A survival analysis displays time-to-event data.
To begin the analysis, see Running the Analyses, then perform the following steps.
To perform a survival analysis:
Click the Advanced Workflow tab, then open the Analysis menu.
Select Survival Analysis.
The Variable Selection section appears.
Define the following variables:
- Time: A numerical measure of duration; for example, Overall Survival Time (Years).
- Category: The groups into which the data will be split in order to compare the time measured; for example, Cancer Stage. This variable is optional. If you do use it, you must enter two nodes for the comparison.
If this variable is continuous, it requires binning.
- Censoring Variable: Specifies which patients had the event whose time is being measured. For example, if the Time variable selected is Overall Survival Time (Years), an appropriate event variable is Alive.
Optionally, select Enable binning.
For details, see Data Binning Using Survival Analysis.
Your analysis appears below:
Data Binning Using Survival Analysis
Data binning is used in survival analyses if the variable you want to use is continuous (for example, age) but needs to be viewed as categorical data. Alternatively, binning can be used to regroup categorical data to consider it as a single variable. For example, if histological grade with values such as Well Defined, Moderately Well Defined, and Poorly Defined are selected, you can group Moderately Well Defined with Poorly Defined and treat them as one group for the purposes of this analysis.
To use the data binning feature with a survival analysis:
Begin to set up a Survival Analysis by following the instructions in section Survival Analysis.
Enable binning by selecting Enable binning.
Define the following and then click Run.
- Variable Type
Select whether the variable you have defined above is continuous or categorical.
A continuous variable can be treated as a categorical variable when you use the binning feature.
- Number of Bins
Type the number of bins you would like data to be organized in.
This step may require trial and error based on how you want to display data.
- Bin Assignments
Select how you would like data to be binned.
This feature can only be used when the variable type selected above is continuous.
Evenly Distribute Population: Assigns bins based on the underlying data.
For example, if the majority of the subjects in the study were elderly, bins based on age could look like: [(1-40), (40-80), (81-85), (86-90), (90-92)].
Evenly Spaced Bins: Creates bins based on the overall range of the variable.
For example, if the majority of the subjects in the study were elderly, bins based on age could look like: [(1-20), (21-40), (41-60), (61-80), (81-100)].
- Manual Binning
Select the checkbox if you want to bin manually.
This is the only binning method available if you are trying to bin a categorical variable type.
Complete the binning form that populates as a result of checking the Manual Binning box.
For continuous data:
For categorical data: