Karen Yirmiya

Objectives: Mentalization-Based Treatment (MBT) requires rigorous fidelity assessment to ensure accurate delivery and validate treatment efficacy. This study introduces the Mentalization-Based Treatment Research Adherence and Competence Scale (MBT-RACS), a new instrument developed initially for research purposes to align with contemporary MBT principles and address psychometric and conceptual limitations found in earlier adherence assessment approaches. Methods: Inter-rater reliability of the MBT-RACS was evaluated using 126 recorded MBT sessions (104 group, 22 individual), rated by 17 trained coders. Results: The results indicated strong overall reliability, with most domains demonstrating good to excellent inter-rater agreement across both group and individual sessions, irrespective of ratings from two or three raters. Total adherence intraclass correlation coefficients (ICCs) were notably high for both group (.84) and individual (.95) sessions rated by two coders, substantially exceeding the reliability typically reported for comparable adherence instruments. Conclusions: These findings suggest that the MBT-RACS's format, which emphasizes broader, clinically meaningful domains, may contribute to improved consistency in ratings. The scale's robust reliability supports its applicability in research and clinical supervision, enhancing methodological rigour, quality assurance and targeted feedback for effective MBT training and implementation.