Course overview
Clinical prediction models combine multiple pieces of patient information in order to predict a clinical outcome. This half-day course includes lectures and practical sessions, and aims to give its delegates an introduction to the clinical reasoning and statistical methodology behind clinical prediction models.
Who should attend? Is it right for me?
This course is aimed at researchers and PhD students within the Faculty of Health and Life Sciences who require a basic understanding of prediction modelling as part of their research studies. It is suitable for those who have no previous experience of prediction modelling, as well as those seeking to refresh their skills.
What will delegates learn?
By the end of this course delegates will have an understanding of:
• What a prediction model is
• How to develop a clinical prediction model
• How to validate a clinical prediction model
• How to present a clinical prediction model for a clinical audience
What does the course cover?
• Logistic regression & survival analysis for model development
• Discrimination & calibration statistics
• Internal & external validation
• Point score systems, graphical score charts, nomograms & websites
