Supplementary Material for: Efficacy of a Stepped Care Approach for Adolescents and Adults with Attention-Deficit/Hyperactivity Disorder (ADHD): An Adaptive Intervention Study Including Randomized Controlled Trials (ESCAlate)

Introduction: This study evaluated the effectiveness of a stepped-care approach — a staged model that escalates from lower- to higher-intensity treatments according to clinical response — for the treatment of adolescents and adults with ADHD, taking into account symptom severity and prior treatment response. Methods: In a multicenter study, adolescents and adults with ADHD (16-45y) participated in a two-step treatment program including randomized controlled trials. Step1: Patients were (block-)randomized to 3 months of Psychoeducation (PE), Telephone-assisted self-help (TASH), or waiting control (WC). Step2: Based on Step1-response (full, partial, none), patients received either counseling or were randomized to counseling with/without neurofeedback (NF) or pharmacological treatment (with/without NF) for 6 months. The primary outcome was change in clinician-rated ADHD symptoms, analyzed using linear (mixed-)effects models for repeated measures to account for correlations within participants over time. Results: Between 01/2015-09/2020, N=299 (mean age=28y, 55.2% male) patients were randomized in Step1. The primary outcome showed no significant between-group differences (PE vs. TASH: d=-0.12, 95%-CI [-3.18,1.19], p=0.64; PE vs. Control: d=-0.26, 95%-CI [-4.25,0.05], p=0.13; TASH vs. Control: d=-0.14, 95%-CI [-3.24,1.04], p=0.57). However, significant within-group effects emerged (PE: d=-0.60; TASH: d=-0.48; WC: d=-0.34; p<0.001). Step2 also showed no between-group differences but significant within-group effects (MPH: d=-0.59; MPH+NF: d=-0.76; counseling: d=-0.57/-1.35, p=0.01/p<0.001). Mixed-models revealed symptom reduction in all Step1 responders and Step1 non-responders in Step2. Conclusion: The lack of Step1-RCT differences questions the specific effects of PE/TASH. Similar patterns emerged in Step2, but intensified treatment for Step1 non-responders improved outcomes in MPH groups. Step1 response influenced later treatment success. Some stepped-care combinations did not further reduce symptoms, but no rebound effects occurred. The main limitations of this study are the two-step design complexity, limited information on certain (additional) psychosocial components, and the need to make assumptions about missing data. Nevertheless, findings support the feasibility and partial effectiveness of a stepped-care approach.