Lack of Insight in Late-Life Schizophrenia: A Function of Illness Severity and Premorbid Intellectual Functioning


      Lack of insight in schizophrenia is a common, but poorly understood phenomenon. It is associated with illness severity, medication non-adherence and poor treatment outcomes (1, 2). Several studies in the adult schizophrenia population have shown that lack of insight is associated with executive dysfunction (assessed using the Wisconsin Card Sorting Test and Trails Making Test - B), global cognitive functioning (i.e. IQ), and to a lesser degree attention (as assessed by the Continuous Performance Task) (3, 4). Aging has been associated with a decline in these cognitive functions. Thus, the relationship between cognition and insight in late-life schizophrenia could differ from that which has been described in adult life. To date, few studies have explored the phenomena of lack of insight in schizophrenia in later life (defined here as age ≥60 years). In one study of middle aged and older adult persons with schizophrenia (≥40 years, mean ∼50 years) poorer insight was associated with negative symptom severity and poorer quality of life (5). In another study, investigating the relationship between insight and treatment outcomes with Cognitive Behavioral and Social Skills Training (CBSST) in older persons with Schizophrenia, higher insight was associated with poorer everyday functioning and hopelessness, but not depression (6). In neither of these was cognition assessed. As such, the aim of this study was to assess this relationship using measures of cognition previously shown to be associated with poor insight in younger adult populations. Based on the adult literature, we hypothesized that lack of insight in late-life schizophrenia would similarly be associated with greater illness severity, and poorer premorbid IQ and executive functioning.


      This study analyzed data collected from baseline assessments of individuals with schizophrenia, age ≥60 years, who had consented to particpate in a CBSST trial for older persons with schizophrenia ( Identifier:NCT00832845). Insight was assessed using a clinican-rated measure - the Positive and Negative Syndrome Scale (PANSS) Insight and Judgment item (G12 item), and a self-report measure - the Birchwood Insight Scale (BIS). Cognition was assessed with the Wechsler Test of Adult Reading (WTAR); Trail Making Test-B (TMT-B); Letter fluency (FAS); Continuous Performance Test (CPT); Letter Number Span (LNS); Hopkins Verbal Fluency Test-Revised, perseveration and delayed free recall scores; and the Brief Visuospatial memory Test-Revised (BVMT-R). Participants' cognitive insight was also assessed, using the Beck Cognitive Insight Scale (BCIS). Pearson correlations were used to evaluate the bivariate associations between insight, cognition and cognitive insight. Subsequently, a hierarchal regression analyis was performed to assess the degree to which these cognitive functions explained the phenomenon of lack of insight.


      Data from 59 participants were analyzed: Mean Age=65.2, SD=5.5, Range=60-79; Male:Female, 25:24; Education=11.6 years, SD=2.8; PANSS Total without item G12=50.9, SD=12.7; PANSS G12=2.7, SD=1.7; BIS=8.4, SD=2.4. PANSS G12 item insight scores correlated with premorbid IQ (r=-0.554, p<0.001), TMT-B time (r=0.451, p=0.001), LNS raw score (r=-0.408, p=0.003), BACS raw score (r=-0.421, p=0.003), and CPT (r=-0.456, p=0.003). In contrast, BIS scores did not correlate significantly with either the PANSS G12 item or other cognitive measures. There was also no significant relationship between either PANSS G12 item or BIS and BCIS scores. The regression analysis revealed that illness severity (PANSS total score excluding item G12) explained 20.1% (p=0.003) and premorbid intellectual functioning (WTAR) explained 23.4% (p<0.001) of the variance of PANSS G12 item. No variables explained a significant amount of the variance of BIS scores.


      Lack of insight in late-late schizophrenia appears to be, in part, a function of illness severity and premorbid intellectual functioning; and unexplained by conventional measures of cognition and cognitive insight. These findings suggest that lack of insight is to a large part independent of these other measures in late-life schizophrenia and that any relationship that exists is accounted for by premorbid intellectual functioning. Alternatively, age-related changes in cognition may attenuate any association between cognition and insight in late-life schizophrenia while preserving the relationship with premorbid cognitive function. As in the younger adult literature, lack of clinical insight was not associated with the cognitive insight (7, 8).