1
$\begingroup$
library(survival)
s <- with(lung,Surv(time,status))
fKM <- survfit(s ~ 1,data=lung)
summary(fKM)
> summary(fKM)
Call: survfit(formula = s ~ 1, data = lung)

 time n.risk n.event survival std.err lower 95% CI upper 95% CI
    5    228       1   0.9956 0.00438       0.9871        1.000
   11    227       3   0.9825 0.00869       0.9656        1.000
   12    224       1   0.9781 0.00970       0.9592        0.997
   13    223       2   0.9693 0.01142       0.9472        0.992
   15    221       1   0.9649 0.01219       0.9413        0.989
   26    220       1   0.9605 0.01290       0.9356        0.986
   30    219       1   0.9561 0.01356       0.9299        0.983
   31    218       1   0.9518 0.01419       0.9243        0.980
   53    217       2   0.9430 0.01536       0.9134        0.974
   54    215       1   0.9386 0.01590       0.9079        0.970
   59    214       1   0.9342 0.01642       0.9026        0.967
   60    213       2   0.9254 0.01740       0.8920        0.960
   61    211       1   0.9211 0.01786       0.8867        0.957
   62    210       1   0.9167 0.01830       0.8815        0.953
   65    209       2   0.9079 0.01915       0.8711        0.946
   71    207       1   0.9035 0.01955       0.8660        0.943
   79    206       1   0.8991 0.01995       0.8609        0.939
   81    205       2   0.8904 0.02069       0.8507        0.932
   88    203       2   0.8816 0.02140       0.8406        0.925
   92    201       1   0.8772 0.02174       0.8356        0.921
   93    199       1   0.8728 0.02207       0.8306        0.917
   95    198       2   0.8640 0.02271       0.8206        0.910
  105    196       1   0.8596 0.02302       0.8156        0.906
  107    194       2   0.8507 0.02362       0.8056        0.898
  110    192       1   0.8463 0.02391       0.8007        0.894
  116    191       1   0.8418 0.02419       0.7957        0.891
  118    190       1   0.8374 0.02446       0.7908        0.887
  122    189       1   0.8330 0.02473       0.7859        0.883
  131    188       1   0.8285 0.02500       0.7810        0.879
  132    187       2   0.8197 0.02550       0.7712        0.871
  ... omitted ... 

Here I have a very simple KM fit. Suppose that I only had knowledge of the total number of individuals at risk in the beginning, i.e. n.risk = 228 and the survival probability at several time points, let's say, at time = 60, survival = 0.9254, and at time = 110, survival = 0.8463. In that case, how would I be able to obtain an estimate of the corresponding n.risk at time = 60 and time = 110? Is this even possible?

I've tried calculating 0.9254 * 228 and 0.8463 * 228, but those do not give me the corresponding n.risk I'm looking for.

$\endgroup$
  • $\begingroup$ How about solving $\frac{227 - n}{228-n} = 0.9254 * \frac{228}{227}$ for time $60$? This gives a close answer.. although I assumed $dN(u) = 1$ $\endgroup$ – moreblue Oct 21 '18 at 11:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.