Hi Dimitrios:
First approach is definitely WRONG. Take a look at the paper
"Inappropriate Use of Bivariable Analysis to Screen Risk Factors for Use
in Multivariable Analysis", J Clin Epidemiol Vol. 49, No. 8, pp.
907-916, 1996.
Regarding the second approach, you should include ALL relevant
variables: known risk factors of death in kidney transplanted patients,
age & gender as usual suspects, other potential confounders, effect
modifiers ("is the effect of NODAT on survival modified by other
factors?")... If any of them violate the PH assumption, there are
statistical methods to control that (stratification, time-dependent
covariate Cox model...), but an important predictor should not be left
out of the model just because the PH assumption is not met.
Regression models are built for two main purposes: prediction and
explanation (see "To Explain or to Predict?", Statistical Science
2010, Vol. 25, No. 3, 289–310). You should know what type of model you
are trying to build before attempting to construct it.
HTH,
Marta GG
El 16/04/2014 13:22, dimitrios escribió:
> Dear All,
>
> I would be grateful to you if you could help me with the following.
>
> I am doing a cox regression analysis regarding patient survival after kidney
> transplantation aiming to explore the effect of NODAT.
>
> Most of the papers report one of the following:
>
> 1. They do univariate analysis and then proceed with multivariate cox
> regression analysis including only variables which were significant in the
> univariate analysis. They do not report about assumptions.
>
> 2. They include in the model all the variables which did not violate the PH
> assumptions.
>
> Which approach is considered the ideal one?
>
> Thank you in advance,
> Dimitris
>
>
>
> --
> View this message in context:
http://spssx-discussion.1045642.n5.nabble.com/Cox-regression-analysis-question-tp5725476.html> Sent from the SPSSX Discussion mailing list archive at Nabble.com.
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
>
[hidden email] (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD
>
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
Locked
