Week 1: General Modules

• Population health, the core of epidemiology
• Measures of occurrence and association
• Exposure and outcome measurements in epidemiology
• Study designs: Cohort studies, case-control studies, cross sectional studies and intervention studies
• Measures of association and attributable risk
• Bias and confounding
• DAGs (Directed Acyclic Graphs)
• Introduction to the bladder cancer dataset
• Temporal trends and geographical variations

The course comprises a combination of lectures and computer practical sessions. The statistical package Stata is used throughout. The course begins with sessions which review key concepts in the frequentist approach to statistical inference: confidence intervals and the use of statistical tests and interpretation of p-values. This is followed by two sessions on likelihood theory to explain its underlying key ideas and the link to the confidence intervals and p-values of the frequentist approach. These more theoretical sessions then lead to applications of the theory to the analysis of an epidemiological dataset, focussing on the estimation of risks, odds and rates as well as introducing survival analysis. There is also a session on the use of stratification to investigate confounding and effect modification.