alvarez-Jimenez et al., 2009) which often result in more relapses, increased symptoms, social dysfunction and rehospitalisation (Mintz et al., 2003, Lincoln et al., 2007).

Beck et al. (2004) have developed a scale to assess a different aspect of insight that they called cognitive insight (not to be confused with the awareness of neurocognitive problems). The Beck Cognitive Insight Scale (BCIS) is self-report measure with two dimensions: introspection and willingness to acknowledge fallibility (called self-reflectiveness), and confidence in their beliefs and judgments (called self-certainty). Thus, while previous assessments of insight focus on the ability to identify the illness and its consequences (clinical insight), the concept of cognitive insight involves the metacognitive ability to examine and question one's beliefs and misinterpretations beyond illness matters. Although cognitive insight is moderately correlated with measures of clinical insight, it is thought to be a basic ability necessary for the development of adequate clinical insight (Riggs et al., 2012).

Beck et al. (2004) postulated that cognitive insight would be mainly associated with delusional thinking. With some inconsistencies, subsequent research has showed significant correlations between low cognitive insight and delusions as predicted by the theoretical model of the BCIS (Pedrelli et al., 2004, Warman et al., 2007, Engh et al., 2010). Similarly, overconfidence in one's own beliefs, a component of the BCIS, has been found to be related to delusion proneness in the normal population (Warman and Martin, 2006).

Importantly, gains in cognitive insight derived from the cognitive-behavioral therapy (CBT) have been associated with improvement in psychotic symptoms at the end of treatment (Granholm et al., 2005, Perivoliotis et al., 2010), suggesting that cognitive insight may be a potential mechanism of symptom change in CBT (Granholm et al., 2005). Moreover, a number of studies in patients with schizophrenia and schizoaffective disorders have shown an association between the metacognitive abilities underpinning cognitive insight, overall work performance (Lysaker et al., 2011) and living independently (Favrod et al., 2008).

Yet little is known about the associations of cognitive insight with pretreatment variables such as duration of untreated psychosis, premorbid functioning and cognitive functioning in FEP, before the potential confounding effects of long term treatments and chronicity are apparent. This has important implications for understanding the nature and treatment of cognitive insight.

The primary aim of the present study was to examine the relationship of cognitive insight to pretreatment variables in a sample of first-episode psychosis (FEP).