[R] Survival analysis with interval censored data

[Audrey Bologna]

Hi!

Does anyone know about survival analysis using interval censored data and
the intcox package?

I fed different ant colonies with different food (called treatment) and I
measured every 2 days the death of ants. I would like to see if the
treatment can explain a different death of individuals.
Since I checked the death every 2 days, I assume that my datas are interval
censored.
Here is a part of my dataset, an exemple for one colony (1A):

date date2 fate colony condition
1 2 3     1A before
1 4 3     1A before
1 4 3     1A before
1 8 3      1A before
1 14 3      1A before
1 16 3       1A before
1 20 0      1 A before
1 20 0      1A before
1 20 0      1A before
1 20 0      1A before
1 20 0       1A before






date corresponds to the beginning of my experiment: the 21st of January and
I called it 1.
date 2 corresponds to the date of the death: 2= 23rd of Janurary, 3= 25th
of January and so one.
Fate: 3 a death event for interval censored data, 0 when individuals are
still alive.

I created a survival object this way:

surv.object<-Surv(test$date1,test$date,test$fate,type="interval") and here
is a part of what I obtain as a survival object:


[1901] [1,  4] [1,  6] [1,  6] [1,  8] [1,  8] [1,  8] [1,  8] [1, 10] [1,
10] [1, 10]
[1911] [1, 10] [1, 12] [1, 16] [1, 16] [1, 16] [1, 18] [1, 20] 1+      1+
   1+
[1921] 1+      1+      1+      1+      1+      1+      1+      1+      1+
   1+
[1931] 1+      1+      1+      1+      1+      1+      1+      1+      1+
   1+
[1941] 1+      1+      1+      1+      1+      1+      1+      1+      1+
   1+
[1951] 1+      1+      1+      1+      1+      1+      1+      1+      1+
   1+
[1961] 1+      1+      1+      1+      1+      1+      1+      1+      1+
   1+
[1971] 1+      1+      1+      1+      1+      1+      1+      1+      1+
   1+
[1981] 1+      1+      1+      1+


However when I try to apply the intcox function, it returns me warnings
messages:

 ex<-intcox(surv.object~test$colony, data=test)
Erreur dans if (any(derivs.wert$g1 <= 0)) { :
  valeur manquante là où TRUE / FALSE est requis
De plus : Message d'avis :
In coxph(formula, data) :
  X matrix deemed to be singular; variable 1 2 3 4 5 6 7 8 9 10 11 12 13 14
15 16 17 18 19 22


Can anybody help me?

Thanks a lot


Audrey Bologna - PhD Student
Unit of Social Ecology
Université libre de Bruxelles, CP 231
Boulevard du Triomphe
B-1050 Brussels
Belgium

 http://www.ulb.ac.be/sciences/use/bologna.html

	[[alternative HTML version deleted]]