Compared with children of nondepressed parents, children of depressed parents have a higher risk for depression (1
) as well as for attention problems, behavior management problems, and conduct disorder (2
). While there has been considerable research on mechanisms of risk in these families, no study has provided a direct test of the extent to which there is an environmental effect of parental major depressive disorder (that is, separate from genetic influences) on psychopathology in children. In this study, we investigated the influence of environmental factors on risk for psychopathology in adopted and nonadopted adolescents of depressed parents.
Most empirically supported models of risk in families of depressed parents include factors that are conceptualized as environmental variables, such as harsh parenting and family conflict (6
). Many of these family “environment” variables, however, are genetically influenced (7
) and share common genetic influences with psychopathology in adolescents (9
)). For example, family conflict is characteristic of families with depressed parents (11
) and is associated with adolescent depression (12
). Rice and colleagues (14
) found that family conflict has a stronger association with depression in youths who have an elevated genetic risk for depression. In other words, these findings suggest that associations between family conflict (an environmental risk variable) and psychopathology in youths are at least partially mediated by genetic influences. This raises questions about the extent to which depressed parents create a rearing environment that increases risk for psychopathology in their children.
Most research on the risk for psychopathology in children of depressed parents includes only biological offspring and thus cannot isolate the influence of environmental factors. Some of these studies clearly support environmental mediation. Kim-Cohen and colleagues (15
) showed that maternal depression occurring after but not before their child's birth had a dose-response relationship with child antisocial behavior, and Weissman and colleagues (16
) reported associations between remission in maternal depressive symptoms following pharmacological treatment and improvements in child diagnoses and symptoms. These studies, however, do not provide a direct test of environmental mediation that is unambiguously separate from genetic effects. For example, the possibility that the associations are explained by genetic influences on chronicity of maternal depression cannot be ruled out.
Adoption studies provide a method for studying the environmental effect, disentangled from genetic influences, of having a depressed parent. Parent-child adoption designs typically compare risk conferred by biological parents who were not involved in rearing their children and risk conferred by adoptive, rearing parents. The design of the Sibling Interaction and Behavior Study (SIBS), whose sample we used in this investigation, provides a unique opportunity to compare the effect on adolescents of depression in rearing biological parents and in rearing adoptive parents. One methodological concern in adoption designs is the possibility that adopted individuals are exposed to a more restrictive range of home environments than nonadopted individuals as a result of self-selection by adoptive parents and screening of parents during the adoption process (17
). Using the SIBS sample, McGue and colleagues (18
) showed that while there is evidence of range restriction in adoptive families on parental disinhibitory psychopathology and socioeconomic status, adoptive families did not differ significantly from nonadoptive families in parental depression and family functioning, which are key variables in our study. In addition, while we cannot rule out the possibility of selective placement, it seems unlikely that biological risk for psychopathology differs in the adolescents of depressed and nondepressed adoptive parents. Information about the mental health of birth parents was not available.
In this article, we report the first parent-child adoption study to investigate the environmental effect of parental major depression, including both maternal and paternal depression, on risk for DSM-IV-TR psychiatric disorders during adolescence. Adolescence is a developmental period when rates of depression and other disorders are rising (19
), and adolescents are exposed daily to environmental influences associated with parental depression. Studying the effects of maternal and paternal depression separately is important. While a large literature has documented the effect of maternal depression on child psychopathology (6
), until recently the effect of paternal depression had largely been ignored. The emerging research shows inconsistent support for the effect of paternal depression on psychopathology in adolescents. Brennan and colleagues (20
) found a paternal depression effect on externalizing disorders but not on major depression in 15-year-olds, but their study included only fathers who had substantial contact with the children, whether they were biological fathers or stepfathers. Klein and colleagues (1
) found that paternal depression did not predict depression in adolescents and young adults unless the offspring's depressive episodes were at least moderate in severity. Lieb and colleagues (4
) found a significant effect of paternal depression on offspring depression but not substance use disorder by late adolescence or young adulthood. However, their study had the significant limitation of using offspring reports for depression diagnoses for both offspring and fathers. These contradictory findings suggest that paternal depression is in need of further study.
We hypothesized that, consistent with psychosocial models of the transmission of risk in families with a depressed parent, environmental influences contribute to an increased risk in families with a depressed mother. In light of the inconsistent evidence for an effect of paternal depression, we made no a priori prediction about risk in families with a depressed father. To the extent that parental depression has an effect, we predicted that it would be associated with an elevated risk for psychopathology both in adolescents with depressed biological parents and in adolescents with depressed adoptive parents, thus providing support for an environmental liability of parental depression.