Subscribe Bookmark RSS Feed

How to make adjusted survival curve?

Ted

Community Trekker

Joined:

Mar 29, 2016

There are two groups of patients. How to make adjusted (for only one factor -- the age) survival curve?

5 REPLIES
markbailey

Staff

Joined:

Jun 23, 2011

I do not know what you mean by 'adjusted survival curve.'

I interpret your question to mean that you want to fit a sample of life data from two populations to a distribution model.

  • Select Analyze > Reliability and Survival > Life Distribution.
  • Select the Compare Groups tab at the top of the dialog.
  • Select the data column with the life data and click Y, Time to Event.
  • Select the data column with the groups and click Grouping.
  • If you have censoring, select the data column with the exact/censored indicator and click Censor. Then select the censor code.
  • Click OK.
  • Click the red triangle next to Life Distribution and select Show Survival Curve.
  • Select a Distribution model and appropriate Scale (e.g., lognormal).

Please see Help > Books > Reliability and Survival > Chapter 3: Life Distribution for more details. See also other chapters for the Survival, Fit Parametric Survival, and Fit Proportional Hazards platforms in case you need such analyses.

Learn it once, use it forever!
Ted

Community Trekker

Joined:

Mar 29, 2016

Excuse me: I clarify the question. There are two groups of patients unbalanced by age: in one of them mostly young, in the other - mostly older. How to make adjusted (by age) Kaplan-Meier curves?

P.S. In any case: where is Compare Groups tab?

markbailey

Staff

Joined:

Jun 23, 2011

Do you want to compare the survival of the two groups, account for patient age (covariate), or both?

The Compare Groups tab is in the launch dialog for Life Distribution. You should see it as you followed my detailed instructions.

Learn it once, use it forever!
Ted

Community Trekker

Joined:

Mar 29, 2016

I would like to obtain Kaplan-Meier curves (for two groups), taking into account the method of which it is said: "Adjusted survival curves are often presented in medical research articles. The most commonly used method for calculating such curves is the mean of covariates method, in which average values of covariates are entered into a proportional hazards regression equation."

markbailey

Staff

Joined:

Jun 23, 2011

Please see Help > Books > Reliability and Survival > Chapter 15: Fit Proportional Hazards.

See also Help > Sample Data > Reliability/Survival > VA Lung Cancer data table. Run the saved Fit Proportional Hazards table script to see an example of the Baseline Survival at the Mean plot.

Learn it once, use it forever!