Applied Survival Analysis and Lifetime Data Analysis

Description

This course introduces the fundamental concepts of time-to-event data analysis. Following a review of the basic principles (censoring, truncation, survival distributions, hazard functions, partial likelihood estimation techniques), students will be given a range of applied case studies in the field of medicine to analyse in SAS software.

Availability

Distance Education - Callaghan

  • Semester 1 - 2017

Learning Outcomes

On successful completion of the course students will be able to:

1. Understand the basic theoretic and applied principles of survival analysis

2. Analyse survival data using appropriate routines in SAS software

3. Check underlying model assumptions

4. Generate diagnostic plots

5. Compute sample size and power for survival models

6. Prepare publication-ready summary tables and write-up of results

Content

Survival analysis remains one of the most important tools in the field of medical research. Topics covered in this course will include basic parametric (Coale-McNeil, Exponential, Generalized F, Generalized Log Gamma, Gompertz, Weibull), non-parametric (Kaplan-Meier, Nelson-Aalen, Mantel-Haenszel), and semi-parametric (proportional hazards) methods of survival analysis.  Additional topics will include competing risks, time-dependent covariates, frailty models, accelerated failure time models, the Fine-Gray model, Makeham-type models, Wei-Lin-Weissfeld multivariate failure time models, and piecewise exponential survival models.  Methods of model checking such as Martingale and Schoenfeld deviance residuals plots also will be covered.

Requisite

Must be enrolled in Graduate Diploma of Medical Biostatistics or Master of Medical Statistics to enrol in this course. Pre-requisites: must have successfully completed BIOS6040, BIOS6050, BIOS6070, BIOS6170, and EPID6420.

Assessment Items

Written Assignment: Essays / Written Assignments

Written Assignment: Short answer exercises