Relative Risk in logistic regression

classic Classic list List threaded Threaded
17 messages Options
Reply | Threaded
Open this post in threaded view
|

Relative Risk in logistic regression

aminreza Aamini
Hi all,
I am very grateful to all those who write to me
1) how i  can  obtain relative risk (risk ratio) in logistic regression in R.
2) how to obtain  the predicted risk for a certain individual using
fitted regression model in R.

Many thanks, in advance, for your help.

Amin.

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: Relative Risk in logistic regression

nalluri pratap
Relative risk = exp(coef(model))

--- On Wed, 30/1/13, aminreza Aamini <[hidden email]> wrote:


From: aminreza Aamini <[hidden email]>
Subject: [R] Relative Risk in logistic regression
To: "R-help" <[hidden email]>
Date: Wednesday, 30 January, 2013, 4:19 PM


Hi all,
I am very grateful to all those who write to me
1) how i  can  obtain relative risk (risk ratio) in logistic regression in R.
2) how to obtain  the predicted risk for a certain individual using
fitted regression model in R.

Many thanks, in advance, for your help.

Amin.

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

        [[alternative HTML version deleted]]


______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: Relative Risk in logistic regression

nalluri pratap
In reply to this post by aminreza Aamini
Example from linear regression help (?lm)
 
ctl <- c(4.17,5.58,5.18,6.11,4.50,4.61,5.17,4.53,5.33,5.14)
trt <- c(4.81,4.17,4.41,3.59,5.87,3.83,6.03,4.89,4.32,4.69)
group <- gl(2,10,20, labels=c("Ctl","Trt"))
weight <- c(ctl, trt)
lm.D9 <- lm(weight ~ group)
lm.D90 <- lm(weight ~ group - 1)
 
prediction_error= lm.D90$residuals

--- On Wed, 30/1/13, aminreza Aamini <[hidden email]> wrote:


From: aminreza Aamini <[hidden email]>
Subject: [R] Relative Risk in logistic regression
To: "R-help" <[hidden email]>
Date: Wednesday, 30 January, 2013, 4:19 PM


Hi all,
I am very grateful to all those who write to me
1) how i  can  obtain relative risk (risk ratio) in logistic regression in R.
2) how to obtain  the predicted risk for a certain individual using
fitted regression model in R.

Many thanks, in advance, for your help.

Amin.

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

        [[alternative HTML version deleted]]


______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: Relative Risk in logistic regression

Kevin Thorpe
In reply to this post by nalluri pratap
On 01/30/2013 09:02 AM, nalluri pratap wrote:
> Relative risk = exp(coef(model))
>

Only if you fit using the log link.  Using the logit link, this gives
odds ratios.

> --- On Wed, 30/1/13, aminreza Aamini <[hidden email]> wrote:
>
>
> From: aminreza Aamini <[hidden email]>
> Subject: [R] Relative Risk in logistic regression
> To: "R-help" <[hidden email]>
> Date: Wednesday, 30 January, 2013, 4:19 PM
>
>
> Hi all,
> I am very grateful to all those who write to me
> 1) how i  can  obtain relative risk (risk ratio) in logistic regression in R.
> 2) how to obtain  the predicted risk for a certain individual using
> fitted regression model in R.
>
> Many thanks, in advance, for your help.
>
> Amin.
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
> [[alternative HTML version deleted]]
>
>
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>


--
Kevin E. Thorpe
Head of Biostatistics,  Applied Health Research Centre (AHRC)
Li Ka Shing Knowledge Institute of St. Michael's
Assistant Professor, Dalla Lana School of Public Health
University of Toronto
email: [hidden email]  Tel: 416.864.5776  Fax: 416.864.3016

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: Relative Risk in logistic regression

John Sorkin
I am not sure why one would want a relative risk from a logistic regression. The measure of association from a logistic regression is the odds ratio, not the relative risk.
John

 
John David Sorkin M.D., Ph.D.
Chief, Biostatistics and Informatics
University of Maryland School of Medicine Division of Gerontology
Baltimore VA Medical Center
10 North Greene Street
GRECC (BT/18/GR)
Baltimore, MD 21201-1524
(Phone) 410-605-7119
(Fax) 410-605-7913 (Please call phone number above prior to faxing)>>> "Kevin E. Thorpe" <[hidden email]> 1/30/2013 11:14 AM >>>
On 01/30/2013 09:02 AM, nalluri pratap wrote:
> Relative risk = exp(coef(model))
>

Only if you fit using the log link.  Using the logit link, this gives
odds ratios.

> --- On Wed, 30/1/13, aminreza Aamini <[hidden email]> wrote:
>
>
> From: aminreza Aamini <[hidden email]>
> Subject: [R] Relative Risk in logistic regression
> To: "R-help" <[hidden email]>
> Date: Wednesday, 30 January, 2013, 4:19 PM
>
>
> Hi all,
> I am very grateful to all those who write to me
> 1) how i  can  obtain relative risk (risk ratio) in logistic regression in R.
> 2) how to obtain  the predicted risk for a certain individual using
> fitted regression model in R.
>
> Many thanks, in advance, for your help.
>
> Amin.
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
> [[alternative HTML version deleted]]
>
>
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>

--
Kevin E. Thorpe
Head of Biostatistics,  Applied Health Research Centre (AHRC)
Li Ka Shing Knowledge Institute of St. Michael's
Assistant Professor, Dalla Lana School of Public Health
University of Toronto
email: [hidden email]  Tel: 416.864.5776  Fax: 416.864.3016

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Confidentiality Statement:
This email message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information.  Any unauthorized use, disclosure or distribution is prohibited.  If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message.

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: Relative Risk in logistic regression

Kevin Thorpe
On 01/30/2013 11:17 AM, John Sorkin wrote:
> I am not sure why one would want a relative risk from a logistic
> regression. The measure of association from a logistic regression is the
> odds ratio, not the relative risk.
> John
>

Yes, the natural measure, when using the logit link, is the OR.  I
always use that myself, but if you use the log link, you get RR.  As for
why you would, I would suggest it's because physicians think they
understand a RR better than and OR.


> "Kevin E. Thorpe" <[hidden email]> 1/30/2013 11:14 AM >>>
> On 01/30/2013 09:02 AM, nalluri pratap wrote:
>  > Relative risk = exp(coef(model))
>  >
>
> Only if you fit using the log link.  Using the logit link, this gives
> odds ratios.
>
>  > --- On Wed, 30/1/13, aminreza Aamini <[hidden email]> wrote:
>  >
>  >
>  > From: aminreza Aamini <[hidden email]>
>  > Subject: [R] Relative Risk in logistic regression
>  > To: "R-help" <[hidden email]>
>  > Date: Wednesday, 30 January, 2013, 4:19 PM
>  >
>  >
>  > Hi all,
>  > I am very grateful to all those who write to me
>  > 1) how i  can  obtain relative risk (risk ratio) in logistic
> regression in R.
>  > 2) how to obtain  the predicted risk for a certain individual using
>  > fitted regression model in R.
>  >
>  > Many thanks, in advance, for your help.
>  >
>  > Amin.
>  >
>  > ______________________________________________
>  > [hidden email] mailing list
>  > https://stat.ethz.ch/mailman/listinfo/r-help
>  > PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
>  > and provide commented, minimal, self-contained, reproducible code.
>  >
>  > [[alternative HTML version deleted]]
>  >
>  >
>  >
>  > ______________________________________________
>  > [hidden email] mailing list
>  > https://stat.ethz.ch/mailman/listinfo/r-help
>  > PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
>  > and provide commented, minimal, self-contained, reproducible code.
>  >
>
>


--
Kevin E. Thorpe
Head of Biostatistics,  Applied Health Research Centre (AHRC)
Li Ka Shing Knowledge Institute of St. Michael's
Assistant Professor, Dalla Lana School of Public Health
University of Toronto
email: [hidden email]  Tel: 416.864.5776  Fax: 416.864.3016

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: Relative Risk in logistic regression

John Sorkin
If you use a log link, you are not, I believe, performing a logistic regression!

 
John David Sorkin M.D., Ph.D.
Chief, Biostatistics and Informatics
University of Maryland School of Medicine Division of Gerontology
Baltimore VA Medical Center
10 North Greene Street
GRECC (BT/18/GR)
Baltimore, MD 21201-1524
(Phone) 410-605-7119
(Fax) 410-605-7913 (Please call phone number above prior to faxing)>>> "Kevin E. Thorpe" <[hidden email]> 1/30/2013 11:22 AM >>>
On 01/30/2013 11:17 AM, John Sorkin wrote:
> I am not sure why one would want a relative risk from a logistic
> regression. The measure of association from a logistic regression is the
> odds ratio, not the relative risk.
> John
>

Yes, the natural measure, when using the logit link, is the OR.  I
always use that myself, but if you use the log link, you get RR.  As for
why you would, I would suggest it's because physicians think they
understand a RR better than and OR.


> "Kevin E. Thorpe" <[hidden email]> 1/30/2013 11:14 AM >>>
> On 01/30/2013 09:02 AM, nalluri pratap wrote:
>  > Relative risk = exp(coef(model))
>  >
>
> Only if you fit using the log link.  Using the logit link, this gives
> odds ratios.
>
>  > --- On Wed, 30/1/13, aminreza Aamini <[hidden email]> wrote:
>  >
>  >
>  > From: aminreza Aamini <[hidden email]>
>  > Subject: [R] Relative Risk in logistic regression
>  > To: "R-help" <[hidden email]>
>  > Date: Wednesday, 30 January, 2013, 4:19 PM
>  >
>  >
>  > Hi all,
>  > I am very grateful to all those who write to me
>  > 1) how i  can  obtain relative risk (risk ratio) in logistic
> regression in R.
>  > 2) how to obtain  the predicted risk for a certain individual using
>  > fitted regression model in R.
>  >
>  > Many thanks, in advance, for your help.
>  >
>  > Amin.
>  >
>  > ______________________________________________
>  > [hidden email] mailing list
>  > https://stat.ethz.ch/mailman/listinfo/r-help
>  > PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
>  > and provide commented, minimal, self-contained, reproducible code.
>  >
>  > [[alternative HTML version deleted]]
>  >
>  >
>  >
>  > ______________________________________________
>  > [hidden email] mailing list
>  > https://stat.ethz.ch/mailman/listinfo/r-help
>  > PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
>  > and provide commented, minimal, self-contained, reproducible code.
>  >
>
>

--
Kevin E. Thorpe
Head of Biostatistics,  Applied Health Research Centre (AHRC)
Li Ka Shing Knowledge Institute of St. Michael's
Assistant Professor, Dalla Lana School of Public Health
University of Toronto
email: [hidden email]  Tel: 416.864.5776  Fax: 416.864.3016


Confidentiality Statement:
This email message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information.  Any unauthorized use, disclosure or distribution is prohibited.  If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message.

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: Relative Risk in logistic regression

Kevin Thorpe
On 01/30/2013 11:26 AM, John Sorkin wrote:
> If you use a log link, you are not, I believe, performing a logistic
> regression!

I guess strictly speaking, that is true.  I was being a little sloppy in
terminology.

>
> "Kevin E. Thorpe" <[hidden email]> 1/30/2013 11:22 AM >>>
> On 01/30/2013 11:17 AM, John Sorkin wrote:
>  > I am not sure why one would want a relative risk from a logistic
>  > regression. The measure of association from a logistic regression is the
>  > odds ratio, not the relative risk.
>  > John
>  >
>
> Yes, the natural measure, when using the logit link, is the OR.  I
> always use that myself, but if you use the log link, you get RR.  As for
> why you would, I would suggest it's because physicians think they
> understand a RR better than and OR.
>
>
>  > "Kevin E. Thorpe" <[hidden email]> 1/30/2013 11:14 AM >>>
>  > On 01/30/2013 09:02 AM, nalluri pratap wrote:
>  >  > Relative risk = exp(coef(model))
>  >  >
>  >
>  > Only if you fit using the log link.  Using the logit link, this gives
>  > odds ratios.
>  >
>  >  > --- On Wed, 30/1/13, aminreza Aamini <[hidden email]> wrote:
>  >  >
>  >  >
>  >  > From: aminreza Aamini <[hidden email]>
>  >  > Subject: [R] Relative Risk in logistic regression
>  >  > To: "R-help" <[hidden email]>
>  >  > Date: Wednesday, 30 January, 2013, 4:19 PM
>  >  >
>  >  >
>  >  > Hi all,
>  >  > I am very grateful to all those who write to me
>  >  > 1) how i  can  obtain relative risk (risk ratio) in logistic
>  > regression in R.
>  >  > 2) how to obtain  the predicted risk for a certain individual using
>  >  > fitted regression model in R.
>  >  >
>  >  > Many thanks, in advance, for your help.
>  >  >
>  >  > Amin.
>  >  >
>  >  > ______________________________________________
>  >  > [hidden email] mailing list
>  >  > https://stat.ethz.ch/mailman/listinfo/r-help
>  >  > PLEASE do read the posting guide
>  > http://www.R-project.org/posting-guide.html
>  >  > and provide commented, minimal, self-contained, reproducible code.
>  >  >
>  >  > [[alternative HTML version deleted]]
>  >  >
>  >  >


--
Kevin E. Thorpe
Head of Biostatistics,  Applied Health Research Centre (AHRC)
Li Ka Shing Knowledge Institute of St. Michael's
Assistant Professor, Dalla Lana School of Public Health
University of Toronto
email: [hidden email]  Tel: 416.864.5776  Fax: 416.864.3016

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: Relative Risk in logistic regression

David Winsemius
In reply to this post by aminreza Aamini

On Jan 30, 2013, at 5:49 AM, aminreza Aamini wrote:

> Hi all,
> I am very grateful to all those who write to me
> 1) how i  can  obtain relative risk (risk ratio) in logistic  
> regression in R.
> 2) how to obtain  the predicted risk for a certain individual using
> fitted regression model in R.

You obtain the predicted probabilities with something like:

predict(model, data.frame(x1="a", x2=30), type = "response")

See ?predict.glm

This would give the odds ratios (similar but larger than the risk  
ratios):

exp(coef(model))

--
David Winsemius, MD
Alameda, CA, USA

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: Relative Risk in logistic regression

Frank Harrell
I am curious why one would want risk ratios.  Unlike odds ratios, they are not interpretable without reference to the base risk.  For example a risk ratio of 2 cannot possibly apply to anyone with a starting risk exceeding 1/2.

I think it is most helpful to use one of the existing nomograms to show someone how the base risk and odds ratio translate to final risk, for a range of base risk.

Frank
David Winsemius wrote
On Jan 30, 2013, at 5:49 AM, aminreza Aamini wrote:

> Hi all,
> I am very grateful to all those who write to me
> 1) how i  can  obtain relative risk (risk ratio) in logistic  
> regression in R.
> 2) how to obtain  the predicted risk for a certain individual using
> fitted regression model in R.

You obtain the predicted probabilities with something like:

predict(model, data.frame(x1="a", x2=30), type = "response")

See ?predict.glm

This would give the odds ratios (similar but larger than the risk  
ratios):

exp(coef(model))

--
David Winsemius, MD
Alameda, CA, USA

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Frank Harrell
Department of Biostatistics, Vanderbilt University
Reply | Threaded
Open this post in threaded view
|

Re: Relative Risk in logistic regression

Ivan-K
Dear colleagues, I have 2 points: One opinion and one question.

1)
In one paper in a peer-reviewed journal, I read about the idea of using a logit regression as a surrogate for the log-binomial, just adding the numerator to the denominator ...
It’s tempting to immediately get the RR instead of OR ...
I tried it and I think it's a bad idea, the confidence intervals dramatically inflated!
Any opinions?

2)
What would be the criteria for selection of link - functions for binary data?
Usually I use the logit - just for simplest interpretation of parameters.
Using logit, probit, log-log, and log-log, I get identical values of the maximum-likelihood, Pearson statistics, overdispersion parameter, etc. However, the regression coefficients and its standard errors are different (for logit b is the maximum, for the probit – min., for log-log & C- log-log are between them). LRs close, but the maximum has the log-log. Wald criterion - the maximum for the probit.
?What are the interpretations for regression parameters (except logit)???

Ivan, IPAE RAS
Reply | Threaded
Open this post in threaded view
|

Re: Relative Risk in logistic regression

glsnow
Ivan,

In reference to your part 2), in 1989 Li and Duan published a paper where
they examined the effect of using the "wrong" link function (
http://projecteuclid.org/DPubS?service=UI&version=1.0&verb=Display&handle=euclid.aos/1176347254).
 The short version is that they found that in common models (lm and glm and
others) and when the x-variables meet certain conditions, then the
estimates of the slopes will change by a multiplicative constant as will
the variance co-variance matrix. Many of the tests are still well behaved
in this condition.  The link functions you mentioned are all very similar
to each other, so the impact of using a "wrong" one will be very minor.


On Fri, Feb 1, 2013 at 8:33 AM, Ivan-K <[hidden email]> wrote:

> Dear colleagues, I have 2 points: One opinion and one question.
>
> 1)
> In one paper in a peer-reviewed journal, I read about the idea of using a
> logit regression as a surrogate for the log-binomial, just adding the
> numerator to the denominator ...
> It’s tempting to immediately get the RR instead of OR ...
> I tried it and I think it's a bad idea, the confidence intervals
> dramatically inflated!
> Any opinions?
>
> 2)
> What would be the criteria for selection of link - functions for binary
> data?
> Usually I use the logit - just for simplest interpretation of parameters.
> Using logit, probit, log-log, and log-log, I get identical values of the
> maximum-likelihood, Pearson statistics, overdispersion parameter, etc.
> However, the regression coefficients and its standard errors are different
> (for logit b is the maximum, for the probit – min., for log-log & C-
> log-log
> are between them). LRs close, but the maximum has the log-log. Wald
> criterion - the maximum for the probit.
> ?What are the interpretations for regression parameters (except logit)???
>
> Ivan, IPAE RAS
>
>
>
>
> --
> View this message in context:
> http://r.789695.n4.nabble.com/Relative-Risk-in-logistic-regression-tp4657040p4657297.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>


--
Gregory (Greg) L. Snow Ph.D.
[hidden email]

        [[alternative HTML version deleted]]


______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: Relative Risk in logistic regression

Michael Dewey
In reply to this post by aminreza Aamini
At 10:49 30/01/2013, aminreza Aamini wrote:
>Hi all,
>I am very grateful to all those who write to me
>1) how i  can  obtain relative risk (risk ratio) in logistic regression in R.

@TECHREPORT{lumley06,
   author = {Lumley, T and Kronmal, R and Ma, S},
   year = 2006,
   title = {Relative risk regression in medical research: models, contrasts,
           estimators, and algorithms},
   number = 293,
   institution = {{UW} Biostatistics Working Paper Series},
   keywords = {glm, Poisson},
   url = {http://www.bepress.com/uwbiostat/paper293}
}


>2) how to obtain  the predicted risk for a certain individual using
>fitted regression model in R.
>
>Many thanks, in advance, for your help.
>
>Amin.

Michael Dewey
[hidden email]
http://www.aghmed.fsnet.co.uk/home.html

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: Relative Risk in logistic regression

aminreza Aamini
Dear Coleagues ,
As my friend John  mentined,* the measure of association from a logistic
regression is the odds ratio, not the relative risk*. but the point is in
follow-up studies, it is commonly preferred to estimate a risk ratio rather
than an odds ratio. Thats why im looking for RR in logistic models.
Bytheway thank you all for ur consideration.
Amin


On Sun, Feb 3, 2013 at 1:42 PM, Michael Dewey <[hidden email]>wrote:

> At 10:49 30/01/2013, aminreza Aamini wrote:
>
>> Hi all,
>> I am very grateful to all those who write to me
>> 1) how i  can  obtain relative risk (risk ratio) in logistic regression
>> in R.
>>
>
> @TECHREPORT{lumley06,
>   author = {Lumley, T and Kronmal, R and Ma, S},
>   year = 2006,
>   title = {Relative risk regression in medical research: models, contrasts,
>           estimators, and algorithms},
>   number = 293,
>   institution = {{UW} Biostatistics Working Paper Series},
>   keywords = {glm, Poisson},
>   url = {http://www.bepress.com/**uwbiostat/paper293<http://www.bepress.com/uwbiostat/paper293>
> }
>
> }
>
>
>  2) how to obtain  the predicted risk for a certain individual using
>> fitted regression model in R.
>>
>> Many thanks, in advance, for your help.
>>
>> Amin.
>>
>
> Michael Dewey
> [hidden email]
> http://www.aghmed.fsnet.co.uk/**home.html<http://www.aghmed.fsnet.co.uk/home.html>
>
>

        [[alternative HTML version deleted]]

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: Relative Risk in logistic regression

John Sorkin
Amin,
It is incorrect to use the relative risk as a measure of association in a logistic regression.  The measure of association in a logistic regression is the odds ratio. The odds ratio is an approximation of the relative risk. The approximation becomes progressively better as the disease becomes progressively rarer. Regardless of whether the disease is rare or not, inferences drawn from a logistic regression are valid. Please do not report a logistic regression using relative risk. It is not correct to do so.
John  

 
John David Sorkin M.D., Ph.D.
Chief, Biostatistics and Informatics
University of Maryland School of Medicine Division of Gerontology
Baltimore VA Medical Center
10 North Greene Street
GRECC (BT/18/GR)
Baltimore, MD 21201-1524
(Phone) 410-605-7119
(Fax) 410-605-7913 (Please call phone number above prior to faxing)>>> aminreza Aamini <[hidden email]> 2/3/2013 9:15 AM >>>
Dear Coleagues ,
As my friend John  mentined,* the measure of association from a logistic
regression is the odds ratio, not the relative risk*. but the point is in
follow-up studies, it is commonly preferred to estimate a risk ratio rather
than an odds ratio. Thats why im looking for RR in logistic models.
Bytheway thank you all for ur consideration.
Amin


On Sun, Feb 3, 2013 at 1:42 PM, Michael Dewey <[hidden email]>wrote:

> At 10:49 30/01/2013, aminreza Aamini wrote:
>
>> Hi all,
>> I am very grateful to all those who write to me
>> 1) how i  can  obtain relative risk (risk ratio) in logistic regression
>> in R.
>>
>
> @TECHREPORT{lumley06,
>   author = {Lumley, T and Kronmal, R and Ma, S},
>   year = 2006,
>   title = {Relative risk regression in medical research: models, contrasts,
>           estimators, and algorithms},
>   number = 293,
>   institution = {{UW} Biostatistics Working Paper Series},
>   keywords = {glm, Poisson},
>   url = {http://www.bepress.com/**uwbiostat/paper293<http://www.bepress.com/uwbiostat/paper293>
> }
>
> }
>
>
>  2) how to obtain  the predicted risk for a certain individual using
>> fitted regression model in R.
>>
>> Many thanks, in advance, for your help.
>>
>> Amin.
>>
>
> Michael Dewey
> [hidden email]
> http://www.aghmed.fsnet.co.uk/**home.html<http://www.aghmed.fsnet.co.uk/home.html>
>
>
[[alternative HTML version deleted]]

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.


Confidentiality Statement:
This email message, including any attachments, is for the sole use of the intended recipient(s) and may contain confidential and privileged information.  Any unauthorized use, disclosure or distribution is prohibited.  If you are not the intended recipient, please contact the sender by reply email and destroy all copies of the original message.

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: Relative Risk in logistic regression

David Winsemius
In reply to this post by aminreza Aamini

On Feb 3, 2013, at 8:15 AM, aminreza Aamini wrote:

> Dear Coleagues ,
> As my friend John  mentined,* the measure of association from a  
> logistic
> regression is the odds ratio, not the relative risk*. but the point  
> is in
> follow-up studies, it is commonly preferred to estimate a risk ratio  
> rather
> than an odds ratio. Thats why im looking for RR in logistic models.
> Bytheway thank you all for ur consideration.
> Amin

I agree that the relative risk is generally preferred in presenting  
the results of follow-up studies. The question should be:  why do you  
want to use a logistic link? The technical report out of the  
University of Washington Biostatistics Depeartment explains a variety  
of approaches including using a log-binomial model and Poisson  
regression. Either of those can be done in R with glm. The Poisson  
regression model is particularly simple to develop.  You should  
explain a)  what sort of data you have in greater detail and b) your  
reasons for using the logistic link when arguably better alternatives  
are available if you want a more specific answer.

--
David

>
>
> On Sun, Feb 3, 2013 at 1:42 PM, Michael Dewey  
> <[hidden email]>wrote:
>
>> At 10:49 30/01/2013, aminreza Aamini wrote:
>>
>>> Hi all,
>>> I am very grateful to all those who write to me
>>> 1) how i  can  obtain relative risk (risk ratio) in logistic  
>>> regression
>>> in R.
>>>
>>
>> @TECHREPORT{lumley06,
>>  author = {Lumley, T and Kronmal, R and Ma, S},
>>  year = 2006,
>>  title = {Relative risk regression in medical research: models,  
>> contrasts,
>>          estimators, and algorithms},
>>  number = 293,
>>  institution = {{UW} Biostatistics Working Paper Series},
>>  keywords = {glm, Poisson},
>>  url = {http://www.bepress.com/**uwbiostat/paper293<http://www.bepress.com/uwbiostat/paper293 
>> >
>> }
>>
>> }
>>
>>
>> 2) how to obtain  the predicted risk for a certain individual using
>>> fitted regression model in R.
>>>
>>> Many thanks, in advance, for your help.
>>>
>>> Amin.
>>>
>>
>> Michael Dewey
>> [hidden email]
>> http://www.aghmed.fsnet.co.uk/**home.html<http://www.aghmed.fsnet.co.uk/home.html 
>> >
>>
>>
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

David Winsemius, MD
Alameda, CA, USA

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Reply | Threaded
Open this post in threaded view
|

Re: Relative Risk in logistic regression

Frank Harrell
This post has NOT been accepted by the mailing list yet.
I can't quite agree.  For follow-up studies, as for case-control studies, I prefer relative odds for reasons cited earlier (transportability/applicability).

Frank
David Winsemius wrote
On Feb 3, 2013, at 8:15 AM, aminreza Aamini wrote:

> Dear Coleagues ,
> As my friend John  mentined,* the measure of association from a  
> logistic
> regression is the odds ratio, not the relative risk*. but the point  
> is in
> follow-up studies, it is commonly preferred to estimate a risk ratio  
> rather
> than an odds ratio. Thats why im looking for RR in logistic models.
> Bytheway thank you all for ur consideration.
> Amin

I agree that the relative risk is generally preferred in presenting  
the results of follow-up studies. The question should be:  why do you  
want to use a logistic link? The technical report out of the  
University of Washington Biostatistics Depeartment explains a variety  
of approaches including using a log-binomial model and Poisson  
regression. Either of those can be done in R with glm. The Poisson  
regression model is particularly simple to develop.  You should  
explain a)  what sort of data you have in greater detail and b) your  
reasons for using the logistic link when arguably better alternatives  
are available if you want a more specific answer.

--
David
>
>
> On Sun, Feb 3, 2013 at 1:42 PM, Michael Dewey  
> <[hidden email]>wrote:
>
>> At 10:49 30/01/2013, aminreza Aamini wrote:
>>
>>> Hi all,
>>> I am very grateful to all those who write to me
>>> 1) how i  can  obtain relative risk (risk ratio) in logistic  
>>> regression
>>> in R.
>>>
>>
>> @TECHREPORT{lumley06,
>>  author = {Lumley, T and Kronmal, R and Ma, S},
>>  year = 2006,
>>  title = {Relative risk regression in medical research: models,  
>> contrasts,
>>          estimators, and algorithms},
>>  number = 293,
>>  institution = {{UW} Biostatistics Working Paper Series},
>>  keywords = {glm, Poisson},
>>  url = {http://www.bepress.com/**uwbiostat/paper293<http://www.bepress.com/uwbiostat/paper293 
>> >
>> }
>>
>> }
>>
>>
>> 2) how to obtain  the predicted risk for a certain individual using
>>> fitted regression model in R.
>>>
>>> Many thanks, in advance, for your help.
>>>
>>> Amin.
>>>
>>
>> Michael Dewey
>> [hidden email]
>> http://www.aghmed.fsnet.co.uk/**home.html<http://www.aghmed.fsnet.co.uk/home.html 
>> >
>>
>>
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> [hidden email] mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.

David Winsemius, MD
Alameda, CA, USA

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Frank Harrell
Department of Biostatistics, Vanderbilt University