Neuropsychopharmacology: The Fifth Generation of Progress
itability (60% to 68%) for mother-reported attention problems (97).
Other studies of inattentive and hyperactive symptoms found a high heritability and minimal impact of the shared environment (98,99). Rhee et al. examined gender differ- ences in heritability using twin and sibling pairs from Aus- tralia (99). Specific genetic and environmental influences were highly similar for boys and girls. Slight differences that emerged were related to more influence of the shared environment in girls and some evidence genetic dominance in boys.
FIGURE 43.2. Heritability of ADHD.
their genes. Thus, the occurrence of twinning creates a natu- ral experiment in psychiatric genetics (91). If a disorder is strongly influenced by genetic factors, then the risk to co- twins of ill probands should be greatest when the twins are monozygotic. The risk to dizygotic twins should exceed the risk to controls but should not be greater than the risk to siblings.
Twin data are used to estimate heritability, which mea- sures the degree to which a disorder is influenced by genetic factors. Heritability ranges from zero to one, with higher levels indicating a greater degree of genetic determination. Figure 43.2 presents heritability data from 11 twin studies of ADHD. These data attribute about 80% of the origin of ADHD to genetic factors.
Goodman and Stevenson found the heritability of hyper- activity to be 64% (92,93). In a repeat analysis of these data, Stevenson reported that the heritability of mother- reported activity levels was 75%, and the heritability of a psychometric measure of attention was 76% (94). In a study of ADHD in twins who also had reading disability, Gilger et al. estimated the heritability of attention-related behaviors as 98% (95). In a study of 288 male twin pairs, Sherman et al. examined inattentive and impulsive-hyperactive symp- toms using both mother and teacher reports (96). Within both raters, the heritability of the impulsivity-hyperactivity dimension exceeded that of the inattention dimension; however, mothers’ ratings showed higher heritability than did teachers’ ratings. Specifically, mothers’ ratings produced a heritability of 91% for impulsivity and hyperactivity and 69% for inattention. Teachers’ ratings yielded a heritability of 69% for impulsivity and hyperactivity and 39% for inat- tention. Using the Child Behavioral Checklist as a dimen- sional measure, Hudziak and colleagues found a similar her-
Several twin studies examined the genetic contribution to the comorbidity of ADHD and other disorders. Data from Gilger et al. (95) were consistent with a prior family study (85) in suggesting that ADHD and reading disability were genetically independent; however, the existence of a genetically mediated subtype of both disorders could not be excluded. In contrast, two twin studies suggested that ADHD and reading disability share some genes in common (100,101). That this relationship may be complex is sug- gested by the report by Willicutt et al. of genetic overlap between reading disability and inattention but not between reading disability and hyperactive impulsive symptoms (102).
Nadder et al. examined whether ADHD and comorbid conduct and oppositional defiant disorder symptoms shared genetic risk factors (98). These investigators found that 50% of the correlation between the ADHD and comorbid con- duct was the result of shared genes. Similarly, the twin study of Silberg et al. found that genes influencing variation in hyperactivity scores were also responsible for variation in conduct problems (103). Between 76% and 88% of the correlation between hyperactivity and conduct scores were attributed to genes. These investigators concluded that the results were consistent with the existence of a biologically based group of children who manifest both hyperactivity and conduct disturbances. Further evidence that the ADHD plus comorbid conduct subgroup may be etiologi- cally meaningful comes from a study showing differences in serotonergic functioning between aggressive and nonag- gressive children with ADHD (104).
Like twinning, adoption provides another useful experi- ment for psychiatric genetics (91). Whereas parents can con- fer a disease risk to their biological children by both biologi- cal and environmental pathways, to adoptive children they can confer risk only by an environmental pathway. Thus, by examining both the adoptive and the biological relatives of ill probands, we can disentangle genetic and environmen- tal sources of familial transmission.
Adoption studies of ADHD also implicate genes in its origin. The adoptive relatives of children with ADHD are less likely to have ADHD or associated disorders than are the biological relatives of children with ADHD (105,106). Biological relatives of children with ADHD also do more