Help with stset probable error

[Omotayo Adeniyi]

Please kindly advice me, I am try to do a survival analysis on stroke recurrence, after stset the data, I got the folllowing message:

stset survy died

failure event: died != 0 & died < .
obs. time interval: (0, survy]
exit on or before: failure


2394 total observations
1110 event time missing (survy>=.) PROBABLE ERROR
13 observations end on or before enter()

1271 observations remaining, representing
1271 failures in single-record/single-failure data
2947.978 total analysis time at risk and under observation
at risk from t = 0
earliest observed entry t = 0
last observed exit t = 13.44011

Please does any one know what I need to do to get rid of event time missing data. Thank you

[Christophe Kolodziejczyk]

If survy is the observed survival time and died an indicator for failure I would code

stset survy , f(died)

It would be good idea to show us the structure of your data (by giving us an example or an excerpt) in order to understand why you have missing survival times. I would eliminate these observations unless there is a good reason for it. You can also code

stset survy if survy<. , f(died)

Moreover I would like to draw your attention to the fact that you have observations with zero survival time. What does that mean in your case? Are these observations at risk or not? Normally observations with 0 survival time are irrelevant because not at risk and it is how stset interprets zero survival time. It what the message

13 observations end on or before enter()

indicates (see help stset). But maybe they have been at risk for a very short period of time which makes you code them with survival time = 0 depending on how you measure units of time (days or weeks).

[Omotayo Adeniyi]

Thank you very much for you advice. This is an excerpt from the data. The variable died was not in the original data, I created it myself from Survy variable generate command, the missing 1100 participants are censored observation-those that are still alive, I am concerned if I excluded it from analysis it might affect my survival function. Thank you
survy rec died _st _d _t _t0

7.701574 1
3.477071 1
.7939767 1
. 0
1.765914 1
2.992471 1
4.750171 1
. 0
.0054757 1