Statistical Methods and Challenges in Epidemiology and Biomedical Research

摘要: Abstract This chapter provides an introduction to the role, and use, of statistics in epidemiology biomedical research. The presentation focuses on assessment understanding health-related associations a study cohort. principal context considered is estimation risk health events relation individual subject characteristics, exposures, or treatments, generically referred as ‘covariates’. Descriptive models that focus relative absolute risks preceding covariate histories will be described, along with potential sources bias testing. design, conduct randomized controlled trials also described this prevention research context, well therapeutic Some aspects initial evaluation ideas concepts for preventive interventions discussed. leads naturally discussion role biomarkers research, such purposes exposure assessment, early disease diagnosis, interventions. Recently available biomarkers, including high-dimensional genomic proteomic markers, have add much knowledge about processes specificity intervention development evaluation. These data are attended by many interesting statistical design analysis challenges. A brief ongoing analytic explanatory analyses Women's Health Initiative concludes presentation.