Etiology of Inattention and Hyperactivity/impulsivity in a Community Sample of Twins with Learning Difficulties.

Erik G. Willcutt [1,3]

Bruce F. Pennington [2]

John C. DeFries [1]

A community sample of 373 8 to 18 year-old twin pairs in which at least one twin in each pair exhibited a history of learning difficulties was utilized to examine the etiology of inattention and hyperactivity/impulsivity (hyp/imp). Symptoms of attention-deficit/hyperactivity disorder (ADHD) were assessed by the DSM-III Diagnostic Interview for Children and Adolescents. Inattention and hyp/imp composite scores were created based on results of a factor analysis. Results indicated that extreme ADHD scores were almost entirely attributable to genetic influences across several increasingly extreme diagnostic cutoff scores. Extreme inattention scores were also highly heritable whether or not the proband exhibited extreme hyp/imp. In contrast, the heritability of extreme hyp/imp increased as a linear function of the number of inattention symptoms exhibited by the proband. This finding suggests that extreme hyp/imp may be attributable to different etiological influences in individuals with and without extreme inatten tion. If this result can be replicated in other samples, it would provide evidence that the hyp/imp symptoms exhibited by individuals with Combined Type ADHD and Predominantly Hyp/Imp Type ADHD may be attributable to different etiological influences.

KEY WORDS: ADHD; inattention; hyperactivity; twins; etiology.

Previous twin studies of attention-deficit/hyperactivity disorder (ADHD) have suggested that individual differences in symptoms of ADHD are largely attributable to genetic influences (Edelbrock, Rende, Plomin, & Thompson, 1995; Eaves etal., 1997; Gjone, Stevenson, & Sundet, 1996; Levy, Hay, McStephen, Wood, & Waidman, 1997; Nadder, Silberg, Eaves, Maes, & Meyer, 1998; Schmitz, Fulker, & Mrazek, 1995; Sherman, Jacono, & McGue, 1997, Thapar, Hervas, and McGuffin, 1995). Estimates of heritability ([h.sup.2]), the proportion of individual differences in ADHD symptoms attributable to the influence of genes, have been consistently high in multiple samples utilizing different measures (mean [h.sup.2] [approx] 0.75). Moreover, studies of selected groups have indicated that extreme ADHD scores are also largely attributable to genetic influences, with group heritability estimates ([[h.sup.2].sub.g] = 0.76-0.92) similar to the estimates obtained for individual differences across the entire symptom dimension (Gillis, Gilger, Pennington, & DeFries, 1992; Gjone et al., 1996; Levy et al., 1997; Stevenson, 1992). The similarity of these estimates is consistent with the hypothesis that the same genetic influences may be associated with both extreme ADHD scores and individual differences across the entire distribution of symptoms.

All but one (Sherman et al., 1997) of these studies utilized broad-band measures that combine symptoms of inattention, hyperactivity, and impulsivity to create a single ADHD score. In contrast to this unidimensional conceptualization, exploratory and confirmatory factor analyses have indicated consistently that symptoms of ADHD comprise separate but correlated dimensions of inattention and hyperactivity/impulsivity (Bauermeister, 1992; Bums, Walsh, Patterson, etal., 1997; Burns, Walsh, Owen, and Snell, 1997; Conners, Sitarenios, Parker, & Epstein, 1998a, 1998b; DuPaul et al., 1997, 1998; Healey et al., 1993; Hudziak et al., 1998; Pelham, Gnagy, Greenslade, & Milich, 1992; Sherman et al., 1997; Willcult & Pennington, in press). Consequently, the fourth edition of the Diagnostic and Statistical Manual of Mental Disorders (4th ed.) (DSM-IV; American Psychiatric Association, 1994) described three subtypes of ADHD based on elevations of inattention and hyperactivity/impulsivity (hyp/imp). The Predominantly Inatte ntive Type describes individuals with significant elevations of inattention in the absence of significant hyp/imp; the Predominantly Hyperactive! Impulsive Type (Hyp/Imp Type) includes children with significant elevations of hyp/imp but not inattention; and the Combined Type describes individuals with elevations of both inattention and hyp/imp.

The single twin study (Sherman et al., 1997) that subdivided symptoms of ADHD into factors of inattention and hyp/imp found that individual differences in symptoms of inattention and hyp/imp were each highly heritable. In addition, this study found significant bivariate heritability for the two dimensions, suggesting that the same genetic influences contribute to individual differences in symptoms of inattention and hyp/imp. These results provide preliminary evidence consistent with the DSM-IV model of ADHD. However, the sample utilized by Sherman and colleagues was not large enough to examine the etiology of extreme inattention or hyp/imp, an analysis which has essential implications for the validity of the diagnostic taxonomy of DSM-IV ADHD.

The Current Study

This report describes a study that replicates and extends previous research in several ways. First, multiple diagnostic cutoff scores for ADHD were utilized to test if the etiology of extreme ADHD scores varied as a function of severity. Symptoms of ADHD were then subdivided into dimensions of inattention and hyp/imp, facilitating the comparison of the etiological influences that contribute to extreme scores on these symptom dimensions. Finally, a series of analyses were conducted to test whether the etiology of extreme inattention or hyp/imp differed as a function of scores on the other dimension. Results of these analyses have implications for the validity of the three subtypes of DSM-IV ADHD. Because the DSM-IV subtypes are proposed to represent variations of the same disorder, similar etiological influences would be expected to contribute to extreme inattention or hyp/imp scores regardless of the child's score on the other dimension (e.g., the etiology of extreme hyp/imp would be similar among individual s with the Hyp/Imp Type or the Combined Type). On the other hand, if the etiology of extreme scores on one dimension (e.g., inattention) is different among children with extreme hyp/imp (Combined Type) and children with solitary elevations of inattention (Inattentive Type), this finding would be consistent with the hypothesis that the Combined Type and Inattentive Type were attributable to different etiological influences.

METHOD

Participants

A total of 373 same-sex pairs of 8 to 18-year-old twins from the ongoing Colorado Learning Disabilities Research Center (DeFries et al., 1997) were included in these analyses. The total sample included 215 MZ twin pairs (100 female, 115 male) and 158 DZ pairs (70 female, 88 male) selected because at least one member of the pair exhibited evidence of reading or mathematics difficulties in their school records. Such evidence included low standardized test scores, referrals for academic tutoring, or special education placement. Because children with learning difficulties are at increased risk for ADHD (Fergusson & Horwood, 1992; McGee & Share, 1988; Willcutt & Pennington, in press), we anticipated that the prevalence of ADHD would be higher in this enriched sample than in the general population.

Measures

Zygosity Determination

Zygosity of the twins was determined using a modified version of the Nichols and Bilbro zygosity questionnaire (1966), which has been shown to reliably classify 96% of twins. If zygosity classification was ambiguous based on the …