Familial tendency to foetal loss analysed with Bayesian graphical models by Gibbs sampling

This paper presents several models for investigating whether the HLA alloge notypes DR1/Br, DR3 and DR10 are genetic markers for a predisposition of ex periencing unexplained recurrent foetal losses. A total of 199 women from 1 13 families answered questionnaires concerning their pregnancies and 145 of these women were HLA typed. The analysis of the data is complicated as dep endencies between pregnancy outcomes are expected. The main purpose of the paper is to illustrate how such analyses can be performed using Bayesian gr aphical models and Gibbs sampling. The analyses are made using the programs BUGS and CODA. Markov chain Monte Carlo analyses within a Bayesian framewo rk have become easier with the introduction of these programs. However, exp erience shows that some caution is required so we recommend making some ini tial analyses using very simple models and perhaps approximative methods, f ollowed by a model development introducing increasing complexity. Copyright (C) 2000 John Wiley & Sons, Ltd.