Prediction Models That Include Genetic Data

Abstract

Can alternative modeling approaches that integrate genetic data help to improve the prediction of risk for common diseases? In this issue of Circulation: Cardiovascular Genetics , Stengard et al1 set out to answer this question with regard to several genetic variations of the APOE gene and risk for ischemic heart disease (IHD). In their study, they included 3686 women and 2772 men with no medical history of IHD from the Copenhagen City Heart Study and sought to correlate IHD events to risk factors such as abnormal lipid levels, hypertension, diabetes, smoking history, and various APOE genotype data during a mean of 6.5 years of follow-up. Article see p 22 The traditional statistical approach to this analysis would be to use Cox proportional hazard modeling to map the hazard of developing IHD to a linear combination of significant risk factors. Stengard et al adopted an interesting alternative procedure that is consistent with the intuition that risk factors may have different effects in subjects with different unmeasured exposures (for example different genetic backgrounds and/or other unknown variables). Therefore, rather than looking for the risk factors that have a homogeneous effect on the hazard for IHD, the authors used the rule induction algorithm PRIM to discover subgroups of subjects with varying combinations of phenotypic risk factors for IHD and APOE alleles e2, e3, and e4. They then used the PRIM again to further segregate these subgroups according to additional genotypes in …