Data with 40 eyes and 20 subjects. Plan to do a survival curve (KM curve). Question is how to cluster the eyes. I tried using the same ID for the same subjects. But the thing is, for few subjects the begin time is "0" i.e. time0=0 and the end time is say for example 2 months (i.e. event happened at month2). While running the below command
stset time, fail(outcome) id(ID)
it excludes the observations for the subjects (both eyes) with same start time and end time. what is the option to include that both eyes with same study time while clustering between eyes?
If I understand your question use 0,1 coding for the two factors. Good luck, D Booth
Thank you David Eugene Booth for you response !
I think my question is not so clear. Try to explain below by an example dataset;
ID Name eye outcome time_start time_end
1 A RE 0 0 24
2 B RE 1 0 18
3 C RE 1 0 12
3 C LE 0 0 12
Label:
eye - RE "Right eye"; LE "Left eye"
outcome - 0 "Success" 1 "Failure"
time_start & time_end - in months
For ID 1,2 the subjects with single eye and for ID 3 subjects were both eye patient. Question is how do i cluster between eyes (RE - right eye; LE - Left eye) in KM curve, possibly in STATA. Because of the same start time and end time for ID 3, stset command with id(ID) excludes the subject "C".
CHECK THIS PAPER
Sarkar, K., Chowdhury, R., & Dasgupta, A. (2017). Analysis of Survival Data: Challenges and Algorithm-Based Model Selection. Journal of clinical and diagnostic research : JCDR, 11(6), LC14–LC20. https://doi.org/10.7860/JCDR/2017/21903.10019
Mohammed Sithiq It appears that you are miscodeing your data . I don't know what you want in your set so i can't fix it. You need to speak to a local statistical consultant. Note in your example you appear to have 3 Right eyes and 1 left but
you can't use the same ID for more than 1 EYE with your format statement. You either have to send your code and your data or see the consultant Which I recommend. you must get an error message-what is it ? what gets printed out - anything.?best, D. Booth
Thank you Oyekola Oluyimika Oloyede for you response !
I have got some idea from the paper that you recommended. It explain about the models that used to cluster based on different types. But still looking, is that a way for clustering eyes in KM graphs?
There must be a biological correlation between eyes (right eye and left eye) of a single subject. So the ID variable was created as unique for the subject with one eye (i.e. for patient 1 & 2)and entered two times to mention that subject was both eye patient (ID number 3).
I planned to do survival analysis (need to find the cumulative probability of success & its KM curve to project visually). Because of the correlation between eyes of a single subject, how can i do clustering between eyes.
I understood models like stcox with clustering ID variable, stmixed (i.e. multi level mixed effect modelling, .... were there to predict by clustering between eyes (i.e. adjusting the correlated information of the same subject).
What am i looking is, Is there a way to cluster between eyes in doing KM graph?
Thank you David Eugene Booth
- Badges
- Science topic
- Similar topics
- Biological Science
- Kinesiology
- Running