Estimating a minimum clinically important difference for the Developmental Behaviour Checklist - parent report.

Background: Evaluating the effectiveness of interventions relies on understanding what change in a main outcome is sufficient to be considered meaningful. Our aim was to estimate a Minimum Clinically Important Difference (MCID) for the Developmental Behaviour Checklist, parent-report (DBC-P)- a measure of behavioural and emotional problems in children and adolescents with intellectual disabilities. Methods: We generated distribution-based estimates through meta-analysis of intervention evaluations using the DBC-P as an outcome measure. We also generated anchor-based estimates using case scenarios with 10 parent carers and 21 professionals working with people with intellectual disabilities. Results: 21 studies were included in the meta-analyses and indicated an average DBC total raw score decrease of 3.01 or 4.73 (depending on analytic methods) in randomised controlled trials, and an average decrease of 9.16 points in pre-post designs. Parent carers provided a median MCID estimate of 6 (IQR 4, 7) and professionals provided a median estimate of 8 (IQR 5, 14). Conclusions: These findings contextualise DBC-P score changes in relation to outcomes from other interventions and parent carer and professional views. Which MCID value to choose depends on what factors are prioritised for an intervention.