WORKSHOPS - WEDNESDAY 18 OCTOBER
Workshop Registration Fees
Workshop Registration is for Face to Face delegates only and is by purchase only.
Workshop Registration Inclusions:-
Admission to Workshop Session on Wednesday 18 October OR Friday 20 October ONLY;
Workshop Registration does not provide access to the AEA Conference 19 - 20 October;
Catering will be provided for Workshop Registration on Wednesday 18 October. Full Day workshop - morning tea, lunch and afternoon tea, 90-minute workshop - morning tea only; Half Day workshop lunch and afternoon tea; Friday Breakfast
AEA MEMBER - Half Day $95 / Full Day $185 / 90-min $65 / Workshop 2+3 $70 / FRI BREAKFAST $20
NON - MEMBER - Half Day $155 / Full Day $310 / 90-min $110 / Workshop 2+3 $280 / FRI BREAKFAST $35
FULL STUDENT - Half Day $50 / Full Day $100 / 90-min $40 / Workshop 2+3 $70 / FRI BREAKFAST $10
DOWNLOAD Workshop Flyer
WORKSHOP 1 - Full Day Wednesday 18 Oct - 9:00am to 5:00pm
Title: Causal analysis methods: beyond standard regression
Room: Ballroom 1
Description: In this era of “data science” it is vitally important to articulate clearly the questions that we ask of data, understand the challenges inherent in answering different types of question, and ensure that our analysis methods are suitably aligned. Many clinical and public health research studies ask causal (“what if…”) questions, about the effects of treatments, policies, behaviours and other exposures on health outcomes. Answers to these questions are key to guiding decision making in health policy and practice when the goal is to improve patient outcomes and population health. Causal inference requires carefully structured reasoning to guide appropriate statistical analysis.
This workshop will combine lectures and tutorials to provide an understanding of key concepts relating to the design of causal analyses, and analysis methods beyond standard regression, specifically g-computation and inverse probability (or "propensity score") weighting (IPW) for estimating causal effects. Emphasis will be on the underlying intuition and assumptions, whilst a hands-on computer practical (in R and Stata) will cover implementing estimation methods in practice. All lectures and tutorials will include illustrations from observational epidemiological studies. Electronic copies of presentation materials will be made available to course attendees.
Target Audience: The target audience is epidemiologists and researchers with some statistical background, including some knowledge of regression methods. It is desirable that participants are familiar with the distinction between the three types of research question (description/prediction/causal inference) and the target trial framework for defining causal effects and designing causal analyses (such as covered in the course "Observational studies: Modern concepts & analytic methods(link is external)" to be delivered on 9-13 October 2023 online via Zoom by the Clinical Epidemiology and Biostatistics Unit (CEBU) at the Melbourne Children’s campus).
For the computer practical, students must also have a sound working familiarity with Stata or R and have the corresponding software installed on their laptops.
Laptops: For the computer practical, students must also have a sound working familiarity with Stata or R and have the corresponding software installed on their laptops.
For Stata, no additional packages need to be installed beforehand.
For R, the following packages need to be loaded: “readstata13”, “tableone”, “boot”, “ggplot2”, “gtools”, “survey”, “margins” and "quantreg". These packages are all available on CRAN and can be installed using the command:
install.packages(c("readstata13", "tableone", "boot", "ggplot2", "gtools", "survey", "margins","quantreg"))
A/Prof Margarita Moreno-Betancur, Principal Research Fellow, CEBU Co-Director, MCRI and University of Melbourne.
Professor John Carlin, Senior Principal Research Fellow, MCRI and University of Melbourne
Dr Ghazaleh Dashti, Biostatistician, MCRI and University of Melbourne
Dr Marnie Downes, Biostatistician, MCRI and University of Melbourne
Dr Daisy Shepherd, Postdoctoral Research Fellow, MCRI and University of Melbourne
Dr Rushani Wijesuriya, Biostatistician, MCRI and University of Melbourne
Dr Tong Chen, Biostatistician, MCRI and University of Melbourne
WORKSHOP 2 - 90-minutes Wednesday 18 Oct - 9:00am to 10:30am
Title: Using the SHINE Health Intervention Impact Calculator (HIIC)
Room: Ballroom 2
Description: HIIC is an online tool for estimating the health gain, health expenditure and productivity impacts of interventions.
Before using HIIC, the user first estimates what the intervention will achieve in terms of a percentage change in future disease rates (disease incidence, morbidity or case fatality rates). For example, an urban transport intervention might lower asthma incidence by X% and COPD severity
by Y%. Given these inputs, HIIC outputs the health adjusted life years (HALYs), expenditure and income productivity impacts - by year into the future. The user can then compare their intervention with other more formally evaluated interventions. If the user also has an estimate of the up-front cost of the intervention, HIIC estimates cost-effectiveness.
These HIIC outputs should be of high value to:
- researchers wanting to quantify (e.g. in publications) the impact an intervention they are studying
- policy makers and advocacy groups wanting to prioritise which interventions should (and should not) be implemented.
The workshop will:
- describe how HIIC works
- present an example for eradicating mould in Australian houses and its impact of COPD severity
- show how to use HIIC for interventions solicited from participants prior to (or during) the workshop.
Professor Tony Blakey, Epidemiologist and Public Health Medicine Specialist, Population Interventions Unit, Melbourne University
Dr Tim Wilson, Simulation modeller, Population Interventions Unit, Melbourne University
WORKSHOP 3 - Half Day Wednesday 18 Oct - 1:00pm to 5:30pm
Title: The Science, Art and Opera of Epidemiology
Room: Ballroom 2
Description: The Workshop will cover half a dozen major topics discussing my personal views on what Epidemiology and Statistics are all about (such as trying to find the signal from the noise), understanding the difference between a risk factor and a risk score, the multiple uses and advantages of family and twin study designs especially now that proper control samples are almost impossible to collect, clarifying the difference between population and sample and understanding about whom inference is being made, causes of variation versus causes per se, effects versus associations, making inference about causation using paired and/or family data and novel methods for making statistical inference, understanding what power calculations are all about once you know what the hypothesis and test statistic is, how epidemiological thinking can improve the practice of artifical intelligence (AI), and vice versa, and finally about terminology and how to write an epidemiology paper.
Professor John Hopper, Professor Fellow, Melbourne School of Population and Global Health, University of Melbourne
WORKSHOP 4 - Friday Breakfast 20 Oct - 8:00am to 9:00am
Title: You have the questions, the Generation Victoria cohorts may provide the answers - Collaborating with GenV and preparing to use the Open Science dataset
Description: For many epidemiologists, answering important health, social, educational and economic questions can be hampered by lack of access to contemporary, large and population-representative datasets of Australian children and adults.
Generation Victoria (GenV, genv.org.au) is Australia's largest birth and mid-age cohort and the only mega-cohort mounted during the COVID-19 pandemic. It is both an observational and interventional cohort of children and mid-life adults. Like other internationally-significant cohorts, GenV will provide an Open Science dataset for causal modelling analyses. GenV also offers collaborative opportunities to embed registries, trials, observational studies, measures and biosamples.
In this 1 hour breakfast workshop, senior GenV team members will describe the study and value to epidemiologists (35 minutes, including time to answer your questions), followed by working with you to plan how to work with this cohort (25 minutes).
Key components of the workshop:
GenV’s study design, who is in GenV, data collected to date, and data linkage plans
How to work with GenV
Process and case studies for embedding registries, trials, observational studies, measures and biosamples into GenV
Helping workshop participants start to plan using the Open Science dataset
Dr Susan Clifford Gen V and Me Lead, Gen V, Murdoch Children's Research Institute
Mr William Siero, Cohort Stream Lead, Gen V, Murdoch Children's Research Institute
Ms Suzanne Long, Trials Lead & Integrated Studies Lead Gen V, Murdoch Children's Research Institute
Mr Jatender Mohal, Data Lead Gen V, Murdoch Children's Research Institute
Professor Melissa Wake, Scientific Director GenV, Murdoch Children's Research Institute
Contact: AEA 2023 Conference Secretariat
P: 02 6171 1312