From Paul D on the ADCA update list:

This study contrasts 410 children diagnosed with ADHD against 1156 unrelated (but ethnically matched) controls, looking for large (and so easily detected) copy number variants (or CNVs). These rare chromosomal deletions and repetitions have been shown to be associated with other neurodevelopmental problems including schizophrenia and autism). Findings were replicated in 825 Icelandic patients with ADHD and 35 243 Icelandic controls.

The study detected twice as many large CNVs in the ADHD population than amongst controls. Amongst people diagnosed with ADHD, those with the most severe ADHD (IQ < 70) had six times as many CNVs. Of 15 specific CNVs, 11 were also found in the parents of ADHD sufferers, suggesting heritability. However 4 were not inheritied from parents, suggesting insult or injury in-utero may have caused these particular copy-errors. As there is some overlap with CNVs previously implicated in Schizophrenia, (chromosome 16) the emerging picture appears to be of a heritable neurodevelopmental problem that can be mediated by environmental influence. Only 16% of examined ADHD sufferers showed the “massive” CNVs that were tested for. Future research will examine smaller CNVs. If smaller CNVs are detected in similar locations in a larger percentage of ADHD sufferers, the implication would be that these polymorphisms interfere with neurogenisis or neuronal development in-utero or early infancy, increasing susceptibility to a number of psychiatric conditions. Discussion at NS; http://www.newscientist.com/article/dn19528-have-gene-findings-taken-the-stigma-from-adhd.html