Australian general practice electronic health records contain a wealth of patient information. However, the suitability of these data for developing clinical prediction tools is unclear. This research focussed on determining the suitability of these data for prediction model development using osteoarthritis as a case-study. A comprehensive data quality assessment was conducted and a prediction model for total knee replacement surgery developed. The thesis demonstrated that suitability of Australian general practice electronic health record data for prediction model development should be assessed on a case-by-case basis. A decision aid was developed to assist researchers in determining whether their electronic health record data are fit for prediction model development.