PREREQUISITES
The computing component of the unit assumes competence in the preparation of computing syntax for SPSS, SAS or Stata, and familiarity with the management and analysis of linked data files at an introductory to intermediate level. Participants must have self-sufficient access to a notebook or desktop computer each afternoon (lab provided but most participants may prefer to bring their own laptop), running a suitable statistical program (SPSS, SAS or Stata) and capable of reading a CD-ROM. A working knowledge of statistical concepts, including regression models, used in data analysis in the medical and social sciences is assumed.
COURSE OUTLINE
This is an intensive course of five days in duration, designed to instruct participants in the theory and practice of analysis of large sets of linked administrative data, especially in the health area, but with applications throughout the social sciences, at an intermediate to advanced level. Rapid growth in data linkage projects has led to a shortfall in analysts' skills. In health, 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 short course combines advanced principles of health care and social epidemiology with practical exercises to illustrate and develop the corresponding computing skills.
The modular structure of the unit provides students with 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 on CD-ROM in the hands-on exercises. Professor Holman is available in the computing laboratory session each afternoon.
LEARNING OBJECTIVES
Advanced Analysis of Linked Health Data provides health and social researchers with the opportunity to build on their pre-existing theoretical knowledge and skills in the analysis of linked data by exploring a number of advanced topics. Upon completion the participant will:
have consolidated their grasp of foundation concepts of health care and social epidemiology and data analysis;
possess an advanced understanding of methods for conceptualisation and construction of valid measures and effect measures of health and human service 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; and
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.
COURSE TEXT
The instructor's detailed course notes will serve as the course text.
RELATED COURSES
This course articulates with Introductory Analysis of Linked Data, which was offered in New Zealand by Professor Holman early in 2009. However, previous exposure to the introductory course is not essential provided that the participant has equivalent experience in the analysis of linked administrative data sets at least an introductory level.
