30.1 REVIEW OF AND NEW 25-MONTH OUTCOMES OF A DOUBLE-BLIND NEUROFEEDBACK RANDOMIZED CONTROLLED TRIAL

Abstract

The objective of this session is to review main 25-month follow-up (FU) outcomes as background for new results and present some new secondary outcomes of the International Collaborative ADHD Neurofeedback (ICAN) randomized controlled trial. Children aged 7-10 years were randomized to 38 sessions of theta:beta ratio (TBR) neurofeedback (NF, n = 84) or a control treatment (Tx) of equal duration, intensity, and appearance without TBR downtraining (n = 58). The primary outcome was the average of parent- and teacher-rated inattention. Secondary outcomes included Clinical Global Impression (CGI), medication changes, functional outcome, change in diagnostic presentation, neurocognitive performance, cognitive tempo, IQ, achievement, and remission rate. The statistical analysis was intention-to-treat (ITT). Retention was 90% at 13 months and 85% at 25 months. The primary outcome improved by an effect size (ES) of d = 1.04 for NF, and d = 1.30 for controls, which is not significantly different. The nonsignificant advantage of NF in percent responders (Clinical Global Impression–Improvement [CGI-I] <3) at 13 months (63% vs 54%) reversed at 25 months (59% vs 66%). The NF group needed significantly less medication than controls at 13-month FU and nominally less at 25 months. The functional rating for both treatments by parent (d = 0.97-1.06) and teacher (d = 0.49-0.7) improved through 25 months (p = 0.002-0.0001). Reductions in ADHD presentation (combined to unimodal, inattentive to no ADHD) averaged 0.86 for NF and 0.69 for controls; combined continued at 25 months for 19% of NF, 41% of controls (n.s.). IQ improved significantly with NF but not controls at 13 months; improvement equalized by 25 months (significantly better for both). Sluggish cognitive tempo decreased significantly for both treatments without a significant difference. Reading achievement dropped significantly for controls at 6 months but recovered by 25 months. Math showed no significant change. Spelling declined for both treatments, significantly for NF. The significant advantage of NF in the remission rate at 13 months (40% vs 19%) attenuated by 25 months (33% vs 26%). For the whole sample, most outcomes in a well-controlled randomized controlled trial (ITT, N = 142) did not show a significant advantage of TBR NF over the control Tx despite favorable tendencies. A few significant 13-month FU outcomes attenuated by 25 months. However, moderator analyses in other presentations in this Symposium detect subgroups who respond significantly better to NF on the primary outcome and define biomarkers.