Project Summary Our proposed project will build upon previous research which has shown an association between oral health and systemic health, including the relationship between heart disease and periodontitis. We will benefit from recent methodological advances in computational genomics and be the first study to provide causal evidence about the postulated relationship between poor oral health and heart disease using nationally-representative, longitudinal data. Specifically, we will construct polygenic scores using a GWAS-by-subtraction genomic structural equation model that are suitable for use in instrumental variables estimation. To accomplish our Aims, we will use longitudinal files containing 11 years of data (2006-2016) from the Health and Retirement Study (HRS). This longitudinal database will provide over 22,000 older Americans producing a sufficiently large sample size to overcome prior analytical obstacles. Aim 1. Determine the strength of genetic risk factors for poor oral health. The working hypothesis is that individual genetic composition is a strong predictor of poor oral health. Aim 2. Determine the causal effect of poor oral health on the probability of heart disease (HD). The working hypothesis is that variation in oral health due to individual genetic composition can be used to identify the causal effect of poor oral health on HD.