pbc {SMPracticals}

Mayo Clinic Primary Biliary Cirrhosis Data

Description

Followup of 312 randomised and 108 unrandomised patients with primary biliary cirrhosis, a rare autoimmune liver disease, at Mayo Clinic.

Usage

data(pbc)

Format

A data frame with 418 observations on the following 20 variables.

age

in years

alb

serum albumin

alkphos

alkaline phosphotase

ascites

presence of ascites

bili

serum bilirubin

chol

serum cholesterol

edema

presence of edema

edtrt

0 no edema, 0.5 untreated or successfully treated 1 unsuccessfully treated edema

hepmeg

enlarged liver

time

survival time

platelet

platelet count

protime

standardised blood clotting time

sex

1=male

sgot

liver enzyme (now called AST)

spiders

blood vessel malformations in the skin

stage

histologic stage of disease (needs biopsy)

status

censoring status

trt

1/2/-9 for control, treatment, not randomised

trig

triglycerides

copper

urine copper

Source

Fleming, T. R. and Harrington, D. P. (1991) Counting Processes and Survival Analysis. Wiley: New York.

References

Davison, A. C. (2003) Statistical Models. Cambridge University Press. Page 549.

Examples

data(pbc)  
# to make version of dataset used in book
pbcm <- pbc[(pbc$trt!=-9),]
pbcm$copper[(pbcm$copper==-9)] <- median(pbcm$copper[(pbcm$copper!=-9)])
pbcm$platelet[(pbcm$platelet==-9)] <- median(pbcm$platelet[(pbcm$platelet!=-9)])
attach(pbcm)

library(survival)
par(mfrow=c(1,2),pty="s")
plot(survfit(Surv(time,status)~trt),ylim=c(0,1),lty=c(1,2),
   ylab="Survival probability",xlab="Time (days)")
plot(survfit(coxph(Surv(time,status)~trt+strata(sex))),ylim=c(0,1),lty=c(1,2),
   ylab="Survival probability",xlab="Time (days)")
lines(survfit(coxph(Surv(time,status)~trt)),lwd=2)
# proportional hazards model fit
fit <- coxph(formula = Surv(time, status) ~ age + alb + alkphos + ascites + 
      bili + edtrt + hepmeg + platelet + protime + sex + spiders, data=pbcm)
summary(fit)
step.fit <- step(fit,direction="backward")