R help - May 2010 - Understanding of survfit.formula output

Dear list,

I am not familiar with survival analysis and I would need your help to
understand a result I have obtained.

I have used the following command line to look at number of events and
probability of survival at the first 5 years:
> su = summary(survfit(Surv(a[, Date], a[, Event]) ~ strata(a[,Prediction]),
data = a), times=c(0,1,2,3,4,5))

I have studied two kind of events (disease-free survival and metastasis free
survival), please see the results below.
At year 5, in group C2 I have one more patient with an event when looking at
DFS (13) than when looking at relapse (12). However, the probability is
higher when looking at DFS (0.23) than relapse (0.18), which I cannot
understand as I have one more event.

Can anyone explain or point to documentation explaining this? Is it due to
the fact that there was no more event between year 4 and 5 when looking at
DFS?

  Relapse
0 1 2 3 4 5
patients 40 34 32 24 15 9  C1 events 0 1 1 1 2 2
survival 1 0.97 0.97 0.97 0.92 0.92
patients 19 12 10 5 3 1  C2 events 0 6 8 10 11 12
survival 0.95 0.68 0.57 0.45 0.36 0.18















 DFS
0 1 2 3 4 5
patients 40 34 32 24 15 9  C1 events 0 1 1 1 2 2
survival 1 0.97 0.97 0.97 0.92 0.92
patients 19 12 9 4 2 1  C2 events 0 6 9 12 13 13
survival 0.95 0.68 0.51 0.31 0.23 0.23


Best wishes,
Audrey

-- 
Audrey Kauffmann
Bergonie Cancer Institute
229 Cours de l'Argonne
33076 Bordeaux
France
+33.5.56.33.04.53

	[[alternative HTML version deleted]]

You have not given enough information to reproduce your problem, so it
is difficult to say.  Most of the time results such as you display will
be due to a data error.  It is possible to get a crossing result,
however.
subject     1  2   3   4  5
 -----------------------------
Death      10  -   -   -  4
Relapse     2  -   4   -  3
Last FU    10  6   6   11 4

The Kaplan-Meier curves are
 time   relapse                 death
  2      4/5                     1
  3      4/5*3/4                 1
  4      4/5* 3/4 * 2/3=.4      2/3
  10     .4                     2/3 * 1/2 = .3333

Subject 1 has a relapse early when all 5 are still at risk, dropping the
curve by .2 units.  Their death is late when only 2 are at risk,
dropping the curve by 1/2.

This crossing anomaly usually only happens near the end of a
Kaplan-Meier, when the confidence intervals are as wide as a river.
   Terry Therneau



------------ begin included message --------------------------

At year 5, in group C2 I have one more patient with an event when
looking at
DFS (13) than when looking at relapse (12). However, the probability is
higher when looking at DFS (0.23) than relapse (0.18), which I cannot
understand as I have one more event.