Anne Fleur Kortekaas-Rijlaarsdam

CHAPTER 3 74 between ADHD and academic underperformance is strongest in primary school (Daley & Birchwood, 2010). Fourthly, teacher ratings of ADHD symptoms on the DBDRS were not obtained for the TD group. However, the chance of including children with ADHD in our TD group was minimized through screening for psychiatric morbidity using parental reports about psychiatric morbidity and reports about the consultation of mental health specialists. TD children were excluded if they scored above the 90 th centile on either the Inattentive or Hyperactive/Impulsive scale of the DBDRS or if reported to have ever been diagnosed with a psychiatric disorder based. Also, in our sample the estimated IQ of children with ADHD was lower than the IQ of TD children. We did not control for this because controlling for IQ in neurodevelopmental disorders such as ADHD is doubtful as IQ is not independent of the disorder and differences between ADHD- and TD groups are pre-existent and not due to selection bias (Dennis, Francis, Cirino, Barnes, & Fletcher, 2011). Although we decided to control for socio- economic-status (SES) when comparing our ADHD and TD group, a common causal factor for both lower SES and a diagnosis of ADHD cannot be ruled out. Finally, behavioral outcomes were evaluated by teachers and parents only, not clinicians, as the focus of this research lays on academic achievement, therefore making teachers and parents the most reliable sources of information. Clinical implications. The finding that behavioral improvements with methylphenidate mediate the effects of this number one treatment for ADHD (Antshel et al., 2011) on math productivity are important because improvements in school performance are a major treatment goal. It is promising that methylphenidate-related reductions in parent-rated ADHD symptoms are related to improvements in math productivity with methylphenidate because behavioral response is a determinant for titration of medication. However, our results also confirm previous research that the effects of methylphenidate are selective and only significant for math and not for reading and spelling. This is in contrast with the generally high expectations of parents, teachers and treating physicians on the efficacy of methylphenidate in academic improvement. If further research on the moderating effects of learning abilities confirms the importance of initial academic abilities for positive medication effects on school performance, current grades should be taken into account when choosing optimal treatment. In summary, small to medium effects of methylphenidate on mathematical productivity and accuracy were shown. Methylphenidate did not improve spelling performance or reading productivity. Improvements in math productivity, but not accuracy, were related to parent-rated improvements in ADHD symptoms. Thus, children showing