Objectives:
We report the effect of quality of therapy delivery on outcomes in a randomized, controlled trial of behavioural activation (BA) and guided self-help (GSH) for depression in adults with intellectual disabilities.

Methods:
A study specific measure of quality was used in a linear mixed effect model to determine the effects therapy and therapy quality on therapy outcome.

Results:
There was a significant interaction between quality and treatment type, with lower quality therapy associated with better outcome for GSH but poorer outcome for BA, with little difference in outcomes at higher levels of therapy quality.

Conclusions:
Factors suggesting high quality in individualized BA may indicate problematic engagement for GSH. More research into processes in therapy for people with intellectual disabilities is required.