Can insight be predicted in first-episode psychosis patients? A longitudinal and hierarchical analysis of predictors in a drug-naïve sample.

Poor insight is a ubiquitous phenomenon in psychosis with great repercussions on clinical practise and the outcomes of patients. Poor insight comprises "state" and "trait" components. This paper targeted predictors of global insight and insight dimensions at baseline in the drug-naïve status of first-episode psychosis patients and during a 6-month follow up after episode remission. Seventy-seven consecutive and previously unmedicated patients with first-episode schizophrenia-spectrum disorders (FESSD) completed baseline and 6-month insight, premorbid, symptomatological and neuropsychological assessments. Insight measures served as dependent variables for a set of hierarchical multiple regression models. Premorbid personality abnormalities and duration of untreated psychosis (DUP) significantly predicted 'state' and 'trait' insight global scores. Duration of untreated psychosis (DUP) significantly predicted 'state' insight, measured as refusal of treatment at baseline. Moreover, premorbid personality abnormalities and DUP with minor contributions of demographic variables, cognitive functioning and psychopathological dimensions predicted 'trait insight', defined as insight after remission of the psychosis episode 'Insight improver' FESSD patients showed better late adolescent premorbid adjustment, lower personality disturbances (sociopathic, schizoid and schizotypy dimensions), shorter DUP, and lower positive, negative and disorganisation symptoms and better cognitive performance on the Trail Making B test at the 6-month follow-up assessment. Premorbid personality abnormalities and DUP were predictors of 'state' and 'trait' insight, both at global scores and dimension levels. Moreover, insight improvement in patients with FESSD was related to premorbid abnormalities (in both adjustment and personality), shorter DUP, fewer positive and negative symptoms and better performance in cognitive tests at the 6-month follow up.