Inter-University Postgraduate Programme in Public Health of the Swiss-German Universities

Advanced Methods in Epidemiology: History and Concepts

Geneva 9-11 January 2008

Within the framework of Inter-University Postgraduate Programme in Public Health of the Swiss-German Universities (www.public-health-edu.ch) and under the heading ”Advanced Methods in Epidemiology” a series of short courses were introduced in 2003 to complement the existing epidemiology programme. The focus of these courses, which generally run over three days, are on skills relevant to study design, critical appraisal and data analysis in Stata™. Courses will not run every year, but over several years this series should cover the most important recent developments in epidemiological methodology.

The Geneva Foundation for Medical Education and Research is organising the Course in Advanced Methods in Epidemiology: History and Concepts in January 2008.

Background

In most schools or programs of public health, epidemiologic concepts and methods are taught in two or three courses, each course presenting an increasing degree of complexity. Even though the course topics look the same (cohort and case-control designs, confounding, interaction, bias and causal inference), their content is different. In the basic courses, students are familiarized with the existence of these concepts and methods and taught simple definitions that will allow them to recognize them in their research and understand some of the issues related to them. The courses ”Biostatistics I” and ”Methoden der Epidemiologie” are prerequisites. In this advanced course, the same concepts and methods are revisited but with more extensive exploration of their properties. The students at this level are prepared to adapt the methods to specific research questions and to eventually make them evolve when encountering new types of problems.

Goals

In this advanced course, the epidemiological concepts and methods are revisited with more extensive exploration of their properties. The students at this level will learn to apply the methods to specific research questions.
Students will:

learn how to use the epidemiological methods and concepts to design studies and prepare study protocols
understand the historical origin of current epidemiological methods and concepts
be exposed to current developments in epidemiologic methods and concepts

Content

Students will learn to understand and to apply the design of prospective and retrospective studies, and the concepts of bias, confounding interaction and casual inference. The course will not deal with statistical methods or data analysis.
History of epidemiologic ideas: All lectures are organized to show the historical evolution of each specific method or concept, but the final lecture is an overview of the four stages that characterize the history of epidemiologic methods and concepts, that is, preformal, early, classic and modern epidemiology. The material is based on the following text: Morabia A (ed). History of epidemiologic methods and concepts. Birkhäuser 2004.
Cohort / case-control studies: Students will learn why cohort studies are the natural design for all types of epidemiological investigations and why case-control studies need always to be conceived as nested in hypothetical cohorts. Design of case-control studies in conditions where the study base is not well defined will be discussed. The consequences of case-based, case-concurrent and traditional schemes of control selection on the interpretation of the odds ratio will also be explained.
Confounding: These lectures will present the traditional way to assess the presence of confounding as a mixing of effects of multiple causes under weak or strong conditions, in unmatched and matched observational studies and in randomized controlled trials. It will also cover the newer conceptual developments viewing confounding as an issue of non comparability of groups.
Interaction: These lectures will present how to assess and interpret the presence of additive or multiplicative interactions, using three different approaches.
Bias: A theory of bias based on their mechanism will be presented, differentiating biases related to loss of accuracy (e.g., misclassification bias) from biases related to unbalanced identification of the compared groups (e.g., selection bias).
Causal inference: This lecture will show how the roots of causal inference in epidemiology go back to the 17th century (David Hume) and 19th century (S. Mill) philosophers and discuss why it can be so.

Methods

Lectures and discussion of specific problems.

Target audience, admission

Advanced course for students of the Public Health programme
Students who have already had a basic training in epidemiology
The courses «Biostatistik I» (2020) and «Methoden der Epidemiologie» (3060) are required

Dates: 9-11 January 2008

ECTS Credits: 1.5 ECTS-Credits - in field 1

Exams: Essay at home

Preparation and postprocessing

No preparation, 10 hours postprocessing

Language: English

Location: Geneva

Fee: CHF 1'150 .- (course material included)

Coordination of the Course

Prof. Alfredo Morabia

Lecturer and tutor

Alfredo Morabia, MD, PhD, Professor of Epidemiology, Queens College, City University of New York

Organisation

The Geneva Foundation for Medical Education and Research, 5 Chemin Edouard-Tavan, 1206 Geneva; Telephone: +41.22.346.77.16; Fax: +41.22.346.78.34
secretariat@gfmer.ch

Registration

Koordinationsstelle des Interuniversitären Weiterbildungsprogramms Public Health
mph@ifspm.uzh.ch

Deadline for registration: 9 November 2007

Additional information

A site visit of the WHO in Geneva is a highlight of the course

The Course will only take place if there is a minimum of 16 participants.

Print this page

Edited by Aldo Campana,