I used the following stset command. Then I merged a dataset which
contains multiple records per person which contains the vars after _t0.
I would like to split the data at the points where cross_over = 1. This
represents that an individual switched from non-dual to dual user by our
definition or vice-versa (dual to non-dual user). As you can see this
can happen multiple times in the study period (i.e. at 51, 232, and 260
days). So for this individual, the data would be split 3 times
resulting in intervals of 0-51, 51-232, 232-260, and 260-365.25. This
is what I would ideally like to see but I'm having trouble with the
stsplit syntax.
stset days, id(study_id) failure(died_1yr == 1) scale(365.25)
study_id died_1yr days _st _d _t
_t0 dual_15_85_12mth va_pcnt dual_15_85_claims
dx_to_claim_day cross_over
1 0 365.25 1 0 1 0
1 0 0 0
0
1 0 365.25 1 0 1 0
1 0 0 2
0
1 0 365.25 1 0 1 0
1 0 0 7
0
1 0 365.25 1 0 1 0
1 0.2 1 51
1
1 0 365.25 1 0 1 0
1 0.16666667 1 113
0
1 0 365.25 1 0 1 0
1 0.14285714 0 232
1
1 0 365.25 1 0 1 0
1 0.125 0 243
0
1 0 365.25 1 0 1 0
1 0.22222222 1 260
1
1 0 365.25 1 0 1 0
1 0.2 1 301
0
1 0 365.25 1 0 1 0
1 0.18181818 1 356
0
Tom Weichle
Math Statistician
Center for Management of Complex Chronic Care (CMC3)
Hines VA Hospital, Bldg 1, C202
708-202-8387 ext. 24261
[email protected]
*
* 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/
