Lecturer:

Matthew Fox, Professor in the Department of Global Health and Epidemiology at Boston University School of Public Health

Learning objectives:

By the end of this week, participants should be able to:

• Use the sufficient cause model, counterfactual susceptibility type model, and a causal graph to assist with the design or analysis of an epidemiologic study.

• Calculate adjusted measures of effect and select those that, when collapsible, correspond to the no-confounding condition. Use the adjusted measures of effect to estimate the direction and magnitude of confounding.

• Distinguish effect measure modification, interdependence, and statistical interaction from one another as separate - but related - concepts of interaction.

• Identify the likely magnitude and direction of bias due to misclassification of exposure, outcomes, confounders and modifiers. Weigh the advantages and disadvantages of significance testing.

• Compare the advantages and disadvantages of frequentist and Bayesian approaches to analysis of a single study, to eveidence, and to changing your mind.

Prerequisites:

• Basic knowledge of epidemiology and biostatistics

Fees:

• 750 €

• 510 € for enrolled students (proof required)
  
  

3 ECTS

Contact: tanja.tegude@charite.de