Building Disease Prediction Models by Synthesis Analysis

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): An estimate of a person's risk for coronary heart disease (CHD) is important for many aspects of health promotion and clinical medicine. A risk prediction model on a disease outcome may be obtained through multivariate regression analysis of a longitudinal study. For example, the CHD prediction model derived from the Framingham Heart Study has been widely used and has been incorporated into the latest National Cholesterol Education Program Adult Treatment Plan III guidelines for the management of hypercholesterolemia. However, the study was started long before many currently known risk factors were suspected. Therefore, new risk factors such as serum albumin, homocysteine, C-reactive protein. A common practice of meta-analysis is combining the results of numerous studies on the effects of a risk factor on a disease outcome. If several of these composite relative risks are estimated from the medical literature for a specific disease, they cannot be combined in a multivariate risk model, as is often done in individual studies, because methods are not available to overcome the issues of risk factor colinearity and heterogeneity of the different cohorts. In this proposal, we propose new methods, called synthesis analysis, to combine different risk factors on a disease outcome from diverse published studies into multivariate models. If several composite relative risk models are available from the medical literature for a specific disease, they cannot be combined into a multivariate risk model using standard meth-analysis techniques. In this proposal, we propose new methods, called synthesis analysis, to combine different risk factors on a disease outcome from diverse published studies into multivariate models.