Service user attachments to psychiatric key workers and teams.

Purpose: The first aim of the study was to evaluate whether self-reported attachment styles of individuals with psychosis are consistent with their self-reported attachment in therapeutic relationships with both key workers and mental health teams. The second aim was to evaluate the level of concordance in attachment ratings given by different raters (self-report, key worker informant-report and team informant-report).

Methods: Three self-report versions of the Psychosis Attachment Measure (PAM; attachment in general relationships, attachment towards key worker and attachment in relation to the mental health team) were administered to 24 individuals with a diagnosis of psychosis in psychiatric rehabilitation settings. Key worker and 'team' informant versions of the PAM were also completed.

Results: There were strong, significant correlations among the three self-reported attachment measures. There was less consistent evidence of correlations between key worker ratings of attachment and self-report attachment ratings. The majority of the correlations between team ratings of attachment and self-report attachment were small and non-significant.

Conclusions: Strong correlations among the self-reported PAM scales suggest that self-reported attachment in specific therapeutic relationships is consistent with self-reported attachment in general relationships. The self-ratings were not consistently correlated with informant-ratings and team informant ratings were particularly poorly correlated with self-ratings. This suggests that it is vital that teams consult service users themselves when making decisions about their care.