Relationship of neurocognitive function and impairment of insight in first episode schizophrenia

Abstract

BACKGROUND: Impairment of insight is one the common features of schizophrenia with clinical importance because of its relationship with treatment compliance and hence clinical outcomes. However, its etiology remains poorly understood. Various studies have suggested the link of poor insight with the neurocognitive functions particularly executive function and memory. The present study aim at evaluating the association between insight and measures of executive functions and working memory in patients with first episode schizophrenia and explore the contribution of these neurocognitive functions to the variance of insight. METHODS: Consecutive patients between ages 18 and 55 with first episode schizophrenia based on DSM-IV diagnostic criteria were recruited from adult psychiatric services covering a defined catchment area in Hong Kong. All the assessments were conducted within six months of starting treatment. Clinical symptoms were evaluated using Positive and Negative Syndrome Scale (PANSS). The Scale for Assessment of Unawareness of Mental Disorder (SUMD) was used to assess insight. Executive functioning was measured by the Wisconsin Card Sort Test (WCST) and Verbal fluency. Working memory was measured by the Letter Number Sequencing (LNS) test from the Wechsler Adult Intelligence Scale (WAIS). RESULTS: There were 79 patients completed the study. The average score of the first three items of SUMD were used as overall insight rating. Individual items were analyzed separately. After correction for multiple testing, WCST categories completed was found to be significantly correlated with overall insight rating (r=-0.337, p=0.003). After analyzing the sub-scale of SUMD in detail, both WCST categories (r=0.315, p=0.006) and WCST perseverative error (r=-0.291, p=0.011) were significantly correlated with awareness of consequence of mental illness (SUMD item 2) but not the awareness of having mental illness (SUMD item 1) and the awareness of need for medication (SUMD item 3). No significant correlations were found with other neurocognitive functions. While the symptomatology explained 27% of variance in overall insight rating, WCST categories explained 7.4% of the variance. The symptomatology explained 26.8% of variance in the awareness of consequence of mental illness, WCST categories completed and perseverative errors explained 5.7% of the variance. DISCUSSION: The current study shows a significant association of WCST performance, particularly categories completed and perseverative errors, with poor insight in patients with first episode schizophrenia. No other neurocognitive functions were found to be significantly associated. This is in line with previous findings in both chronic and first episode patients. Though the contribution of these neurocognitive functions to the variance of insight is small, the consistence of the finding indicating that executive function could be one of the important elements of underlying neurocognitive deficit explaining poor insight. Further studies are needed to explore the etiological models for poor insight.