To explore different longitudinal patterns of social functioning before onset of psychotic illness and how they relate to clinical presentation, substance use and acute treatment response.

Inclusion criteria: Drug-naïve first-episode psychosis, 18-50 yo, criteria for Schizophrenia or Other Psychotic Disorders (DSM-IV), excluding Psychotic Disorder due to a General Medical Condition and Substance-Induced Psychotic Disorder.

Mental Retardation, neurological disease, brain injury or drug dependence.

Premorbid Adjustment Scale (PAS), Scale for the Assessment of Positive Symptoms (SAPS), Scale for the Assessment of Negative Symptoms (SANS).

Ward cluster analyses were carried out to differentiate three longitudinal patterns of social premorbid adjustment from childhood to late adolescence: stable good (N=75), stable bad (N=44) and deteriorating (N=35). Chi-square and ANOVA tests were used.

154 subjects (64.5% male, mean age 26.81, SD=6.98) participated in the study.

At baseline the socially stable good group had more positive symptoms, SAPS 13.85 (3.99), than the stable bad group, SAPS 11.82 (3.93) (p=0.023).

At six weeks post-treatment the socially deteriorating group had more negative symptoms, SANS 8 (4.89), than the stable good, SANS 3.85 (4.11), and the stable bad, SANS 5.23 (5.45) (p=0.000).

The stable good group had the highest rates of substance use, followed by the deteriorating group.

A good premorbid social life was related to higher substance use and more positive symptoms at presentation. A social deteriorating pattern was associated with more negative symptoms at baseline and six weeks post-treatment. These differences would argue for different pathogeneses of psychosis.