A sib-pair strategy for the use of restriction fragment length polymorphisms to study the mode of transmission of type II diabetes.

Three models for explaining the joint distribution of DNA insertional elements and type II diabetes in sibships are given as potential candidates for resolving the mode of transmission for this common non-Mendelian disorder. While the distributions predicted from all models are subject to the same rigid marginal constraints, only two models can be considered genetic (a combined major locus multifactorial background model and a two-locus model). A sequential probability ratio test is proposed to distinguish between these two models. Once the probands have been ascertained and using realistic parameter estimates, it is shown that, on the average, fewer than 100 affected sib-pairs are required to reach a decision at the 1% significance level with 99% power.