UCLA

Attention-deficit/hyperactivity disorder (ADHD) is a highly heritable childhood-onset disorder characterized by developmentally extreme and impairing levels of inattention and/or hyperactivity/impulsivity that disrupt critical domains of academic, social, behavioral, emotional, and neuropsychological development. Improved diagnostic procedures are likely to facilitate early identification and timely delivery of interventions and to potentially reduce the risk of negative outcomes. Most research on the assessment of ADHD has focused on the clinical utility of DSM-IV symptoms and related clinical and diagnostic instruments, as well as the optimal symptom thresholds for accurate diagnosis. Unfortunately, examination of optimal strategies to integrate multi-informant data (e.g., parents and teachers), as well as development and evaluation of alternative empirically-derived assessment strategies across contexts (e.g., clinical settings, nationally representative samples), has received considerably less attention. Moreover, there is limited evidence on the association of different assessment strategies with respect to multi-domain impairment. The purpose of this investigation was to describe the utility of multiple ADHD assessment strategies in identifying functionally impaired children across multiple periods of development and in separate clinical and population-based samples. The first study compared multiple strategies for using single and multi-informant ADHD data in their prospective prediction of functional impairment. The second study aimed to develop alternative ADHD symptom algorithms for evaluation against the DSM-IV with respect to predictions of multi-domain impairment. The third study described the base rates and psychometric properties of ADHD symptoms in a nationally representative sample. In addition, this study compared the utility of ADHD symptom algorithms developed within clinical and population-based samples with respect to associations with academic and family functioning, as well as general health outcomes. The use of various symptom algorithms should consider the intended purpose of the assessment - that is, to rule in or rule out ADHD. Whereas more sensitive (i.e., inclusive) algorithms optimally identified individuals experiencing ADHD-related impairment in case-control samples, somewhat more specific (i.e., exclusive) algorithms were most useful for identifying impairment in a nationally representative sample. Finally, several non-DSM-IV algorithms outperformed the DSM-IV in identifying children exhibiting ADHD-related impairment. With additional validation, identifying and implementing alternative symptom algorithms for ADHD and other mental health conditions may improve treatment planning by allowing clinicians to target the most clinically significant symptoms.