CG writes:
>I know of stcrreg for competing risks and how to use stcox in a multiple
>failure setting, but no clue how to "merge" both.
I suspect given how recent the theory behind stcrreg is, any
extensions to multiple failure would be novel territory. I'd be
tempted to run multiple stcrreg analyses; one for the first failure,
another for the second, etc. It's downgrading your data, but at least
you'll be on solid statistical ground.
This might also be useful, although a totally different approach it
might still address your study aims.
Biometrics. 1985 Mar;41(1):49-54."Case-control methods in the presence
of multiple failure times and competing risks" Lubin JH.
Andrew Lover
___________________________________________________________
Epidemiologist
Centre for Infectious Disease Epidemiology Research (CIDER)
Saw Swee Hock School of Public Health
National University of Singapore
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/
