Content area
Full text
Maltreatment poses severe risks to children's health and development and is increasingly coming to the attention of primary care clinicians and other community professionals (Cicchetti & Manly, 2001; Sedlak & Broadhurst, 1996). A major consequence of maltreatment in early childhood is antisocial behavior (Dodge, Bates, & Pettit, 1990; Lansford, Dodge, Pettit, Bates, Crozier, & Kaplow, 2002; Smith & Thornberry, 1995; Widom, 1989; Widom & Maxfield, 2001). Such early-onset antisocial behavior is, in turn, associated with life-long and pervasive mental (Moffitt, Caspi, Harrington, & Milne, 2002), physical (Farrington, 1995), economic (Caspi, Wright, Moffitt, & Silva, 1998), and interpersonal (Moffitt et al., 2002) problems that create an enormous public-health burden (Potter & Mercy, 1997).
However, not all maltreated children develop conduct problems (Widom, 1997), and some maltreated children exhibit adaptive functioning (Cicchetti, Rogosch, Lynch, & Holt, 1993) that is still evident in adulthood (McGloin & Widom, 2001). Very little systematic evidence is available to explain why children show such marked variation in their response to maltreatment. Such variability would be observed if there were genetically influenced individual differences in susceptibility to environmental experiences like maltreatment, a concept referred to as "gene-environment interaction" (GxE; Kendler & Eaves, 1986; Rutter & Silberg, 2002). The GxE concept is familiar to clinicians as the "host-pathogen interaction" in a patient's vulnerability or resistance to disease (Evans & Relling, 1999; Hill, 1999) and to developmentalists as the "diathesis-stress" interaction in an individual's vulnerability or resistance to pathogenic experiences (Monroe & Simons, 1991). In this article, we test the hypothesis that children's conduct problems emerge when genetically vulnerable children encounter family environments in which they are maltreated. That many children are not at genetic risk for conduct problems may explain, in part, why the experience of maltreatment does not always result in the development of conduct problems.
To analyze GxE with real precision, one must specify both the environmental risk and the genotype. Although twin studies of antisocial behavior have shown that genetic factors influence early-emerging conduct problems (Arseneault, Moffitt, Caspi, Taylor, Rijsdijk, Jaffee, Ablow, & Measelle, 2003; van den Oord, Verhulst, & Boomsma, 1996; van der Valk, Verhulst, Neale, & Boomsma, 1998), functional genes for these problems have yet to be reliably identified. There are some promising candidates (Rowe, 2001), and one study...