An attachment perspective on the risk for psychosis: Clinical correlates and the predictive value of attachment patterns and mentalization.

Background: In a longitudinal design, the present study examined: (a) the nature of attachment patterns and reflective functioning (RF) in individuals at ultra-high risk for psychosis (UHR); (b) the association between RF and subclinical psychotic symtoms; and (c) the psychosis-predictive value of attachment patterns, RF, and the interaction between these factors.

Method: The sample comprised 57 UHR subjects and 53 clinical controls (non-UHR). UHR subjects were followed over a mean period of 14 months (SD = 2.7; range 11-19), during which time 11.5% developed psychosis. Attachment patterns and RF were measured. Hierarchical logistic regression was used to examine the predictive value of variables in the transition to psychosis.

Results: At baseline, significant differences were found between groups in secure attachment patterns and RF. RF was negatively associated with the attenuated psychotic symptoms described by the Unusual Thought Content/Delusional Ideas, Suspiciousness/Persecutory Ideas, and Disorganized Communication subscales. Moreover, poor mentalization was related to an increased likelihood of developing a psychotic disorder (OR = 0.427, 95% CI [0.188, 0.970]); conversely, no predictive effects of the attachment variables and their interaction with RF on the transition to psychosis were found. Notably, an optimal threshold value of RF = 1.25 was found to distinguish UHR subjects who made the transition to psychosis from those who did not develop this pathology.

Conclusion: The results suggest that aberrant mentalizing patterns can predict the development of psychosis. Clinical implications of these results for the development of preventive treatments are discussed.