The Alliance and Rupture Observation Scale (AROS): Development and validation of an alliance and rupture measure for repeated observations within psychotherapy sessions.

OBJECTIVE

The aim of this study was to test a new observer-rated instrument, the Alliance and Rupture Observation Scale (AROS). It was designed for repeated measurements of the alliance within sessions and to detect alliance ruptures.

METHOD

Videotaped therapy sessions with depressed adults were analyzed. Reliability was mainly assessed as inter-rater reliability. Convergent, predictive, and discriminant validity of the AROS was assessed by comparing the instrument with both observer-rated and patient-rated measures.

RESULTS

The AROS exhibited excellent inter-rater reliability. Alliance levels measured with the AROS predicted patients' ratings of the alliance in the same session and were highly correlated with another observer-rated alliance measure. Alliance patterns (rupture; repair; and no-rupture) based on AROS scores were significantly correlated with patients' ratings of the alliance.

CONCLUSIONS

Preliminary support for convergent and predictive validity was found. It is yet to be determined whether AROS scores are related to psychotherapy outcomes.