Abstract: Objective To study the relationship between the clinical features, cognitive function, duration of untreated psychosis (DUP), and social function of patients at ultra-high risk of psychosis. Methods The data about clinical information, cognitive function, DUP, and social function of 49 patients were collected and analyzed. Results There was no significant correlation between social function and age, age of onset, education years and DUP (P>0.05). The social function showed a significant negative correlation with total scores of Scale of Psychosis-risk Symptoms (SOPS) and negative symptoms scores (r=-0.357, P<0.05; r=-0.348, P<0.05); It also showed a significant negative correlation with trail making test A (r=-0.32, P<0.05), a positive correlation with the simple visual memory test, the Stroop test and the continuous performance test (CPT) (r=0.324, P<0.05; r=0.295, P<0.05; r=0.407, P=0.000). An optimal linear model was developed by using the stepwise regression (R²=0.231, F=13.491, P=0.001). The relationship between social function and CPT was linear (unstandardized coefficients B=6.046, 95%CI:-0.681-0.494, P=0.001), whereas other factors did not show significant linear relationship with social function (P>0.05). Conclusion The patients' clinical symptoms and cognitive function are closely related with social function, which could be as reference for intervention strategies for ultra-high risk group.

Key words: ultra-high risk for psychosis, social function, affect factors