Covariance component models for multivariate binary traits in family data analysis
2008 (English)In: Statistics in Medicine, ISSN 0277-6715, Vol. 27, no 7, 1086-1105 p.Article in journal (Refereed) Published
For family studies, there is now an established analytical framework for binary-trait outcomes within the generalized linear mixed models (GLMMs). However, the corresponding analysis of multivariate binary-trait (MBT) outcomes is still limited. Certain diseases, such as schizophrenia and bipolar disorder, have similarities in epidemiological features, risk factor patterns and intermediate phenotypes. To have a better etiological understanding, it is important to investigate the common genetic and environmental factors driving the comorbidity of the diseases. In this paper, we develop a suitable GLMM for MBT outcomes from extended families, such as nuclear, paternal- and maternal-halfsib families. We motivate our problem with real questions from psychiatric epidemiology and demonstrate how different substantive issues of comorbidity between two diseases can be put into the analytical framework.
Place, publisher, year, edition, pages
2008. Vol. 27, no 7, 1086-1105 p.
comorbidity, familial aggregation, multivariate binary trait, variance component
Medical and Health Sciences
IdentifiersURN: urn:nbn:se:uu:diva-15919DOI: 10.1002/sim.2996ISI: 000254678900010OAI: oai:DiVA.org:uu-15919DiVA: diva2:43690