This unit introduces the topic of analysing linked health data. The topic is very specialised and is relevant to those who are familiar with writing a basic SAS program, who wish to further develop their knowledge and skills in managing and analysing linked health data, eg. hospital admissions, cancer registry, births and deaths. Contents include: an overview of the theory of data linkage methods and features of comprehensive data linkage systems, sufficient to know the sources and limitations of linked health data sets; design of linked data studies using epidemiological principles; construction of numerators and denominators used for the analysis of disease trends and health care utilisation and outcomes; assessment of the accuracy and reliability of data sources; data linkage checking and quality assurance of the study process; basic statistical analyses of linked longitudinal health data; manipulation of large linked data files; writing syntax to prepare linked data files for analysis, derive exposure and outcome variables, relate numerators and denominators and produce results from statistical procedures at an introductory to intermediate level. The unit is delivered as a workshop over 5 consecutive days. Lectures are delivered in the morning sessions and the afternoon sessions are computer labs where students gain hands-on experience using large health datasets. The unit is usually offered twice a year, once in mid-June and once in mid-November.
Block/intensive mode - 5 days, 9am - 5pm
Reflective journal (30%) and 1x data analysis assignment (70%)
Notes will be distributed in class.
Familiarity with writing a basic SAS program. For data privacy and security reasons, the major assignment can only be completed on the computers in the Sydney School of Public Health Computer Lab. This computer lab is available 24/7 for students enrolled in this unit.
Basic familiarity with SAS computing syntax and methods of basic statistical analysis of fixed-format data files
(PUBH5010 or BSTA5011 or CEPI5100) and (PUBH5211 or PUBH5217 or BSTA5004)