R help - Sep 2004 - Cox regression for prevalence estimates

Hello, I'm an MD working in an eye clinic. I'm learning by myself to use
R
for use in my research works and for implementation in a software project.
There are some authors who recomends the use of Cox regression as a
substitute for Logistic regression (<a
href="http://www.biomedcentral.com/1471-2288/3/21.pdf"> Barros AJD,
Hirakata
VN. BMCMedical Research Methodology, 2003; 3:21 </a>. The use of Cox
regression permit the estimation of the prevalence rates rather than Odds
ratios obtained by logistic regression analysis. 
Cox regression is used for time-to-event data. To obtain prevalence rates
the time has to be constant. One of the problems of Cox regression is that
the confidence intervals are overestimated. To correct this Barros &
Hirakata recommend the use of robust variance estimates.
How can R be used to calculate the prevalence ratios using Cox regression +
robust variance estimates ?

Thanks for your collaboration,

Tomas Karpati MD
The Michaelson Institute for Rehabilitation of Vision
Hadassah Medical Organization

At , The Michaelson Institute wrote:
>How can R be used to calculate the prevalence ratios using Cox regression +
>robust variance estimates ?
Well,

In Design package have a command: cph
This command have a option "robsut" with default=FALSE, but in help is
write:
" ... robust if TRUE a robust variance estimate is returned. ..."

I think that is your response...


bye

Bernardo Rangel Tura, MD, MSc
National Institute of Cardiology Laranjeiras
Rio de Janeiro Brazil

At 09:27 07/09/2004, you wrote:
>Bernardo Rangel Tura wrote:
>>At , The Michaelson Institute wrote:
>>
>>>How can R be used to calculate the prevalence ratios using Cox
regression +
>>>robust variance estimates ?
>>
>>Well,
>>In Design package have a command: cph
>>This command have a option "robsut" with default=FALSE, but in
help is write:
>>" ... robust if TRUE a robust variance estimate is returned.
..."
>>I think that is your response...
>>
>>bye
>>Bernardo Rangel Tura, MD, MSc
>>National Institute of Cardiology Laranjeiras
>>Rio de Janeiro Brazil
>
>No, robust is an option to coxph, not cph.  cph uses 'after the fit'
>correction using the robcov or bootcov functions in Design.
>
>
>--
>Frank E Harrell Jr   Professor and Chair           School of Medicine
>                      Department of Biostatistics   Vanderbilt University
Sorry Frank!
I made a digitation mistake, but I think yours package answers the Tomas 
Karpati?s need.


Thanks in advance

Bernardo Rangel Tura, MD, MSc
National Institute of Cardiology Laranjeiras
Rio de Janeiro Brazil

robust standard errors are available from coxph with the option
robust=TRUE.

	-thomas

On 6 xxx -1, The Michaelson Institute wrote:
> Hello, I'm an MD working in an eye clinic. I'm learning by myself
to use R
> for use in my research works and for implementation in a software project.
> There are some authors who recomends the use of Cox regression as a
> substitute for Logistic regression (<a
> href="http://www.biomedcentral.com/1471-2288/3/21.pdf"> Barros
AJD, Hirakata
> VN. BMCMedical Research Methodology, 2003; 3:21 </a>. The use of Cox
> regression permit the estimation of the prevalence rates rather than Odds
> ratios obtained by logistic regression analysis.
> Cox regression is used for time-to-event data. To obtain prevalence rates
> the time has to be constant. One of the problems of Cox regression is that
> the confidence intervals are overestimated. To correct this Barros &
> Hirakata recommend the use of robust variance estimates.
> How can R be used to calculate the prevalence ratios using Cox regression +
> robust variance estimates ?
>
> Thanks for your collaboration,
>
> Tomas Karpati MD
> The Michaelson Institute for Rehabilitation of Vision
> Hadassah Medical Organization
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
>
Thomas Lumley			Assoc. Professor, Biostatistics
tlumley at u.washington.edu	University of Washington, Seattle

Thnks to Prof. Frank E Harrell Jr , Prof. Bernardo Rangel Tura, and Prof. 
Thomas Lumley for their response.

Tomas Karpati (MD)
The Michaelson Institute for the prevention of Blindness
Hadassah Medical Org.
email: hadasa704 at 012.net.il
Phone: +972-2-6256458, fax: +972-2-6232895


On Tue, 07 Sep 2004 10:11:32 -0300, you wrote:
>At 09:27 07/09/2004, you wrote:
>
>>Bernardo Rangel Tura wrote:
>>>At , The Michaelson Institute wrote:
>>>
>>>>How can R be used to calculate the prevalence ratios using Cox
regression +
>>>>robust variance estimates ?
>>>
>>>Well,
>>>In Design package have a command: cph
>>>This command have a option "robsut" with default=FALSE,
but in help is write:
>>>" ... robust if TRUE a robust variance estimate is returned.
..."
>>>I think that is your response...
>>>
>>>bye
>>>Bernardo Rangel Tura, MD, MSc
>>>National Institute of Cardiology Laranjeiras
>>>Rio de Janeiro Brazil
>>
>>No, robust is an option to coxph, not cph.  cph uses 'after the
fit'
>>correction using the robcov or bootcov functions in Design.
>>
>>
>>--
>>Frank E Harrell Jr   Professor and Chair           School of Medicine
>>                      Department of Biostatistics   Vanderbilt
University
>
>Sorry Frank!
>I made a digitation mistake, but I think yours package answers the Tomas 
>Karpati´s need.
>
>
>Thanks in advance
>
>Bernardo Rangel Tura, MD, MSc
>National Institute of Cardiology Laranjeiras
>Rio de Janeiro Brazil  
>