The Italian version of the Reflective Functioning Questionnaire: Validity data for adults and its association with severity of borderline personality disorder

Impairments in the ability to understand others and the self in terms of internal mental states (reflective functioning [RF] or mentalizing) are thought to play a key role in the development of borderline personality disorder (BPD). The first aim of this study was to validate the Italian version of the Reflective Functioning Questionnaire (RFQ), a brief self-report measure of RF, by examining its factor structure with Principal Component Analyses (PCA), and correlations with constructs that should be theoretically related to RF. In addition, we investigated whether the RFQ could empirically distinguish between healthy controls and carefully diagnosed BPD patients using Research Operating Curve methods, and was related to severity of borderline pathology as measured with the Shedler-Westen Assessment Procedure (SWAP), an observer-rated measure of BPD pathology. An Italian translation of the RFQ was administered to a sample of 154 healthy controls and a clinical sample of 59 BPD patients diagnosed with the Structured Clinical Interview for DSM-IV Axis II disorders. Clinical severity of BPD was assessed with the SWAP. Normal controls completed self-report inventories of constructs related to RF (mindfulness, empathy, Theory of Mind, alexithymia, and autistic traits). PCA confirmed the a priori factor structure in the Italian translation of the RFQ, showing two subscales that measure certainty and uncertainty about mental states, with satisfactory reliability and construct validity. These dimensions also distinguished BPD patients from healthy controls (p < 0.05). ROC analyses showed that the uncertainty subscale discriminated BPD patients from healthy individuals (area under the curve = 78%, cut of 4.5 points, sensitivity = 73%, specificity = 68%). Within the patient group, regression analyses showed uncertainty about mental states to have a significant unique contribution in predicting BPD severity (p < 0.05), explaining 12% of the variance. Results largely supported the reliability and validity of the Italian version of the RFQ. These findings also provide further evidence for the role of impairments in mentalizing and reinforce the rationale for offering mentalization-based interventions to individuals with this disorder.