st: command stsplit and time dependent factor

["Alessio Signori" <S2734525@studenti.unige.it>]

Dear all,

i have  a db with some categorical time-dependent factor and some continuous
covariate.

I used Stsplit to split into two episodes to obtain people at risk for the
two factors after one determined value in the time obtaining:

Upndalli     outcome  follow_u     _st     _d    _t    _t0
1222           2         76          1      0    76     15
1222           .         15          1      0    15      0
1234           2         83          1      0    83      7
1234           .          7          1      0     7      0
1240           2        126          1      0   126     21
1240           .         21          1      0    21      0
1243           2         75          1      0    75     22
1243           .         22          1      0    22      0
1263           2         36          1      0    36      0
1267           2          9          1      0     9      0
1269           0       3748          1      0  3748    100
1269           .        100          1      0   100     15
1269           .         15          1      0    15      0

Where upndalli is id of the patients outcome is 0 for censoring and two for
events.

My question is how function stcox (after have determined in stset .,
id(upndalli)) in a similar situation? There is some problem if I run stcox
on this database for continuous covariates no time-dependent?Particularly
with missing values created in outcome and relative time points.

Thanks to all

Alessio

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/