| Attention-Deficit/Hyperactivity Disorder (ADHD) is a neurodevelopmental disability characterized by symptoms of inattention, hyperactivity and impulsiveness, which may cause impairment in a variety of cognitive, social, behavioral, and emotional domains. Extensive research has been conducted to determine the best course of treatment for individuals suffering from the disorder. The two most popular forms of treatment research are pharmacological (e.g. stimulant medication) and non-pharmacological (e.g. behavior management) in nature. Although studies of both pharmacological and non-pharmacological treatment show significant effects, there is a great degree of variation amongst methodology and findings, which has led to confusion with regards to the most appropriate method of treatment. Thus, there is a need for greater consensus in determining which treatments work best. To help address this issue, researchers will often turn to meta-analyses, a systematic approach to consolidating research findings, to resolve issues related to discrepant results. The current review examined methodological procedures and outcome data from meta-analyses of ADHD interventions in order to identify the most reliable, valid, and replicable methodological processes amongst meta-analyses. Forty-six meta-analyses of pharmacological and non-pharmacological ADHD interventions met search criteria. Coding of meta-analyses focused on four domains related to best practices in conducting meta-analytic studies: 1) Literature Search, 2) Study Characteristics, 3) Research Methods, and 4) Effect Size Computation. The present review assigned a quality index score to meta-analyses based upon the collection of methodological variables specified within the coding manual. Results found variability across meta-analyses in research design, methodological characteristics, types of treatments reviewed, outcome measures, and the reporting of effect sizes. Variability across meta-analyses limited the extent by which quantitative comparisons could be made, thus limiting the bulk of the current review to descriptive data. Some notable findings included significant differences in the types of outcomes reported by pharmacological versus non-pharmacological treatments and significant differences in the quality ratings of meta-analyses as function of time, with recent meta-analyses exhibiting higher quality ratings. |
