Relationship between treatment duration and the efficacy of pharmacological treatment for ADHD: a meta-analysis and meta-regression

Abstract

Background. Attention deficit and hyperactivity disorder (ADHD) is one of the most common mental disorders affecting children and adolescents. When non-pharmacologic therapies are insufficient, clinical guidelines recommend pharmacological treatment. However, these recommendations are limited by the wide heterogeneity of the effect size of these drugs among randomized, double-blind, placebo-controlled clinical trials (RCT). As tolerance or sensitization development may be responsible for the changes over time of pharmacological ADHD treatment efficacy, treatment duration has been proposed as one efficacy moderator. Objectives. To determinate the relationship between the duration of pharmacological treatment in RCT and its efficacy in improving the severity of symptoms in patients with ADHD. In addition, to determinate the efficacy of pharmacological treatment, the variability of the

efficacy between-studies, and its relationship with patient, intervention and study design- related covariates, other than treatment duration.

Methods. A systematic review of RCT investigating the efficacy of pharmacological treatment for patients with ADHD was carried out. Mean difference (MD) and 95% confidence interval (CI) of the efficacy results from each RCT were calculated. MD were pooled using a random effects model. Heterogeneity was assessed using the I2 parameter and publication bias by drawing a funnel plot. The potential confounding effect of patient, intervention and study design-related covariates in the relationship between treatment duration and efficacy were identified using weighted lineal regression and meta-regression. The relationship between treatment duration and efficacy was studied using meta-regression, which included the potential confounding covariates identified before in a stepwise manner. Results. Up to 81 RCT that investigated the efficacy of ADHD treatment were included. These RCT involved 91 drug-placebo comparisons, enrolled 18,055 patients and lasted from 3 to 28 weeks. Overall efficacy was -7.44 (95%CI from -8.18 to -6.69) with a high heterogeneity (I2=73.1%). Funnel plot was reasonably symmetrical. Potential confounding variables identified were: baseline ADHD severity, type of treatment (psycho- vs. non-psychostimulant drugs) and comorbidity. The effect of treatment duration over efficacy was 0.24 (p-value = 0.0016) and 0.18 (p-value = 0.0587), in the unadjusted and adjusted models, respectively. Conclusion. No relationship between treatment duration and efficacy of the pharmacological treatment in improving the severity of the symptoms in patients with ADHD was found