Statistical Methods in Public Health I

Catalog description

Biostatistical methods in the design and analysis of epidemiological (observational) studies. This is a course on biostatistical methods in the design and analysis of epidemiological studies. The course addresses design issues with respect to (1) basic epidemiological (observational) studies, such as case-control, cohort, and cross-sectional studies, and (2) more complex studies, such as nested case-control, case-cohort, and case-crossover designs. Next, the course develops basic statistical inference for risk measures according to the nature of the outcome variables (binary and ordinal, continuous, rate, time-to-event). Confounding and interaction issues are discussed, along with statistical methods for handling them, such as standardization, stratification, and matching. More advanced methods are described based on multiple regression models that are specific to the outcome variables, as well as mediation modeling and propensity scores. Finally, computer-intensive analyses are considered, such as bootstrapping, permutation tests, and multiple imputation for missing data.