Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Survival analysis - specifying age as time


From   Hugh Davies <hdavies@interchange.ubc.ca>
To   statalist@hsphsun2.harvard.edu
Subject   st: Survival analysis - specifying age as time
Date   Sun, 10 Apr 2005 23:10:48 -0700

Title: Survival analysis - specifying age as time
Greetings,

My apologies for a rather simple question.....

I am evaluating the effectiveness of an intervention on the incidence of occupational noise-induced hearing loss.

I am conducting a survival analysis, in which I wish to examine time-to-hearing-loss in my experimental vs. control groups. My hypothesis is that time-to-hearing loss will be longer in the experimental group.

Because age is major risk factor for hearing loss, I wish to use age as my “time” variable.

However, subjects only become at risk after they have had their first exposure to noise. Therefore I have entered individuals age-at-first-exposure as their origin:

STSET AGE_AT_TEST, FAILURE(HL_EVENT) ORIGIN(AGE_AT_FIRST_NOISE_EXP) ID(SUBJ_ID) SCALE(365.25)

But if t=(time-origin)/scale, as the reference manual says – doesn’t that mean I am no longer controlling for age as I wished?

I would very much appreciate assistance in specifying this problem.

Regards,

Hugh

University of British Columbia



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