Short Courses

WCE 2021 Short course details are available now.

The Short courses occur pre-congress and post-congress. Further details on registration pricing can be found here.

Pre congress short course 1

Causal Mediation Analysis

Professor Tyler VanderWeele, Harvard School of Public Health

Mediation analysis concerns assessing the mechanisms and pathways by which causal effects operate. We introduce the fundamentals of causal mediation analysis, and describe the relationship between traditional methods for mediation in the biomedical and the social sciences and new methods in causal inference. For dichotomous, continuous, and time-to-event outcomes, discussion will be given as to when the standard approaches to mediation analysis are or are not valid, and how to extend these to broader settings. The course will present SAS, SPSS, Stata, and R macros to implement these techniques. The course will also cover sensitivity analysis techniques to assess how sensitive the conclusions are to violations of the no-confounding assumptions.

Date and time:
Friday 3 September 10:00-13:30 NZST (New Zealand Standard Time)
Friday 3 September 08:00-11:30 AEST (Australian Eastern Standard Time)
Friday 3 September 07:00-10:30 JST (Japan Standard Time)
Friday 3 September 00:00-03:30 CEST (Central European Summer Time)
Thursday 2 September 23:00-02:30 BST (British Summer Time)
Thursday 2 September 19:00-22:30 ART (Argentina Time)
Thursday 2 September 18:00-21:30 EDT (Eastern Time, USA)
Thursday 2 September 15:00-18:30 PDT (Pacific Time, USA)

Max number of participants: unlimited

Pre congress short course 2

Causal modelling: How to analyse a data set with a large number of variables

Professor Neil Pearce, Department of Medical Statistics, London School of Hygiene and Tropical Medicine

This is a half-day course which will cover: (i) causation, variation and statistical modelling; (ii) the difference between predictive and causal modelling; (iii) use and abuse of DAGs in causal modelling; (iv) minimising bias and multicollinearity; and (v) applying these concepts to analyse a data set with a large number of variables.
This is a lecture/discussion-based course, and it is not planned to have formal data analysis practical sessions. However, participants are welcome to bring examples of their work and current methodological problems for discussion, and where appropriate, these discussions could involve analyses of their data sets using Stata.Readings:
Greenland S, Pearce N. Statistical foundations for model-based adjustments. Annual Review of Public Health 2015; 36: 89-108.
Greenland S, Daniel R, Pearce N. Outcome modelling strategies in epidemiology: traditional methods and basic alternatives. International Journal of Epidemiology 2016; 45: 565-575.
Pearce N. Epidemiology in a changing world: variation, causation and ubiquitous risk factors. International Journal of Epidemiology 2011; 40: 503-512.

Date and time:
Friday 3 September 19:00-22:30 NZST (New Zealand Standard Time)
Friday 3 September 17:00-20:30 AEST (Australian Eastern Standard Time)
Friday 3 September 16:00-19:30 JST (Japan Standard Time)
Friday 3 September 09:00-12:30 CEST (Central European Summer Time)
Friday 3 September 08:00-11:30 BST (British Summer Time)
Friday 3 September 04:00-07:30 ART (Argentina Time)
Friday 3 September 03:00-06:30 EDT (Eastern Time, USA)
Friday 3 September 00:00-03:30 PDT (Pacific Time, USA)

Number of participants: 50

Pre congress short course 3

Introduction to multi-level mixed regression models with applications to linear regression in Stata

Dr Arul Earnest, Associate Professor, Biostatistics Unit, Deputy Head, Reporting and Research Clinical Outcomes data Reporting and Research Program (CORRP), Department of Epidemiology & Preventive Medicine, School of Public Health & Preventive Medicine, Monash University, Australia

This half-day workshop is designed specifically for those interested in applying multi-level mixed regression models to their datasets. A brief comparison between the various analytical approaches to hierarchical (multi-level) data will be undertaken, including Generalised Estimating Equations, Generalized Linear Latent and Mixed Models, followed by a demonstration of application of the linear mixed model to health data using Stata’s ‘xtmixed’ command. Annotated output will be presented, along with relevant published papers. Familiarity with Stata will be useful for this workshop.

Number of participants: 50

Pre congress short course 4

Introduction to Mendelian randomization

Dr Alice Carter and Dr Kaitlin Wade, MRC Integrative Epidemiology Unit, The University of Bristol

Mendelian randomization is a study design that uses human genetic variants as instrumental variables to test the causal effect of a (non-genetic) risk factor on a disease or health-related outcome. Since its first proposal in 2003, it has been increasingly used to determine population causal effects using observational epidemiological data. This course aims to introduce the framework, assumptions, strengths and limitations of Mendelian randomization. Students will learn about one-sample and two-sample Mendelian randomization, including gaining practical experience of how to apply these methods to real data provided. They will also learn about a range of sensitivity analyses that explore likely violations of the assumptions of Mendelian randomization. Prior experience of using Mendelian randomization is not required, but participants should have an understanding of basic genetics and aetiological epidemiological principles.
The workshop will take a blended approach, whereby some pre-course material would be provided to participants to read / listen to in their own time before the workshop so that live sessions are more interactive. There is a choice of two times for the live sessions to better accommodate different time-zones – please choose one session only.

SESSION A:
Friday 3 September 19:00-22:30 NZST (New Zealand Standard Time)
Friday 3 September 17:00-20:30 AEST (Australian Eastern Standard Time)
Friday 3 September 16:00-19:30 JST (Japan Standard Time)
Friday 3 September 09:00-12:30 CEST (Central European Summer Time)
Friday 3 September 08:00-11:30 BST (British Summer Time)
Friday 3 September 04:00-07:30 ART (Argentina Time)
Friday 3 September 03:00-06:30 EDT (Eastern Time, USA)
Friday 3 September 00:00-03:30 PDT (Pacific Time, USA)SESSION B:
Saturday 4 September 01:00-04:30 NZST (New Zealand Standard Time)
Friday 3 September 23:00-02:30 AEST (Australian Eastern Standard Time)
Friday 3 September 22:00-01:30 JST (Japan Standard Time)
Friday 3 September 15:00-18:30 CEST (Central European Summer Time)
Friday 3 September 14:00-17:30 BST (British Summer Time)
Friday 3 September 10:00-13:30 ART (Argentina Time)
Friday 3 September 09:00-12:30 EDT (Eastern Time, USA)
Friday 3 September 06:00-09:30 PDT (Pacific Time, USA)

Number of participants: 30 max in each session

Pre congress short course 5

Risk estimation – new developments

Professor Leigh Blizzard, Menzies Institute for Medical Research, Hobart, Australia

The frequency with which a binary event occurs, relative to the number at risk of the event, is a fundamental measure of occurrence. In epidemiology, it is referred to as risk in prospective studies including clinical trials, and as prevalence in cross-sectional studies. Ratios or differences of these quantities enable easily understood comparisons of occurrence between groups of subjects.

The binomial regression model for binary outcomes provides a mathematical model of the event probability, the theoretical counterpart of relative frequency. Depending on whether the model is estimated with a logarithmic or identity link to the mean of the binary outcome variable, the coefficients of covariates provide estimates of risk ratios or risk differences.

Numerical instability can arise when fitting these models using standard statistical software such as R, SAS, SPSS or Stata. This instability manifests in non-convergence of the fitting algorithm, or convergence to a solution with fitted probabilities outside the (0,1) range.

Fortunately, a complete remedy for all such estimation difficulties is now available. New developments provide a fail-safe method of estimating risk ratios and differences. These advances should engender complete confidence in the use of binomial regression as a primary method of analysis.

This workshop provides a thorough tutorial on the rationale for estimation of risk ratios and risk differences, and their range of applications including outreach to fields in which absolute risk is a common metric. Attention will be given to assessment of statistical interaction on the multiplicative and additive scales. The improved methodology will be demonstrated using R and Stata software. Participants with their own laptops will be able to experience the ease with which the software can be applied to data encountered in common situations.

Friday 3 September 10:00-13:30 NZST (New Zealand Standard Time)
Friday 3 September 08:00-11:30 AEST (Australian Eastern Standard Time)
Friday 3 September 07:00-10:30 JST (Japan Standard Time)
Friday 3 September 00:00-03:30 CEST (Central European Summer Time)
Thursday 2 September 23:00-02:30 BST (British Summer Time)
Thursday 2 September 19:00-22:30 ART (Argentina Time)
Thursday 2 September 18:00-21:30 EDT (Eastern Time, USA)
Thursday 2 September 15:00-18:30 PDT (Pacific Time, USA)

Number of participants: unlimited

Pre congress short course 6

Best Practices in Publishing Public Health, Clinical, and Biomedical Research

Eduardo L. Franco, MPH, DrPH, PhD (Hon), O.C., FRSC, FCAHS;
James McGill Professor, Departments of Oncology and of Epidemiology; Chairman, Department of Oncology, and Director, Division of Cancer Epidemiology, McGill University, Montreal, Canada; Editor-in-Chief, Preventive Medicine and Preventive Medicine Reports; Senior Editor, eLife.

Content: (i) Principles of good scientific practice as applied to research on human subjects; (ii) the value of peer review as the cornerstone of scientific research and knowledge translation; (iii) reading and writing well: using logic and scientific reasoning in research communication; (iv) understanding the business of scholarly publishing; (v) dealing with editors and reviewers; (vi) rules of authorship, strategies to avoid conflict; (vii) selecting journals, understanding metrics of scientific value and impact; (viii) the threat from predatory publishing, vanity presses, and for-profit conferences; (ix) the shifting paradigm of scholarly publishing: the open access revolution, the advent of sound science journals, competing models for peer review, preprint archiving.
The format will be interactive lecturing to take advantage of students’ own experiences as authors and reviewers. There are no prerequisite courses or degrees other than some experience in biomedical or clinical research. This course is intended for early-career researchers and clinician-scientists in any domain of health or life sciences. Those in more advanced career stages may also benefit from the course by contributing to the class discussions their own experiences as authors. Science journalists, public health practitioners and policymakers will also find the contents valuable in understanding the process of scientific research and publishing. Researchers whose mother tongue is not English will find the course helpful in assisting them to navigate the process of preparing, submitting, and revising manuscripts for publication in international biomedical journals.

Number of participants: unlimited

Pre congress short course 7

Mathematical modelling of real-world infectious disease epidemics – an R based hands-on short course

Ashok Krishnamurthy PhD, Associate Professor, Department of Mathematics and Computing at Mount Royal University, Calgary, Alberta, Canada

In this short course I will describe and illustrate participants an understanding of infectious disease models and their value for public health. Mathematical modelling of infectious diseases is an interdisciplinary area of increasing interest. The short course will be based on our real-world experience of tracking the spatial spread of

• Measles in pre-vaccine England and Wales (1944 – 1966),
• Ebola in the Democratic Republic of Congo (2018 – 2020), and
• SARS-CoV-2 in Nigeria and the Czech Republic (2020 – to date).
This interactive short course will be delivered using real-world data and practical simulation exercises using the free, open-source software R. No prior detailed knowledge of modelling infectious diseases or epidemiology is required.

Objectives: By the end of the short course participants will be able to:
• know how to build a compartment model of epidemiology (for ex: SIR, SEIR, SEIRD, SVEIRD etc.) to track the spatial spread of an infectious disease outbreak.
• considerations for tackling global public health issues through data linkage.
• Practical hands on application of a realistic scenario involving tracking the COVID-19/Ebola in a particular country/region.

Target Audience: This short course will be designed for early-career data scientists, epidemiologists, biostatisticians and graduate students. The workshop is open to all WCE 2021 delegates.

Number of participants: 60 max

#WCEpi2021