The Measure of Insight into Cognition: reliability and validity of clinician-rated and self-report scales of neurocognitive insight for schizophrenia

Abstract

Poor insight into the neurocognitive deficits in schizophrenia-spectrum disorders may have significant clinical implications for treatment. The purpose of this study was to replicate and extend the psychometric findings initially reported on the Measure of Insight into Cognition Clinician-Rated (MIC-CR) and Self-Report (MIC-SR) which assess awareness, attribution, and frequency of neurocognitive deficits. Assessments conducted one week apart allowed for internal consistency, test-retest reliability, concurrent validity, and administration order effects to be determined. In this study sample (N=80), 73 demonstrated significant neurocognitive impairment, yet on the MIC-CR, the average awareness and attribution scores indicated only partial awareness and attribution of neurocognitive deficits. The average MIC-SR score indicated rare frequency of cognitive difficulty in everyday life. Cronbach's alphas for the MIC-CR (0.83) and MIC-SR (0.93) were consistent with the initial validation study; test-retest reliability for both MIC assessments was significant (p<0.001). In new analyses, concurrent validity was demonstrated by significant correlations between the MIC-CR and MIC-SR in the first validation sample: r=-0.75 (p<0.001) and the current replication sample: r=-0.70 (p<0.001). Effect sizes were large when frequencies of neurocognitive complaints (MIC-SR) were compared between subjects with good versus poor awareness on the MIC-CR. Reliability and concurrent validity results were similar and significant regardless of order of MIC-CR and MIC-SR administration. The MIC-CR and MIC-SR are thus reliable and valid complementary instruments to assess neurocognitive insight in people with schizophrenia spectrum disorders.