Dear Statalist -
This question is more of a general stats question than a Stata question,
but I am hoping someone might be able to help. I am running a cox model on
data that are clustered by study site. I suspect that there is correlation
among subjects within site. Generally, I would just use the cluster option
to account for this lack of independence. I have been reading in the
manual about shared frailty though, and am now confused about the
appropriate way to handle this situation:
(From the Survival Analysis manual section on cox model with shared frailty)
"One solution would be to fit a standard Cox model, adjusting the standard
errors of the estimated hazard ratios to account for the possible
correlation by specifying cluster(patient).
Alternatively, one can model the correlation by assuming that the
correlation is the result of a latent patient-level effect, or frailty.
That is, rather than fitting a standard model and specifying
cluster(patient), fit a frailty model by specifying shared(patient)."
(p. 129)
I get very different results with each of these solutions, so I am trying
to figure out which one to use. Is one of these the preferred method of
handling design effects and, if so, why? How does one decide which method
to use?
Thank you, thank you.
Sarah
Sarah A. Mustillo, Ph.D
Department of Psychiatry and Behavioral Sciences
Duke University School of Medicine
Box 3454
Durham NC 27710
919 687-4686 x241
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
