Using family based designs to explore causal mechanisms in conduct disorder

Abstract

Conduct Disorder (CD) is one of the most common mental disorders seen in children and adolescents, with conduct problems accounting for 30-40% of referrals to Child and Adolescent Mental Health Services (CAMHS). The majority of studies in this area have adopted case-control correlational designs, demonstrating associations rather than causal links between putative neuropsychological or neuroanatomical risk factors and CD. Family based designs allow us to move closer to inferring the causal role of such mediating risk factors in the developmentof CD, through common genetic and environmental originating factors. One method, which has not yet been applied to CD research, is to study the biological relatives of those with CD, who do not show the disorder themselves, but may still carry risk markers for the condition. Identifying shared neuropsychological or neuroanatomical deficits in both affected and unaffected relatives, would suggest shared genetic and environmental factors may mediate and increase the risk for developing CD.

We therefore adopted a family design to test for the co-segregation of CD and its putative neuropsychological and neuroanatomical risk factors. There were three research groups: adolescents with CD (n=45), unaffected relatives of people with CD (n=24) and typically developing controls with no family history of CD (n=43). The project used a range of techniques, such as semi-structured psychiatric interviews, questionnaire measures, neuropsychological tasks (emotion recognition, fear conditioning, risky decision-making and economic decision-making) and Magnetic Resonance Imaging (MRI) methods, to test whether CD and its putative cognitive and neuroanatomical deficits co-segregate within families.

Clinical and personality characteristics that are thought to increase risk for CD, such as callous-unemotional traits, drug use and impulsivity, did not show a familial pattern but were only elevated in affected probands. The familial pattern of effects varied across neuropsychological domains. Unaffected relatives demonstrated similar impairments to CD probands (albeit to a lesser degree), in emotion recognition and fear conditioning, but there was no evidence of familial effectson decision-making, either under risk or in an economic exchange task. There was mixed evidence for co-segregation of CD and neuroanatomical changes, with CD probands showing reductions in grey matter volume in the cerebellum, inferior frontal gyrus and ventromedial prefrontal cortex and unaffected relatives showing trends towards reductions in these regions. Interestingly, CD probands showed reduced grey matter volume in the anterior insula compared to unaffected relatives.

The current study was the first to directly compare adolescents with CD, unaffected relatives of people with CD and healthy controls. The findings suggest that neuropsychological deficits in emotion recognition and fear conditioning may increase the individual’s risk for developing CD in a probabilistic way and may help to inform future longitudinal studies of the development of CD and interventions for this condition.

More information

Identifiers

ORCID for Graeme Fairchild: orcid.org/0000-0001-7814-9938

Catalogue record

Export record