Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: How to treat variables where all outcomes happens in one interval-apology


From   ssamuels@albany.edu
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: How to treat variables where all outcomes happens in one interval-apology
Date   Thu, 2 Apr 2009 15:08:04 -0400 (EDT)

--

I apologize for the extended subject of my last post.  I haven't been able
to get through to the list and sent it from webmail, but didn't notice I'd
copied too much from my clipboard.  Here it is again.

-Steve



Roland-

When categories with events are compared to categories with no events in a
Cox model, the  partial likelihood is maximized by a beta coefficient of
plus or minus of infinity, giving you the "very large HR" you observed or
to HR = 0.  The same phenomenon would occur if you had a continuous
covariate whose rank correlation with failure time was 1.0.

A similar problem arises in estimating an odds ratio in a 2 x 2 table when
one of the off-diagonal cells has no observations.

If you wish to use Cox, you cannot compare age >45 to age <=45. You cannot
make a comparison involving any stage defined, in part, by having age<=45.
 You may have to exclude all people <=45 and take whatever stages remain.

Try to obtain from the literature information about the distribution of
deaths by age. A sample size calculation (-stpower-) should show why you
observed none in the <=45 group.

-Steve



I am analysing survival in two methods of syrgery for thyroid cancer.
The international classification of stage of disease includes
tumorsize (<2, 2-4, >4 cm within the thyroid and growth outside the
thyroid, presence of distant metastases, metastases to lymphglands and
age>45 years.
In my patients all deaths have occured in patients >age 45 years. When
the dichotomised agevariable is analysed in Coxregression the HR is
very large with very large SE. There is no problem with collinearity.
How should I treat this situation? One solution would be to only
analyse according to the stage classification (which includes age >45
years for stage 3 and 4), but I would like to analyse the importance
of each element of the stageclassification. I may dichotomise with
cutoff point >50 years, but that is not correct according to the
international definition of tumour stage.


I am analysing survival in two methods of syrgery for thyroid cancer.
The international classification of stage of disease includes
tumorsize (<2, 2-4, >4 cm within the thyroid and growth outside the
thyroid, presence of distant metastases, metastases to lymphglands and
age>45 years.
In my patients all deaths have occured in patients >age 45 years. When
the dichotomised agevariable is analysed in Coxregression the HR is
very large with very large SE. There is no problem with collinearity.
How should I treat this situation? One solution would be to only
analyse according to the stage classification (which includes age >45
years for stage 3 and 4), but I would like to analyse the importance
of each element of the stageclassification. I may dichotomise with
cutoff point >50 years, but that is not correct according to the
international definition of tumour stage.

*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index