The applied and theoretical development of statistical methodology is progressing in the areas of biological and epidemiological models, mixture models for describing age changes in distributions of biological markers of morbidity and mortality, multiple comparisons, survival analysis, and the design of experiments, each of which is applicable to longitudinal studies and other studies of aging. The research utilizes various types of statistical models - regression models for studying risk factors' association with outcomes observed in prospective studies, and mixed- effects models for longitudinal data which consider both within- and between-subject variation in analyzing the repeated measurements for all individuals in the study population. Other techniques used include Bayesian, maximum likelihood and numerical computing methods. The methodology created provides original contributions to experimental testing associated with longitudinal studies, simultaneous comparison of various specified experimental effects, epidemiological study of disease states, survival or failure analyses of longitudinal observations representing growth, physical and mental disability, and other biological and behavioral changes in humans and animals. A major emphasis of this research project is on the development of methods which yield cogent yet easily understood results when applied to data. Mixed-effects regression models have been completed for two different aging studies. One study of nursing home patients resulted in the first comprehensive publication of such models in the aging literature. The second study of longitudinal changes in blood pressure found that individuals with rates of change in systolic blood pressure greater and smaller than the average rates of change had an increased prevalence of coronary heart disease.
