Genetics and developmental psychopathology: 1. Phenotypic assessment in the Virginia Twin Study of Adolescent Behavioral Development

Abstract

We introduce an overlapping cohort sequential longitudinal study of behavioral development and psychopathology in a representative sample of 1412 pairs of twins aged 8 through 16 years. Multiple phenotypic assessments involve a full psychiatric interview with each child and each parent, and supplementary parental, teacher, and child interview material and questionnaires. For the first wave of assessments, the numbers of reported DSM-III-R symptoms of Major Depressive Disorder (MDD), Separation Anxiety Disorder (SAD), Overanxious Disorder (OAD), Oppositional Defiant Disorder (ODD), Conduct Disorder (CD), and Attention Deficit Hyperactivity Disorder (ADHD), assessed through interviews, confirm patterns of age and sex trends found in other epidemiological samples, but underscore their dependence on whether the child or the parent is the informant. Correlations across domains for symptoms reported by the same informant are often as large as correlations across informants for the same domain of symptoms. Factor analyses of these symptom counts, taking account of informant view and unreliability of assessment, show the high degree of correlation between SAD and OAD, between MDD and OAD, and between CD and ODD. ADHD symptoms are relatively independent of the other domains, but show moderate correlations with CD, ODD, and MDD. Factorially derived dimensional questionnaire scales, based on child, parental, and teacher reports, show patterns of relationship to symptom counts consistent with both convergent and discriminant validity as indices of liability to clinical symptoms. Across informants, questionnaire scales provide as good a prediction of symptoms as do clinical interviews. Multitrait-multimethod confirmatory factor analysis reveals the patterns of relationship between symptoms of psychiatric disorder in children taking due account of informant and unique sources of variance. Gender differences are consistent within the correlated clusters of ODD/CD and MDD/SAD/OAD, although there are disorder-specific age trends. There are large informant-specific influences on the reporting of symptoms in clinical interviews. Dimensional questionnaire scales provide a useful source of additional information. In subsequent analyses of genetic and environmental etiology of childhood psychopathology we must expect that results may differ by informant and method of assessment. Multivariate and developmental analyses that explore the sources of these differences will shed new light on the relationship between genetic and environmentally influenced vulnerability and the manifestation of psychopathology in specific circumstances.