Abstract
Thomson and Bodmer (1979) explored several conceptual and analytic issues involved in studying the association of HLA haplotypes with disease. One key problem emphasized was the assessment of the statistical significance of the 3rd order linkage disequilibrium between 2 HLA genes and a disease susceptibility gene. An approach to this problem was formulated and this approach was illustrated by deriving statistical tests of true haplotype associations for 2 models of disease susceptibility previously studied by Terasaki and Mickey (1975) and by Thomson and Bodmer (1977). A numerical example using actual phenotypic data was presented to demonstrate the practicality of the methods.