Data from a randomised trial of two treatment regimens for lung cancer. This is a standard survival analysis data set from the classic textbook by Kalbfleisch and Prentice.



The Lung object is a list containing three elements:

  • y:A two column matrix (Surv object) containing the follow-up time (in days) and an indicator variable for whether the patient died while on the study or not.

  • X:Matrix of predictors

  • group:Vector describing how the columns of X are grouped

The matrix X contains the following columns:

  • trt:Treatment indicator (1=control group, 2=treatment group)

  • karno1, karno2, karno3:Orthogonal polynomials of first, second, and third degree representing Karnofsky performance score (0=bad, 100=good)

  • diagtime1, diagtime2:Orthogonal polynomials of first and second degree representing time from diagnosis to randomization (months)

  • age1, age2, age3:Orthogonal polynomials of first, second, and third degree representing the patient's age in years

  • prior:Prior therapy (0=no, 1=yes)

  • squamous, small, adeno, large:Indicators for the lung cancer cell type. For each subject, exactly one of these columns will be 1 and the rest 0.


  • Kalbfleisch D and Prentice RL (1980), The Statistical Analysis of Failure Time Data. Wiley, New York.

See also


hist(Lung$y[,1], xlab="Follow-up time", main="")

#>   0   1 
#>   9 128