Advanced Analysis of Prognosis Studies

Prognostic models are increasingly published in the medical literature each year. But are the results relevant for clinical practice? What are the critical elements of a well developed prognostic model? How can we assume that the model makes accurate predictions for our patients, and not only for the sample that was used to develop the model (generalizability, or external validity)?

In the course we will address these and other questions from a methodological perspective, using examples from the clinical literature.

The participants will be encouraged to participate in interactive discussion to increase their knowledge of the roles that prognostic models may play in clinical practice and the critical factors that determine the validity of predictions from a prognostic model.

Objectives 

  • Increasing the knowledge of the roles that prognostic models may play in clinical practice and the critical factors that determine the validity of predictions from a prognostic model.  
  • Gain insight in the pitfalls in prognostic model development with standard statistical techniques.  
  • Acquire knowledge on advanced statistical techniques in prognostic model development and validation.     

Required book: Clinical Prediction Models: A Practical Approach to Development, Validation, and Updating (Steyerberg, Springer Series: Statistics Statistics for Biology and Health, ISBN: 978-0-387-77243-1)

Books are for sale during the Erasmus Winter Programme.

Course code: EWP13

Faculty: Ewout Steyerberg

Date: January 30 – February 3, 2012

Time: 13:00 – 16:00

Days: Monday to Friday

Course material: Online, download instructions will be sent in advance by e-mail. 
A laptop with SPSS is required.

Prerequisites: Knowledge of basic epidemiological notions, especially cohort studies and randomized controlled trials. Knowledge of statistical concepts (t-test, normal distribution, correlation) and regression techniques (linear, logistic, Cox regression).

Course fee: € 600