ABSTRACT
The present study explores the predictive power of seven neuropsychological assessment tools used in combination in classifying children with attention-deficit/hyperactivity disorder (ADHD). Twenty-one ADHD boys and 22 community control children participated. Group differences were significant on the continuous performance test only; however, battery analysis did increase overall predictive power, which was moderate. This study highlights the difficulty in identifying consistent mean differences on tests of frontal/executive functioning across studies, as well as the need to assess the predictive validity of these tests in classifying children with ADHD. The study suggests that these tests may provide greater predictive validity when used in combination. Inconsistencies in the literature are discussed, with consideration of research methodology, the heterogeneity of the ADHD population, and comorbid diagnoses.
Subject(s)
Attention Deficit Disorder with Hyperactivity/diagnosis , Neuropsychological Tests/statistics & numerical data , Attention , Cognition Disorders/diagnosis , Female , Humans , Male , Predictive Value of Tests , Psychometrics/statistics & numerical data , Psychomotor Performance , Reference Values , Reproducibility of Results , Risk AssessmentABSTRACT
ADHD adults (N = 26) were compared to normal controls (N = 26) on 6 neuropsychological measures believed sensitive to frontal lobe-executive functioning. MANOVA analyses and subsequent univariate tests indicated that most of the neuropsychological measures discriminated between the two groups. To address clinical significance diagnostic classification rates were also generated for each measure individually, and for the battery as a whole. Levels of positive predictive power (PPP) for each of the 6 measures (83-100%) indicated that abnormal scores on these tests were good predictors of ADHD. However, estimates of negative predictive power (NPP) suggested that normal scores poorly predicted the absence of ADHD. When classification rates were calculated for the overall battery classification accuracy improved substantially. Thus, neuropsychological tests can differentiate adults suffering from ADHD from adults without ADHD, while also providing good classification accuracy. Finally, the pattern of neurobehavioral impairments exemplified through the Summary Index scores was interpreted as consistent with conceptualizations of ADHD depicting mild neurologic dysfunction in networks associated with the frontal lobes.
