Summary of a Survival Curve

Returns a list containing the survival curve, confidence limits for the curve, and other information.

Usage

# S3 method for class 'survfit'
summary(object, times, censored=FALSE, scale=1,
  extend=FALSE, rmean=getOption('survfit.rmean'), data.frame=FALSE, dosum, ...)
  # S3 method for class 'survfitms'
summary(object, times, censored=FALSE, scale=1,
  extend=FALSE, rmean=getOption('survfit.rmean'), data.frame=FALSE, dosum, ...)

Arguments

object: the result of a call to the survfit function.
times: vector of times; the returned matrix will contain 1 row for each time. The vector will be sorted into increasing order; missing values are not allowed. If censored=T, the default times vector contains all the unique times in fit, otherwise the default times vector uses only the event (death) times.
censored: logical value: should the censoring times be included in the output? This is ignored if the times argument is present.
scale: numeric value to rescale the survival time, e.g., if the input data to survfit were in days, scale = 365.25 would scale the output to years.
extend: logical value: if TRUE, prints information for all specified times, even if there are no subjects left at the end of the specified times. This is only used if the times argument is present.
rmean: Show restricted mean: see print.survfit for details
data.frame: if TRUE, return the results as a data frame, rather than a summary.survfit object
dosum: only applicable if times is present, see details below
...: for future methods

Value

if data.frame = TRUE, a data frame with columns of time, n.risk, n.event, n.censor, surv, strata (if present) and, for survfit.coxph, the relevant row of newdata. Also std.err, upper and lower if the curve had se.fit=TRUE.

if data.frame = FALSE, a list with the following components:

surv: the estimate of survival at time t+0.
time: the timepoints on the curve.
n.risk: the number of subjects at risk at time t-0 (but see the comments on weights in the survfit help file).
n.event: if the times argument is missing, then this column is the number of events that occurred at time t. Otherwise, it is the cumulative number of events that have occurred since the last time listed until time t+0.
n.entered: This is present only for counting process survival data. If the times argument is missing, this column is the number of subjects that entered at time t. Otherwise, it is the cumulative number of subjects that have entered since the last time listed until time t.
n.exit.censored: if the times argument is missing, this column is the number of subjects that left without an event at time t. Otherwise, it is the cumulative number of subjects that have left without an event since the last time listed until time t+0. This is only present for counting process survival data.
std.err: the standard error of the survival value.
conf.int: level of confidence for the confidence intervals of survival.
lower: lower confidence limits for the curve.
upper: upper confidence limits for the curve.
strata: indicates stratification of curve estimation. If strata is not NULL, there are multiple curves in the result and the surv, time, n.risk, etc. vectors will contain multiple curves, pasted end to end. The levels of strata (a factor) are the labels for the curves.
call: the statement used to create the fit object.
na.action: same as for fit, if present.
table: table of information that is returned from print.survfit function.
type: type of data censoring. Passed through from the fit object.

Details

This routine has two uses: printing out a survival curve at specified time points (often yearly), or extracting the values at specified time points for further processing. In the first case we normally want extend=FALSE, i.e., don't print out data past the end of the curve. If the times option only contains values beyond the last point in the curve then there is nothing to print and an error message will result. For the second usage we often want extend=TRUE, so that the results will have a predictable length. If data.frame = TRUE then either might be desired. Be aware, however, that these extended values will often be badly biased; we are essentialy treating the final censored subjects as immortal.

The underlying survival object will have a row for each unique event or censoring time. When the times argument contains values not in the data, the routine can only use a best guess for the number at risk, i.e., the number at risk at the next event/censoring time. When the routine is called with counting process data many users are confused by counts that appear too large. For example, Surv(c(0,0, 1, 5), c(2, 3, 8, 10), c(1, 0, 1, 0)), which prints as (0,2] (0, 3+] (1, 8] (5,10+]. Do survfit followed by summary with a request for the values at time 0. The survfit object has entries only at times 2, 3, 8, and 10; there are 3 subjects at risk at time 2, so that is what will be printed for time 0.

For a printout at fixed times, for example yearly values for a curve, the printed number of events will by default be the total number of events that have occured since the prior line of printout, and likewise for number of censored and number at entry, dosum = TRUE. Alternately, the routine can return the number of events/censors/entry at that time, dosum=FALSE. This feature was added at the request of a user who essentially wanted to use the times argument as a subscript to pick off selected rows of the output, e.g., to select survival values corresponding to the last follow-up times of a new set of observations. The default for dosum is TRUE if the times vector is strictly increasing and FALSE otherwise.

For a survfitms object replace the surv component with pstate. Also, a data frame will not include the cumulative hazard nor its standard error, since it has a different multiplicity: one column per transition rather than one per state.

Examples

summary( survfit( Surv(futime, fustat)~1, data=ovarian))
#> Call: survfit(formula = Surv(futime, fustat) ~ 1, data = ovarian)
#> 
#>  time n.risk n.event survival std.err lower 95% CI upper 95% CI
#>    59     26       1    0.962  0.0377        0.890        1.000
#>   115     25       1    0.923  0.0523        0.826        1.000
#>   156     24       1    0.885  0.0627        0.770        1.000
#>   268     23       1    0.846  0.0708        0.718        0.997
#>   329     22       1    0.808  0.0773        0.670        0.974
#>   353     21       1    0.769  0.0826        0.623        0.949
#>   365     20       1    0.731  0.0870        0.579        0.923
#>   431     17       1    0.688  0.0919        0.529        0.894
#>   464     15       1    0.642  0.0965        0.478        0.862
#>   475     14       1    0.596  0.0999        0.429        0.828
#>   563     12       1    0.546  0.1032        0.377        0.791
#>   638     11       1    0.497  0.1051        0.328        0.752
summary( survfit( Surv(futime, fustat)~rx, data=ovarian))
#> Call: survfit(formula = Surv(futime, fustat) ~ rx, data = ovarian)
#> 
#>                 rx=1 
#>  time n.risk n.event survival std.err lower 95% CI upper 95% CI
#>    59     13       1    0.923  0.0739        0.789        1.000
#>   115     12       1    0.846  0.1001        0.671        1.000
#>   156     11       1    0.769  0.1169        0.571        1.000
#>   268     10       1    0.692  0.1280        0.482        0.995
#>   329      9       1    0.615  0.1349        0.400        0.946
#>   431      8       1    0.538  0.1383        0.326        0.891
#>   638      5       1    0.431  0.1467        0.221        0.840
#> 
#>                 rx=2 
#>  time n.risk n.event survival std.err lower 95% CI upper 95% CI
#>   353     13       1    0.923  0.0739        0.789        1.000
#>   365     12       1    0.846  0.1001        0.671        1.000
#>   464      9       1    0.752  0.1256        0.542        1.000
#>   475      8       1    0.658  0.1407        0.433        1.000
#>   563      7       1    0.564  0.1488        0.336        0.946
#>

Usage

Arguments

Value

Details

See also

Examples