survreport {IPDfromKM}

Survival analysis on the reconstructed IPD

Description

Graph the Kaplan-Meier curves and the cumulative hazard curves for the reconstructed IPD (from the output of getIPD function). Also report the survival times with confidence intervals for a given vector of survival probabilities, as well as the landmark survival probabilities of interest.(for example, if set interval=6, the survival probability will be reported at every six months)

Usage

survreport(ipd1,ipd2=NULL,arms=1,interval=6,s=c(0.75,0.5,0.25),showplots=TRUE)

Arguments

ipd1

a three-column (i.e., time, status, and treatment indicator)table of IPD for treatment 1.

ipd2

a three-column (i.e., time, status, and treatment indicator)table of IPD for treatment 2.

arms

number of treatment group. Can be either 1 or 2.

interval

length of the time interval for which the landmark survival probabilities are of interest. The default is at every 6 months.

s

a vector with survival probabilities for which the corresponding survival times are reported. e.g., s=0.5 means that the median survival time is desired.

showplots

indicate if the survival plots are displayed or not in the plot window

Value

survreport() returns a list object.

References

Guyot P, Ades AE, Ouwens MJ, Welton NJ. Enhanced secondary analysis of survival data: reconstructing the data from published Kaplan-Meier survival curves. BMC Med Res Methodol.2012; 1:9.

Examples

### Get data from the sample dataset=======================
radio <- Radiationdata$radio
radioplus <- Radiationdata$radioplus
trisk <- Radiationdata$trisk
nrisk_radio <- Radiationdata$nrisk.radio
nrisk_radioplus <- Radiationdata$nrisk.radioplus
### Estimate the IPD for the Radiotherapy treatment group ====================
pre_radio <- preprocess(dat=radio, trisk=trisk,nrisk=nrisk_radio,maxy=100)
est_radio <- getIPD(prep=pre_radio,armID=0,tot.events=NULL)
### Estimate the IPD for the Radiotherapy plus treatment group ====================
pre_radio_plus <- preprocess(dat=radioplus, trisk=trisk,nrisk=nrisk_radioplus,maxy=100)
est_radio_plus <- getIPD(prep=pre_radio_plus,armID=1,tot.events=NULL)
### survival report for one arm ===================
surv1 <- survreport(ipd1=est_radio$IPD,arms=1,interval=6,s=c(0.75,0.5,0.25),showplots=FALSE)
print(surv1)
surv1 <- survreport(ipd1=est_radio$IPD,arms=1,interval=10,s=seq(0,1,0.2),showplots=TRUE)
print(surv1)
### survival report for two arms ===================
surv2 <- survreport(ipd1=est_radio$IPD,ipd2=est_radio_plus$IPD,arms=2,
                    interval=8,s=c(0.75,0.5,0.25),showplots=TRUE)
print(surv2)