Who is the course for?

This is an introductory course for anyone who is interested in finding out about mathematical tools related to the development of diagnostic and prognostic models in clinical practice. This course will suit researchers, clinicians or non-clinicians, who have an interest in optimizing diagnostic and prognostic systems and who need to present their work for publication or grant application. The basic concepts including how to develop and evaluate the performance of diagnostic/prognostic models will be explained through the use of real examples. Available software will also be discussed.

Course description

There is an increasing importance of diagnostic and prognostic tools in medicine and healthcare. More than ever the practice of medicine and delivery of healthcare depends upon making an accurate diagnosis or a prognosis based on the use of these tools. These models are increasingly used to complement clinical reasoning and medical decision-making. The reliability of a test has to be known in making a diagnosis or a prognosis. This course aims to equip participants to further develop their understanding on interpreting parameters of diagnostic and prognostic tests and on comparing the quality of these tools.

Learning objectives

By the end of the session, participants should be able to:

  • Identify the correct statistical test for evaluating diagnostic tests
  • Design and conduct a ROC study.
  • Increasing the knowledge of the roles that prognostic models may play in clinical practice and the factors that determine the validity of predictions from a prognostic model.
  • Understand the various steps to develop a prognostic model, from the selection of predictors to the appraisal of the external validity.
  • Summarise the results of diagnostic and prognostic studies in a publication or grant application.

About the trainer

Dr Clare Flach is a Lecturer in Medical Statistics in the School of Medicine, King’s College London. She also acts as a consultant advisor to clinical researchers.