Dental carries or tooth decay is one of the most prevalent chronic diseases in the world. People remain susceptible to it throughout their lifetime, and so it is important that we understand all the factors and covariates that could affect dental health, so that high-risk groups can be properly targeted. The National Health and Nutrition Examination Survey (NHANES) is a program of studies that assesses the health and nutritional status of adults and children in the United States. In home interviews and physical examinations are conducted for survey participants, and each person is assigned a sample weight based on the probability of selection. I will present Bayesian models for analyzing dental caries in the NHANES data, particularly focusing on the presence/absence of teeth and health/non-health of specific teeth surfaces. The Potts and autologistic models that account for spatial correlations within and between teeth, missingness in the data, and small area estimation, which we have developed via advance Markov Chain Monte Carlo computing techniques, will be discussed.