R help - Nov 2011 - Sensitivity and Specificity Forest Plots

Dear R Users,

Do you know of an existing function that allows the production of
sensitivity and specificity forest plots?
See the following for an example:
http://www.google.co.uk/imgres?q=forest+plots+of+sensitivity+and+specificity&um=1&hl=en&authuser=0&biw=1920&bih=989&tbm=isch&tbnid=JLxXNU7iQ2N_CM:&imgrefurl=http://www.medscape.com/viewarticle/717751_3&docid=NnthE8_KgNQZdM&imgurl=http://img.medscape.com/article/717/751/717751-fig3.jpg&w=857&h=734&ei=UnzKTvfTGIWC8gPnq4R7&zoom=1&iact=hc&vpx=1300&vpy=282&dur=19&hovh=208&hovw=243&tx=140&ty=108&sig=112892243842455187632&page=1&tbnh=123&tbnw=144&start=0&ndsp=69&ved=1t:429,r:20,s:0

The standard meta-analysis packages (metafor etc) do not seem to have this
type of forest plot.

If I were to construct a bespoke function within R, would you recommend
using the forest function within metafor, or the dotplot function within
lattice, or something else entirely as a starting point for constructing
the function?

Many thanks,

-- 
Dr Jon Minton
Decision Support Unit
Health Economics & Decision Science
School of Health and Related Research
University of Sheffield
Tel: 0114 222 0836

	[[alternative HTML version deleted]]

Essentially, this is a side-by-side forest plot, where the plot on the left is
for sensitivity and the plot on the right is for specificity. For 2x2 table data
from diagnostic studies, it is easy to calculate the sensitivity and specificity
values (and corresponding sampling variances) by hand. In fact, since
sensitivity and specificity values are actually simply proportions, you can use
the escalc() function in the metafor package to do this for you (just pass the
appropriate cell counts to escalc() and use one of the outcome measures listed
under "Proportions and Transformations Thereof" -- see help(escalc).
Then, you could set up a plotting device with layout() and put the forest plot
for sensitivity on the left and the plot for specificity on the right. So,
everything is already there to do this.

Best,

Wolfgang

-- 
Wolfgang Viechtbauer, Ph.D., Statistician 
Department of Psychiatry and Psychology 
School for Mental Health and Neuroscience 
Faculty of Health, Medicine, and Life Sciences 
Maastricht University, P.O. Box 616 
6200 MD Maastricht, The Netherlands 
+31 (43) 368-5248 | http://www.wvbauer.com 

> -----Original Message-----
> From: r-help-bounces at r-project.org [mailto:r-help-bounces at
r-project.org]
> On Behalf Of Jonathan Minton
> Sent: Monday, November 21, 2011 17:34
> To: r-help at r-project.org
> Subject: [R] Sensitivity and Specificity Forest Plots
> 
> Dear R Users,
> 
> Do you know of an existing function that allows the production of
> sensitivity and specificity forest plots?
> See the following for an example:
> http://www.google.co.uk/imgres?q=forest+plots+of+sensitivity+and+specifici
>
ty&um=1&hl=en&authuser=0&biw=1920&bih=989&tbm=isch&tbnid=JLxXNU7iQ2N_CM:&i
>
mgrefurl=http://www.medscape.com/viewarticle/717751_3&docid=NnthE8_KgNQZdM
> &imgurl=http://img.medscape.com/article/717/751/717751-
>
fig3.jpg&w=857&h=734&ei=UnzKTvfTGIWC8gPnq4R7&zoom=1&iact=hc&vpx=1300&vpy=2
>
82&dur=19&hovh=208&hovw=243&tx=140&ty=108&sig=112892243842455187632&page=1
> &tbnh=123&tbnw=144&start=0&ndsp=69&ved=1t:429,r:20,s:0
> 
> The standard meta-analysis packages (metafor etc) do not seem to have this
> type of forest plot.
> 
> If I were to construct a bespoke function within R, would you recommend
> using the forest function within metafor, or the dotplot function within
> lattice, or something else entirely as a starting point for constructing
> the function?
> 
> Many thanks,
> 
> --
> Dr Jon Minton
> Decision Support Unit
> Health Economics & Decision Science
> School of Health and Related Research
> University of Sheffield
> Tel: 0114 222 0836
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> R-help at r-project.org 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.