A Structural Approach to the Familial Coaggregation of Disorders

• DOI: 10.1097/ede.0b013e31816a9de7
• Publication Date: 2009-03-03T22:40:07Z
• Authors: James I. Hudson; Kristin N. Javaras; Nan M. Laird; Tyler J. VanderWeele; Harrison G. Pope; Miguel A. Hernán
• Keywords: Association (psychology); Causal model

The demonstration that 2 disorders coaggregate in families is often the first step exploration of genetic factors common to disorders. Previous methods analyzing familial coaggregation have used either (1) a typical measure association (eg, odds ratio) between disorder an individual and another family members, or (2) linear structural equation model (SEM). method accommodates case-control sampling families, but may not assess causal effect interest because it based on underlying model. SEM model, cannot easily accommodate direct effects 1 other within individuals. We develop new directed acyclic graphs. Because this generalization models uses measures effects, combines strengths both previous methods. In absence disorders, our approach provides valid estimate effect. presence upper-bound (assuming additive latent nonfamilial factors) lower-bound For illustration, we applied study binge eating bipolar disorder.