Cox Regression is a semi-parametric method used to determine the effect of several variables (age, treatment, stage) upon the time a specified event takes to happen.
The output of a Cox model is the Hazard Ratio. It compares the hazard rates between groups.
| HR Value | Interpretation |
|---|---|
| HR = 1 | No effect. |
| HR > 1 | Increased risk (e.g., HR=2 means 2x risk of death). |
| HR < 1 | Protective effect (e.g., HR=0.5 means 50% reduction in risk). |
Cox regression assumes that the ratio of hazards between groups remains constant over time. For example, if Treatment A has half the risk of Treatment B at 1 month, it should also have half the risk at 1 year. If lines cross on a survival plot, this assumption may be violated.