[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: Survival analysis - comparing a subset of the sample to the entire sample
It seems to me that what you should be doing is comparing the RFS in
patients with genotype information to that in patients without information.
Then your difficulties disappear.
At 06:53 AM 4/09/03 -0400, you wrote:
>I have a dataset with 1200 patients and time to event (relapse and
>death). Of these patients, 500 have data on a particular genotype, of
>which the variant allele is associated with an increased relapse risk.
>I would like to compare the relapse free survival estimate (RFS) of the
>overall study group to the RFS estimate of the patients with genotypes.
> That is, I want to show that the patients with genotype information
>(both wildtype and variant genotypes combined) do not overall have a
>significantly different risk of relapse than all patients on the trial.
>Thus, I have 2 questions:
>1. How do I do this in STATA?
>2. Are there issues of lack of independance in the data that make this
>comparison problematic from a statistical standpoint?
>Thank you very much in advance.
>Richard Aplenc, MD, MSCE
>Pediatric Oncology/Stem Cell Transplant
>Children's Hospital of Philadelphia
>Center for Clinical Epidemiology and Biostatistics
>University of Pennsylvania
>* For searches and help try:
* For searches and help try: