P.Mean >> Category >> Survival analysis (created 2007-09-12). 

Survival data represents data that indicates with information about the time to a certain event (often death, but it can represent other events as well). A common feature for most survival data is the process of censoring. These pages discuss the various ways you can analysis survival data. Also see Category: Modeling issues. You can find outside resources at the bottom of this page.


10. What is a Kaplan-Meier survival curve? (November 2009)

9. P.Mean: Changes in the adjusted hazard ratio, but not in the precision of the ratio (created 2009-01-19). Does anyone know a good reference on why, in Cox regression of a clinical trial, including covariates often changes the treatment hazard ratio rather than narrowing the confidence interval? I can remember attending a talk on this years ago, but cannot remember the details.

Other resources:

Time-Dependent Covariates In The Cox Proportional-Hazards Regression Model. Lloyd D. Fisher and D. Y. Lin. Annu. Rev. Public Health. 1999. 20:145�57. [Full text] [PDF]. Description: This is an excellent review article on time-dependent covariates that offer some specific examples of how and when you might incorporate these in a Cox proportional hazards model. The authors also make pointed warnings about possible problems with time-varying covariate.

Creative Commons License All of the material above this paragraph is licensed under a Creative Commons Attribution 3.0 United States License. This page was written by Steve Simon and was last modified on 2017-06-15. The material below this paragraph links to my old website, StATS. Although I wrote all of the material listed below, my ex-employer, Children's Mercy Hospital, has claimed copyright ownership of this material. The brief excerpts shown here are included under the fair use provisions of U.S. Copyright laws.


8. Stats: A simple example of a Kaplan-Meier curve (January 24, 2008). In response to a query, I wanted to write up a simple example of how to calculate survival probabilities when you have censored data. It is adapted from Chapter 6 of my book, Statistical Evidence in Medical Trials. I have updated and simplified the example, for possible use in a second edition of the book, if I am so lucky.


7. Stats: Conditional Frailty Models (January 20, 2006). One of the people I am working with is interested in using gap time analysis with a conditional frailty model. I was impressed with this request and asked her to send any relevant references that she had. She gave me a pointer to the following PDF file: Repeated events survival models: the conditional frailty model.


6. Stats: More than 90% censored values (April 22, 2005). Someone asked me about running a Cox proportional hazards regression model when over 90% of the observations were censored. That means (if the outcome of interest was death), that your research subjects did not cooperate and die fast enough. Good news from the patients' perspective, but bad news for the statistician. 90% censored observations is not a problem, though, as long as your sample size is adequate. As a rough rule of thumb, you need to have 25 to 50 events (uncensored observations) in each treatment group to have reasonable precision. Of course, if you have fewer events, the model is still valid, but your confidence intervals may end up being wider than you'd really like.

5. Stats: Stratified Cox regression models (March 22, 2005). Someone sent me an email asking about a Cox regression model that included a strata for clinics. What's the best way to handle strata? That's a tricky question to answer. The first question you might want to ask is whether it makes sense to include the clinic factor as a strata at all. When you include strata, you allow the Cox model to estimate an entirely separate hazard function for each clinic. That's quite different from including clinic as a fixed effect in the Cox regression model, where you would be assuming that the clinics differ only in that the hazard function for one clinic is a multiple of the hazard function for the other clinic.


4. Stats: The price of Kaplan-Meier [Incomplete] (September 23, 2004). I rarely find time to read the Statistics journals anymore, but I did run across an excellent article in the September 2004 issue of JASA. The Price of Kaplan-Meier. Meier P, Karrison T, Chappell R, Xie H. Journal of the American Statistical Association 2004: 99(467); 890-896.


3. Stats: Guidelines for survival data models (October 11, 2002). There are three steps in a typical survival analysis. Know how much data you have, Graph the survival function, Compare the survival times.

2. Stats: Data management for survival data (August 27, 2002). Survival data will involve calculating the time between the various dates and noting when certain dates are present or absent. In a study of bone marrow transplants for childhood cancer, we have up to four dates: Date of bone marrow transplant (always known) Date of last follow-up (always known) Date of relapse (sometimes censored) Date of death (sometimes censored) The dates of relapse and death are censored because either they did not occur, or they occurred after the date of last follow-up.


1. Stats: Kaplan Meier (June 27, 2000). Dear Professor Mean: When I read my medical journals, I keep on coming across terms like "Kaplan-Meier Product Limit Estimate" or "Kaplan-Meier survival curve." What do these terms mean and when are they used?

What now?

Browse other categories at this site

Browse through the most recent entries

Get help