Advanced methods in biostatistics and epidemiology

Course content

** Learning activity 1: N. Speybroeck
Module 1: The use of routine data for the generation of epidemiological information;
Module 2: Review of the basic concepts in epidemiology;
Module 3: Bias Control (Bias: revision; Control of confounding (random sampling, pairing, standardization,…); Adjustment by a regression model: example: logistic regression;
Module 4: Analyzing and understanding incidence rates (Logistic and Poisson regression);
Module 5: Simulation Modeling in epidemiology;
Module 6: Study of some advanced epidemiological approaches and illustrations (Space-time models, Classification and regression Trees; Decomposing the inequalities of health.).

** Learning activity 2: F. Kirakoya
Part 1: Introduction, why perform a meta-analysis? How a meta-analysis works?
Part 2: Effect size, Effect size based on binary data, effect size based on means;
Part 3: Fixed effect versus random-effects models;
Part 4: Heterogeneity;
Part 5: Publication biais.

Objectives (and/or specific learning outcomes)

** Learning activity 1: N. Speybroeck The course consists of six modules formulated in teaching objectives. Each module tackles one or several key competences in professional situations.
The teaching objectives for each module are:
Module 1: Revise basic concepts in epidemiology;
Module 2: Understand how to control bias (Bias: revision; control of confounding (random sampling, pairing, standardization,…); adjustment by a regression model: example: logistic regression);
Module 3: Learn how analyzing and understanding incidence rates (Logistic and Poisson regressions);
Module 4: Learn how to build simulation models;
Module 5 (if time allows): Introduce students to advanced epidemiological approaches and illustrations (Space-time models, Classification and regression Trees; Decomposing the inequalities of health.).

** Learning activity 2: F. Kirakoya to provide an introduction to meta-analysis and to discuss the rationale for this type of research and other general considerations. Methods used to perform a meta-analysis will be highlighted and some aspects of presentation and interpretation of meta-analysis will be discussed.

Teaching methods and learning activities

**Learning activity 1: N. Speybroeck The lectures will be illustrated by concrete cases extracted from literature. Sessions of exercises will go along with the lectures. The exercises will be conducted in small groups, worked out by the students and discussed together in class. The exercises are simple applications (related to the knowledge acquired in the theoretical part), or exercises combining several principles (related to the teaching objectives) which will allow the use of a diversity of skills and which will be the object of group works at specific times (the methodology will be explained during the course). Software: R R is an interactive programming language containing a very large collection of statistical methods and important graphic facilities. It is a free clone of the S-Plus software marketed by MathSoft and developed by Statistical Sciences around the language S. The internet site of the "R core-development TEAM", http://www.r-project.org, is the best source of information on the software R.

** Learning activity 2: F. Kirakoya - Course notes with examples drawn from the health sciences literature. - Homework assignments. - Practical exercises with real world data. - A data analysis project performed with statistical software STATA.

Contribution to the teaching profile

Contribution to: * SKILL 1. Applying a corpus of pluridisciplinary methodological knowledge to the analysis of various public health issues , specifically « Analyse data gathered using appropriate methods » and « Evaluate the quality and limits of the methods used to gather, save, analyse, and share research data ». * SKILL 3. Organizing individual and collective work in collaboration with various partners, more specifically « Plan work in order to achieve results within the intended timeframe » and « Work as part of a team, with shared tasks and group dynamics ».

Other information

Contacts

Niko Speybroeck: Niko.Speybroeck@uclouvain.be

Fati Kirakoya: fati.kirakoya@ulb.ac.be

Campus

Erasme, Other campus

Evaluation

Method(s) of evaluation

Other

Other

Learning activity 1: N. Speybroeck Closed book (theory) & open book exam (practical exercise) and excercises during the teaching sessions.
Learning activity 2: F. Kirakoya data analysis project + oral presentation and defense.

Mark calculation method (including weighting of intermediary marks)

Learning activity 1: 70% (40% exam + 30% data analysis project).
Learning activity 2: 30% (15% data analysis project + 15% oral presentation).

The Unit is validated when the weighted average of all learning activities is >= 10/20.
When the Unit is not validated at the 1st session, the 2d session will include an examination regarding only the failed activity part(s) at the first session (<10/20). Marks equal or higher than 10/20 at the first session will automatically be reported at the 2d session.

Language(s) of evaluation

english