Development of a scale to measure fidelity to manualized group-based cognitive behavioural interventions for people with intellectual disabilities.

The context for the present study was a cluster-randomized controlled trial of a group-based anger-management intervention, delivered by day-service staff. We aimed to develop a scale to measure the fidelity of manualized cognitive-behavioural therapy (CBT) delivered to adults with intellectual disabilities in group-based settings. A 30-item monitoring instrument (the MAnualized Group Intervention Check: MAGIC) was adapted from an existing fidelity-monitor instrument for individual CBT. Two sessions for 27 groups were observed by pairs of monitors who had no other contact with the intervention. 16 observers participated, in 15 unique pairings. Observers recorded high levels of inter-rater reliability and the scale had good internal consistency. Fidelity ratings predicted two key outcomes of the intervention, and were themselves predicted by the therapists' clinical supervisors.