RESUMO
PURPOSE: Disorganised attachment is a key concept in understanding the development of psychosis. However, existing questionnaires of adult attachment do not adequately measure this construct hindering future research into the psychosocial causes of psychosis. The most widely measure of adult attachment in people experiencing psychosis is the Psychosis Attachment Measure (PAM). The measure has recently been revised to include disorganised attachment items. This study develops previous research by providing a rigorous examination the psychometric properties of the revised questionnaire (PAM-R). METHODS: A total of 407 participants with self-reported experiences of psychosis completed a battery of questionnaires which included the PAM-R and other measures which were conceptually related to the concept of disorganised attachment. RESULTS: Confirmatory factor analysis (CFA) indicated a three-factor solution with factors corresponding to anxious, avoidant, and disorganised attachment. The majority of the fit statistics were acceptable with the exception of the RMSEA statistic. Internal consistency and test-retest reliability were good for all subscales. The disorganised subscale correlated in expected directions with other measures of attachment, dissociation, trauma, and psychotic experiences. CONCLUSION: The PAM-R is a valid and reliable measure of adult attachment. It is a practical assessment tool for clinicians and researchers to measure insecure and disorganised attachment patterns that is acceptable to people experiencing psychosis.
RESUMO
BACKGROUND: Secure attachment in adulthood is associated with many markers of adaptive functioning. Valid and reliable self-report measures of attachment security could provide a practical tool to help advance strengths-based research and clinical work. Previous reviews have not specifically examined the psychometric properties of self-report instruments with respect to secure attachment or systematically appraised the methodological quality of relevant validation studies. METHOD: A systematic review was completed in accordance with the COSMIN guidelines for reviews of patient-reported outcome measures. The methodological quality of individual studies was evaluated, and results were rated against criteria for good measurement properties. RESULTS: A total of 40 studies were included in the review, which collectively reported on 24 self-report instruments. The methodological quality of individual studies was variable, and no single instrument was identified as having sufficient evidence of a range of psychometric properties. However, the Attachment Style Questionnaire-Short Form (ASQ-SF), the Cartes-Modèles Individuels de Relations (CAMIR), Cartes-Modèles Individuels de Relations-Reduced (CAMIR-R), and the Psychological Treatment Inventory-Attachment Style Scales (PTI-ASS) had the most robust evidence for the properties assessed. CONCLUSION: Existing self-report measures assessing adult secure attachment have limited psychometric support. More methodologically robust studies of content validity, reliability, measurement invariance, and construct validity in particular are needed.
