Mismatch between DEATHTIME and OUTTIME

[hugoych]

Hi,
Thanks a lot for your work, I really appreciate it!
When using multi-tasking, there are some samples that have stays that go beyond their DEATHTIME.

An example of this is ID 24603 as shown below.

I believe this leads to inconsistency between the decompensation label and the LOS label.
A fix would be to consider as OUTTIME the minimum value between OUTTIME and DEATHTIME and compute LOS from theirs.

Hope this helps!