The National Health and Nutrition Examination Survey (NHANES) I and its Follow-up form an integral part of the NIA EDB scientific data base. More than 100 different analyses are planned by EDB and its collaborators. To make appropriate use of these data in regression analyses, adjustments must be made in usual regression techniques because the data from these complex samples do not meet the assumptions required by ordinary regression analysis. The purpose of the project is to study and develop statistical methodology for multiple linear regression models that can be appropriately applied to the NHANES. The NIA entered into an agreement with NICHD for the purpose of expanding the scope of the contract with the Research Triangle Institute to develop statistical methodology for multiple linear regression models that can be appropriately applied to data from the NHANES and ts Follow-up Survey. The contractors have completed work on the development of the methods, and we will soon illustrate them with a reanalysis of an NHANES data set relating baseline bone density measures and vitamin D intake to hearing loss in the NHANES respondents.
