Week 3: Special Modules

Summer Course 2019
32nd Residential Summer Course in Epidemiology

1 July – 5 July 2019

Week 3, parallel morning module 1
Advanced statistical topics
Per Kragh Andersen with Corrado Lagazio and Michaela Baccini

  • Competing risks
  • Recurrent events and longitudinal data
  • Cohort sampling
  • Propensity score
  • Causal modelling

 

Week 3, parallel morning module 2
Causal methods in epidemiology:
Mendelian randomization and triangulation
Debbie Lawlor and Carolina Borges 
A major aim of epidemiology is to identify causes of disease and health related outcomes in populations. This is necessary to provide the evidence base for identifying prevention and treatment targets for which interventions can be developed and their effectiveness tested. This has often been undertaken by applying multivariable regression analyses (or similar methods) to observational data (e.g. cohort or case control studies). Results from these studies may give biased causal estimates because of confounding, reverse causality, selection bias or other sources of bias.

In recent years several novel (to epidemiology) methods have been developed to explore causality and the idea that ‘triangulating’ results from different methods can provide more robust causal understanding is gaining traction. The idea of triangulation is that if different methods that all have very different key sources of bias point to the same causal answer we have more confidence that is the correct answer than if we just had information from one of those methods.

In this course students will learn about cross-context comparisons, negative control studies, matched designs (specifically within sibship comparisons) and instrumental variable analyses (with a particular focus on Mendelian randomization in which genetic instruments are used). They will also learn about triangulating findings from these different methods and more conventional multivariable regression approaches to improve causal understanding.

By the end of this module students should be able to:

  1. Understand the principles and assumptions, strengths and limitations of each of the following methods:
    1. Cross-context comparisons;
    2. Negative control studies;
    3. Matched within sibship analyses;
    4. Mendelian randomization
    5. Other instrumental variable methods.
  2. understand the concepts behind sensitivity analyses to explore violation of key assumptions of each of these methods
  3. Complete a (straight-forward) one-sample and two-sample MR analysis
  4. Understand how different methods might be triangulated to improve causal inference

The course will be taught with lectures, paper & pen and computer (using Stata) practicals.

There is no requirement to understand genetics (we will provide a brief background about genes to the level needed for their use in MR), but students should have a good grounding in the principles of epidemiology, including a clear understanding of confounding and the assumptions of multivariable regression analyses.

 

 Week 3, parallel morning module 3
Environmental epidemiology
Josep M. Antó and Jordi Sunyer

  • Epidemiological design and applications
  • Exposure measurements : ecological and individual
  • Spatial and temporal clusters
  • Effects measurements. Biological markers

Risk assessment

 

Week 3, parallel morning module 4
From epidemiology to the burden of disease: putting risks in perspective
Nino Künzli and Thomas Fürst

Epidemiology is a core science to investigate and quantify the association between risk factors and health outcomes. However, public health professionals and policy makers need to understand the public health relevance of risks to plan and prioritize prevention and policy making. The epidemiology-based assessment of the risk related burden of disease provides the bridge between public health science and policy. This course will familiarize students with the use of epidemiology in quantitative risk assessment and the comparison across risks. Based on a range of examples, students will i) learn how epidemiology contributes to quantitative risk assessment, ii) understand the tools to assess the health burden and iii) critically interpret the derived outputs. Exercises based on the burden of disease data will train critical thinking for the comparison of different public health risks.

Approach: Lectures on concepts, self-studies with on-line exercises and group discussions will foster the understanding of how epidemiology is used in risk assessment. Students are expected to bring their personal note books to independently work on line, on the web.

 

Week 3, parallel afternoon module 1
Clinical Epidemiology: the evaluation of medical tests
Patrick M Bossuyt and Miranda Langendam

In this one-week course, we focus on three topics:

Diagnostic testing: Diagnostic accuracy, bias and study design in diagnostic testing.

Prognostic modeling: The concept of a prognostic model, building and testing the validity of a model, interpreting data from prognostic models.

Intervention study design: Introduction to RCTs and innovative design, comparison between RCT and observational studies, confounding by indication.

Approach: A combination of lectures and personal assignments, to be completed in pairs and discussed in the group.

 

Week 3, parallel afternoon module 2
Advanced topics in epidemiology
Irene Petersen and Jan Vandenbroucke

The origins and usefulness of several advance study desings and methods of analyses will be studied – each time ending with a current positioning: what is state-of-the art and what are the potential applications and pitfalls (with practical examples)

  • Instrumental variable analysis and regression discontinuity designs/analysis
  • Missing Data and Multiple Imputation part I
  • Missing Data and Multiple Imputation part II

 

Week 3, parallel afternoon module 3
Principles of prevention in the precision medicine and Big data era
Rodolfo Saracci

This module presents to researchers, health professionals and clinicians particularly interested in prevention a perspective critically examining whether the population and the individualized approaches, as a classically outlined by Geoffrey Rose in the 1980s, still represent useful concepts and operational principles or whether the availability of massive health data on each person makes them obsolete leading to a unified ‘precision prevention’ approach. Relevant methodological aspects will be reviewed, involving an introductory presentation of causal versus predictive models and of machine learning instruments. Specific ethical issues that prevention research and measures raise will be sketched for discussion.

Approach: lectures, reading of papers with critical discussion and ‘pros and cons’ arguments.

 

Week 3, parallel afternoon module 4
Infectious disease epidemiology
Tyra Grove Krause and Steen Ethelberg

Vaccines, antibiotics and hygiene measures have played an important role in the fight against infectious diseases. However, worldwide inequalities in accessing health care including treatments and vaccines, re-emergence of vaccine preventable diseases, and the threat of antimicrobial resistance as well as the risk of emerging new pathogens underline the fact that infectious diseases remain a global public health challenge.

This course will introduce the epidemiological fields of transmissibility, vaccinology, disease surveillance and outbreak investigations. By the end of this module, the student should be able to understand:

  1. The terminology and definitions used in infectious disease epidemiology
  2. Principles of disease transmission including mathematical models for epidemics
  3. Basic concepts of vaccinology
  4. Principles of infectious disease surveillance and interpretation of surveillance data
  5. The 10 steps of an outbreak investigation
  6. The use and interaction of microbiological and epidemiological methods in outbreak detection and control.
  7. The use of epidemiological study designs in infectious disease epidemiology

The course will use a mix of lectures and small group case studies.

Prior knowledge of infectious diseases is not needed but students should have a good understanding of principles of epidemiology, including a basic knowledge of measures of frequency and associations and epidemiological study designs.