Per Kragh Andersen
(Department of Biostatistics, University of Copenhagen)
and
Hein Putter
(Department of Medical Statistics and Bioinformatics Leiden University Medical Centre)
28 - 29 May 2009
Location: Erasmus Medical Faculty
Room: Colloquium B
Deadline for registration: 22 May 2009. For registration and further
practical information, please contact Dymph Wijnen, Department of Biostatistics, Erasmus MC, Dr.
Molewaterplein 50, 3015 GE Rotterdam, Ee 2124 (21st floor), tel: +31-10-7044514
Erasmus MC University Medical Center, Department of Biostatistics
News & Events
COURSE: Multi-state models and models for competing risks
Abstract
This course will survey methods for the analysis of multiple events per subject: multi-state models. These are either mutually exclusive, competing risks, or can occur sequentially. We will discuss ways of extending the situation of survival analysis with a single endpoint to multi-state models through concepts like transition intensities or hazards. The special case of competing risks is discussed in some detail
(including concepts like cause-specific and sub-distribution hazards and cumulative incidence functions).
Incorporation of covariates will receive special attention. For general multi-state models, regression models for the transition hazards are typically used. In the special case of competing risks, both regression models based on transition intensities (“cause-specific hazards”) and transition probabilities (or “sub-distribution hazards”, the Fine & Gray model) are available. We will discuss ways of estimating regression coefficients and transition intensities, and on how to obtain (dynamic) predictions from the model using the Aalen-Johansen estimator. In the last part a (partial) overview of ongoing research in the field is given. Many relevant examples are given, mostly from medical statistics. Examples of analysis with standard software and more specialized R packages are shown throughout the course.
The course is directed to people actively involved in the field of medical statistics or epidemiology with some background in survival analysis.
Suggested literature:
Putter H, Fiocco M, Geskus RB (2007). Tutorial in biostatistics: Competing risks and multi-state models. Statist. Med. 26: 2389–2430.
Andersen PK, Keiding N (2002). Multi-state models for event history analysis. Statist Meth Med Res. 11: 91–115.
Andersen PK, Pohar Perme M (2008). Inference for outcome probabilities in multi-state models. Lifetime Data Analysis 14: 405–431.
(including concepts like cause-specific and sub-distribution hazards and cumulative incidence functions).
Incorporation of covariates will receive special attention. For general multi-state models, regression models for the transition hazards are typically used. In the special case of competing risks, both regression models based on transition intensities (“cause-specific hazards”) and transition probabilities (or “sub-distribution hazards”, the Fine & Gray model) are available. We will discuss ways of estimating regression coefficients and transition intensities, and on how to obtain (dynamic) predictions from the model using the Aalen-Johansen estimator. In the last part a (partial) overview of ongoing research in the field is given. Many relevant examples are given, mostly from medical statistics. Examples of analysis with standard software and more specialized R packages are shown throughout the course.
The course is directed to people actively involved in the field of medical statistics or epidemiology with some background in survival analysis.
Suggested literature:
Putter H, Fiocco M, Geskus RB (2007). Tutorial in biostatistics: Competing risks and multi-state models. Statist. Med. 26: 2389–2430.
Andersen PK, Keiding N (2002). Multi-state models for event history analysis. Statist Meth Med Res. 11: 91–115.
Andersen PK, Pohar Perme M (2008). Inference for outcome probabilities in multi-state models. Lifetime Data Analysis 14: 405–431.
Contents
Day 1
9.00 - 9.15 Welcome (EL)
9.15 - 10.45 Introduction (PA)
10.45 - 11.15 Coffee break
11.15 - 12.45 Competing risks (HP)
12.45 - 14.00 Lunch
14.00 - 15.30 Multi-state models, non-parametric and Cox (HP)
15.30 - Questions (PA + HP)
Day 2
9.00 - 10.30 Prediction in multi-state models (HP)
10.30 - 11.00 Coffee break
11.00 - 12.00 Interval censoring (PA)
12.00 - 13.15 Lunch
13.15 - 15.00 Pseudo-values, additive hazards (PA) + Predicting by landmarking (HP)
9.00 - 9.15 Welcome (EL)
9.15 - 10.45 Introduction (PA)
10.45 - 11.15 Coffee break
11.15 - 12.45 Competing risks (HP)
12.45 - 14.00 Lunch
14.00 - 15.30 Multi-state models, non-parametric and Cox (HP)
15.30 - Questions (PA + HP)
Day 2
9.00 - 10.30 Prediction in multi-state models (HP)
10.30 - 11.00 Coffee break
11.00 - 12.00 Interval censoring (PA)
12.00 - 13.15 Lunch
13.15 - 15.00 Pseudo-values, additive hazards (PA) + Predicting by landmarking (HP)
Administrative information
Coffee breaks
2 coffee breaks (one in morning and one in afternoon) are foreseen and are included in the registration costs. Lunch is not included.
Course materials
Copies of the transparencies used in the course are included in the registration costs.
Costs
Erasmus University: € 150,00
Other universities/governmental: € 250,00
Commercial organizations: € 500,00
Registration is only effective upon receipt of payment. For payment send us your invoice address and for Erasmus personnel also the "kostenplaats".
2 coffee breaks (one in morning and one in afternoon) are foreseen and are included in the registration costs. Lunch is not included.
Course materials
Copies of the transparencies used in the course are included in the registration costs.
Costs
Erasmus University: € 150,00
Other universities/governmental: € 250,00
Commercial organizations: € 500,00
Registration is only effective upon receipt of payment. For payment send us your invoice address and for Erasmus personnel also the "kostenplaats".