Confounding in Datasets
Date: Monday 29 August
Time: 1:00-4:30 pm
Venue: Hotel Realm, Canberra
There is global interest in using computerised databases for comparative effectiveness and medication safety studies. However, the data represents patients in practice. Patient and physician characteristics may affect the likelihood of receiving treatments. This creates the potential problem of confounding, which must be overcome to obtain robust results from observational studies.
The technical symposium will explore the issue of confounding in Australian and international datasets.
The methods to deal with confounding in observational studies will be presented including: propensity scores; instrumental variables and the self-controlled case-series design.
Specific examples will be presented to highlight how researchers can apply these techniques to evaluate the risks and benefits of medicines.
1:00 PM – Welcome
1:10 PM – What is confounding?
1:30 PM – Dealing with confounding in effectiveness studies
2:10 PM – Dealing with confounding in safety studies
3:15 PM – Dealing with confounding in health economic studies
4:00 PM – Panel Discussion
4:30 PM – Close
Dr Nicole Pratt
Senior research fellow, Quality Use of Medicines and Pharmacy Research Centre, Sansom Institute, University of South Australia.
Professor Libby Roughead
Sansom Institute, School of Pharmacy and Medical Sciences, University of South Australia.
Professor Rosalie Viney
Associate Professor of Health Economics and Deputy Director of the Centre for Health Economics Research and Evaluation at the University of Technology, Sydney.
Mr Kees van Gool
Health Economist and Senior Lecturer at the Centre for Health Economics Research and Evaluation, University of Technology, Sydney.