Biostatistics II: Logistic regression for epidemiologists

Learning outcomes

After successfully completing this course you as a student are expected to be able to:

• choose a suitable regression model for assessing a specific research hypothesis using data collected from an epidemiological study, fit the model using standard statistical software, evaluate the fit of the model, and interpret the results.
• explain the concept of confounding in epidemiological studies and demonstrate how to control/adjust for confounding using statistical models.
• apply and interpret appropriate statistical models for studying effect modification.
• critically evaluate the methodological aspects (design and analysis) of a scientific article reporting an epidemiological study.

Learning outcomes are classified according to Bloom's taxonomy: knowledge, comprehension, application, analysis, synthesis, and evaluation.

Contents of the course

This course focuses on the application of linear and logistic regression in the analysis of epidemiological studies. Topics covered include a brief introduction to continuous and binary outcome data, univariable and multivariable models, interpretation of parameters for continuous and categorical predictors, flexible modelling of quantitative predictors, confounding and interaction, model fitting and model diagnostics.

Course director and contact person