Course / Module Description

 

1.   Title:                                    Advanced Epidemiological Methods  

 

2.   Institution:                          Master Programme International Health

                                                   Institute of Tropical Medicine, Spandauer Damm 130, 14050 Berlin

                                                   Charité University-Medicine Berlin, Free and Humboldt University Berlin

                                                               

3.   Course coordinators:       Dankmar Böhning, James Gallagher (both University of Reading, Great Britain)

 

4.   Type:                                   Advanced optional

 

5.   Duration:                             2 weeks (first half of July)

 

6.   Credit points                      3.0 (90 hrs SIT)

 

7.   Language:                           English

 

8.   Objectives:                         At the end of the module, students will be able to understand the various effects of bias that might occur in
                                                   assessing the relationship between disease occurrence and a risk factor. This includes the forms of   
                                                   confounding (inflation and masking) as well as effect modification. Students will be able to model this
                                                   relationship in the light of several other factors. They will  be able to assess the effect of several factors on
                                                   a quantitative outcome (linear regression model) as well as on a binary outcome (logistic regression
                                                   model) and on a count number (Poisson regression model).  Students will understand the relationship of
                                                   standardization methods (direct and indirect standardization) and confounder adjustment by means of
                                                   regression analyses and will be able to select an appropriate regression model for epidemiological data.
                                                   Students will be able to perform multiple regression analyses using the statistical programme package
                                                   Stata.

 

9.  Content:                               Introduction to the statistical programme package Stata;                       

                                                   The model of dependency  in epidemiology;              

                                                   Epidemiological measures of exposure effects and their estimation including standard errors (in 
                                                   connection with study type);

                                                   Confounding (inflation and masking) and interaction (synergistic and antagonistic);

                                                   Direct and indirect standardization;

                                                   Mantel-Haenszel approaches for risk ratios, rate ratios and odds ratios;

                                                   Regression models for quantitative outcome variables;

                                                   Regression models for binary  outcome variables;

                                                   Regression models for count data;

                                                   Model selection and variable selection;

                                                   Outlook on further advanced regression methods (regression models for categorical outcome variables,                                                    regression models for survival time data, random effect models, latent variable models, threshold models,
                                                   inverse regression)

 

10. Learning Methods:            Morning sessions: Seminar, introduction to the subjects areas above;

                                                   Afternoon sessions: Project work, modelling exercises, tutorials with individual consultation.     

 

11. Assessment
      procedures:                       Students obtain a data file, a description of data, and a scientific question at about half time through the                                                    course. Students are asked to contribute to the answer to the question by means of statistical analyses                                                    including regression analysis. Students will have to select an appropriate model and to perform the                                                    analysis using the statistical programme package. Its results have to be described and discussed in a                                                    short 2-page report. A 1-hour open book examination using MC and essay questions is performed on the
                                                   last day of class.

 

12. Prerequisites:                    Successful completion of a basic course in epidemiology;

                                                   Elementary knowledge of biostatistics;

                                                   Interest in theory and practice of epidemiology.

 

13. Attendance:                        Max 15 students

 

14. Selection criteria:              Preferential admission of tropEd Masters students

 

15. Tuition Fees:                      EUR 580,00                     

 

16. Scholarships:                     None                    

 

back to list