This is an intensive five-day unit on the theory and practice of analysis of large sets of linked administrative health data at an intermediate to advanced level.
Advanced principles of health care epidemiology are combined with hands-on practical exercises in the implementation of computing solutions.
Professor Holman provides students with a theoretical grounding in the classroom on each topic, followed by a training session on the corresponding technological computing solutions. Students use fictitious but realistic linked data files on CD-ROM in the hands-on exercises.
Professor Preen is available in the computing laboratory session each afternoon and conducts an end-of-day plenary discussion where he and class participants share perspectives on how to problem-solve complex research scenarios.
Advanced Analysis of Linked Health Data provides health researchers with the opportunity to build on their pre-existing theoretical knowledge and skills in the analysis of linked health data by exploring a number of advanced topics. Upon completion the participant will:
- have consolidated their grasp of foundation concepts of health care epidemiology and data analysis
- possess an advanced understanding of methods for the conceptualisation and construction of valid measures and effect measures of health care utilisation and outcomes based on complex, multi-sourced linked data sets
- understand complex longitudinal research designs and how to implement them using multi-sourced linked data sets
- understand case distribution designs and how to implement them using multi-sourced linked data sets
- have skills in the analysis of linked mortality, institutional, pharmaceutical and primary care health data
- be able to write computing syntax to prepare complex linked data files for analysis, derive exposure and outcome variables, relate numerators and denominators and produce results from advanced statistical procedures
The course assumes that students have completed Introductory Analysis of Linked Health Data or have equivalent knowledge. The computing component of the unit assumes a facile competence in the preparation of computing syntax for either SPSS, SAS or STATA and familiarity with the statistical analysis of linked data files at an introductory to intermediate level.