Patient characteristics' moderation of multinutrients treatment response in school-age children with attention deficit hyperactivity disorder.

Objective: This study analyzed data from the Micronutrients for ADHD in Youth (MADDY) 8-week RCT to identify moderators of treatment response to multinutrients, i.e., patient characteristics likely to predict optimal benefit based on the Clinical Global Impression-Improvement (CGI-I)- scale.

Methods: Classification tree analysis investigated demographic and clinical history variables as potential moderators of treatment response. Effect sizes (ES) with confidence intervals (CI) were generated using bivariate analysis of risk difference of proportions derived from the classification tree between-treatment effects stratified by the moderator variable.

Results: The classification tree showed a higher response rate among multinutrient-treated children (n = 70) with university-educated parents compared to those with parents with less than a university education (64% vs. 36%); a finding also supported by the bivariate analysis (ES = 0.35, 95% CI = 0.03-0.67). Within university-educated parents, a larger percentage of children with past ADHD medication use were responders compared to no past ADHD medication use (83% vs. 52%). Classification tree also revealed that among multinutrient-treated children with university-educated parents without past ADHD medication, higher baseline CGI-Severity score was responders compared to lower baseline score (67% vs. 33%). Bivariate analysis predicted higher response in participants whose parents reported alcohol use disorder (ES = 0.43, 95% CI = 0.10-0.76), or illegal drug use (ES = 0.68, 95% CI = 0.52-0.84), or learning problems (ES = 0.44, 95% CI = 0.13-0.75).

Conclusions: Parent education level, substance use disorder or learning problems, and child's previous medication use moderated multinutrient response in children with ADHD. These findings have implications for future research and clinical practice.