I send a special note of thanks to Doug Altman,Robin
Prescott, Pierluigi Ballabeni, Maurille Feudjo and
Hans Hockey for their valuable advice regarding the
problem.
Ndema.
> Dear allstatters,
>
> How can one interprete a case, when relative risk
> (hazard ratio) estimate of a continuous explanatory
> variable in a multivariate Cox PH model being
> approximately equal to 1 i.e 0.9999 or 1.0001 etc
> given that the p-value shows significance i.e <
0.05.
> Also, the relative risk estimate for this variable
is
> approximately equal to the lower limit for its 95%
CI
> for relative risk.
>
__________________________________________________
Do you Yahoo!?
Yahoo! Platinum - Watch CBS' NCAA March Madness, live on your desktop!
http://platinum.yahoo.com
|