Epidemiology and Applied Statistics

Students will learn about the concepts, principles, and methods in epidemiology. The participants will familiarize themselves with study designs, methodological problems, statistical analyses and presenting and interpreting results. The skills learned in this course are of importance when conducting research and analyzing data as well as when evaluating published research.

In the first week the student will become acquainted with the field of epidemiology and gain knowledge on research designs, measures of frequency and association, and how to formulate a good research question. During this week a research question is formulated (in groups of 3 students), for which data have to be analyzed during the course (a dataset will be provided). Further, a start is made with the applied statistics (starting at a very basic level with descriptive statistics, non-parametric and parametric tests, and correlations, and also more advanced techniques such as linear and logistic regression analysis). All statistical analyses will be performed during practicals using SPSS.

In the second week the student will learn the concepts of confounding and effect modification. Also causal inference theory, the use of Directec Acyclic Graphs (DAGs) and survival analysis are introduced.  All these techniques will be used to answer the reseacrh question. Students are given a personal grade based on the final  oral presentation of 10 minutes on their research question (introduction, analyses, results, conclusion, discussion), followed by a general discussion.

• Study population
• Study designs (cross-sectional/ longitudinal, descriptive/ experimental, case-control/ cohort/ intervention
• How to formulate a research question and to make it operational
• Measures of frequency: prevalence/ incidence
• Measures of effect: relative risk (RR), odds ratio (OR)
• Methodological problems: bias, confounding and effect modification
• Descriptive statistics (plots, distribution, group differences)
• Associations (correlation, linear and logistic regressions, survival analysis)
• How to deal with confounding in the analysis (stratification, interaction and effect modification)
• Deriving inferences (statistical significance, clinical relevance, causality)
• Presentation and interpretation of results

NB. Before the first lecture we ask the students to make a home-assignment which includes reading a paper (provided by us) and answering a few questions about this paper.