The participant will be introduced to the everywhere used descriptive survival curve of censored or truncated time to event data; The Cox proportional hazards model and its validation; Pitfalls in the analysis of time to event data; The extension to time-varying exposures, competing risks and dynamic prediction; Risk estimation using pseudo values; The design of survival studies, power considerations, various types of censoring and presentation of obtained results. Methods will be demonstrated using the statistical software packages R and Stata and the participant will get the opportunity to try out hands-on analyses.
This course gives an introduction to survival analysis for clinical and observational studies. The students will gain insight to modern survival analysis ranging from design of study to analysis and interpretation of publishable results. The students learn how to investigate and analyse time to event data in health science, eg. time to diagnosis under various exposures and conditions.
Course credit: 1,8 ECTS.
Course arrangements: 2 days, one week apart followed by a small take-home exam
Lectures and practicals using statistical software, Stata and/or R. Example studies and solutions to often encountered problems are provided.
Course leader: Professor Jacob Hjelmborg, Epidemiology, Biostatistics & Biodemography, SDU.
Signe Bedsted Clemmensen, Martin D Villumsen, Simon BM Kristensen and Jacob vB Hjelmborg (University of Southern Denmark).
Knowledge about statistics corresponding to what is acquired from a introductory level PhD course, eg Biostatistics I or similar. Basic experience with statistical software is an advantage.
The course is free of charge for PhD students enrolled in Universities that have joined the "Open market agreement".
For other participants there is a course fee of DKK 2320,-