Session 8: Survival Analysis 3

Lecture

Learning Objectives

1. Check model assumptions and fit of the Cox model
   • residuals analysis
   • log-minus-log plot
2. Fit and interpret multivariate Cox models
   • perform tests for trend
   • predict survival for specific covariate patterns
   • predict survival for adjusted coefficients
3. Explain stratified analysis
4. Identify situations of competing risks
5. Describe the application of Propensity Score analysis

• Vittinghoff sections 6.2-6.4

Outline

1. Review
2. Assumptions of Cox PH model
3. Tests for trend
4. Predictions for specific covariate patterns
5. Stratification
6. Competing risks
7. Propensity Score analysis to control for confounding

Lab

Learning Objectives

1. Make stratified and unstratified Kaplan-Meier plots
2. Perform Cox proportional hazards regression
3. Assess proportional hazards assumption

Exercises

1. Load the Primary Biliary Cirrhosis (pbc) dataset from the survival package
2. Create a Surv object using variables "time" and "status", add this to the pbc dataframe
3. Plot a KM curve for all participants using library(survminer) function ggsurvplot()
4. Stratify by treatment and add a p-value to this plot (see ?ggsurvplot)
5. Check whether these p-values correspond to results from a log-rank test
6. Perform a Cox proportional hazards regression, using the "trt" variable as a predictor
7. Create a log-minus-log plot to test the proportional hazards assumption
8. Plot Schoenfeld residuals and perform Schoenfeld test for the above Cox model