This is an intensive five-day course on the theory and practice of analysis of large sets of linked administrative health data at an introductory to intermediate level.
Rapid growth in data linkage projects has led to a shortfall in analyst skills.
Some researchers understand epidemiological principles, but are unfamiliar with the specialised computing skills needed to analyse linked data files.
Others have a strong grasp of computing concepts, but lack an adequate theoretical base to design high quality applications to answer research questions. This endeavours to fill a gap in research training opportunities to cater to these two areas of need.
Professor Preen provides a theoretical grounding in the classroom on each topic, followed by a training session on the corresponding computing solutions. Students use fictitious but realistic linked data files in the hands-on exercises. A lecturer will be available in the computing laboratory session each afternoon and conducts an end-of-day tutorial for those who need additional assistance.
The course acquaints health researchers, clinical practitioners and managers with the theory and skills needed to analyse linked health data at the introductory to intermediate level. Upon completion the participant will:
- possess an overview of the theory of data linkage methods and features of comprehensive data linkage systems, sufficient to understand the sources and limitations of linked health data sets
- understand the principles of epidemiologic measurement and research methods for the conceptualisation and construction of numerators and denominators used in the analysis of disease trends and health care utilisation and outcomes
- understand sources of error in epidemiologic measurement, the difference between confounding and effect modification, and use of regression models in risk adjustment in health services research
- be able to perform statistical analyses on linked longitudinal health data
- be able to conceptualise and perform the manipulation of large linked data files
- be able to write syntax to prepare linked data files for analysis, derive exposure and outcome variables, relate numerators and denominators and produce results from statistical procedures
Basic familiarity with computing syntax used in programs such as SPSS, SAS or Stata and methods of basic statistical analysis of fixed-format data files.
There are no formal prerequisites in epidemiology for the course. However, it is recommend that participants who have not previously completed an introductory course in epidemiology, familiarise themselves with the basic principles and terms used in that discipline. A working knowledge of statistical concepts, including regression models, used in data analysis in the medical and social sciences is assumed.