compound.Cox (version 3.30)

PBC: Primary biliary cirrhosis (PBC) of the liver data

Description

A subset of primary biliary cirrhosis (PBC) of the liver data in the book "Counting Process & Survival Analysis" by Fleming & Harrington (1991). This subset is used in Tibshirani (1997).

Usage

data(PBC)

Format

A data frame with 276 observations on the following 19 variables.

T

Survival times (either time to death or censoring) in days

d

Censoring indicator, 1=death, 0=censoring

trt

Treatment indicator, 1=treatment by D-penicillamine, 0=placebo

age

Age in years (days divided by 365.25)

sex

Sex, 0=male, 1=female

asc

Presence of ascites, 0=no, 1=yes

hep

Presence of hepatomegaly, 0=no, 1=yes

spi

Presence of spiders, 0=no, 1=yes

ede

Presence of edema, 0=no edema, 0.5=edema resolved by therapy, 1=edema not resolved by therapy

bil

log(bililubin, mg/dl)

cho

log(cholesterol, mg/dl)

alb

log(albumin, gm/dl)

cop

log(urine copper, mg/day)

alk

log(alkarine, U/liter)

SGO

log(SGOT, in U/ml)

tri

log(triglycerides, in mg/dl)

pla

log(platelet count, [the number of platelets per-cubic-milliliter of blood]/1000)

pro

log(prothrombin time, in seconds)

gra

Histologic stage of disease, graded 1, 2, 3, or 4

Details

Survival data consisting of 276 patients with 17 covariates. Among them, 111 patients died (d=1) while others were censored (d=0). The covariates consist of a treatment indicator (trt), age, sex, 5 categorical variables (ascites, hepatomegaly, spider, edema, and stage of disease) and 9 log-transformed continuous variables (bilirubin, cholesterol, albumin, urine copper, alkarine, SGOT, triglycerides, platelet count, and prothrombine).

References

Tibshirani R (1997), The Lasso method for variable selection in the Cox model, Statistics in Medicine, 385-395.

Examples

data(PBC)
PBC[1:5,] ### profiles for the first 5 patients ###
# See also Appendix D.1 of Fleming & Harrington, Counting Process & Survival Analysis (1991) #