<div dir='auto'>Dear Guido, <div dir="auto"><br></div><div dir="auto">thank you very much for your reply. Great that you have implemented it! </div><div dir="auto"><br></div><div dir="auto"><br></div><div dir="auto">Regards </div><div dir="auto"><br></div><div dir="auto">Tobias</div></div><div class="gmail_extra"><br><div class="gmail_quote">Am 30.09.2020 18:02 schrieb Guido Schwarzer <sc@imbi.uni-freiburg.de>:<br type="attribution" /><blockquote class="quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div>
<p>Am 10.08.20 um 12:06 schrieb Tobias Saueressig:<br />
</p>
<blockquote>
<div style="font-family:'verdana';font-size:12px">
<div>Hello,</div>
<div> </div>
<div>I am performing a diagnostic accuracy meta-analysis. As a
check for publication bias/small study effects I want to use
the Deeks' funnel plot asymmetry test for publication bias. I
could not find a implementation in a software program. Does
anybody know if there is a package that contains this test?</div>
</div>
</blockquote>
<p>Probably, a little bit late, however, for the record and future
queries. ;-)<br />
</p>
<p>I implemented the Deeks' test in R package <b>meta</b>. In order
to use it one has to conduct a meta-analysis with <i>metabin()</i>
using the diagnostic odds ratio as summary measure. Then, R
function <i>metabias()</i> conducts the linear regression test
using the inverse of the square root of the effective study size.</p>
<p>Using Wolfgang's example dataset:</p>
<p>dat <- read.table(header = TRUE, text = "<br />
tp np tn nn<br />
10 25 47 53<br />
15 24 29 32<br />
10 14 44 47<br />
13 23 123 123<br />
10 32 55 56<br />
16 61 184 186<br />
12 49 137 138<br />
13 26 66 74<br />
16 80 NA NA<br />
24 42 55 57<br />
17 27 85 91<br />
5 22 45 46<br />
18 31 240 245<br />
24 28 104 104<br />
26 55 81 89")<br />
<br />
library(meta)<br />
settings.meta(digits = 2)<br />
<br />
## Drop study with missing information and conduct meta-analysis<br />
##<br />
dat.nomiss <- subset(dat, !is.na(tn))<br />
##<br />
m <- metabin(tp, np, nn - tn, nn, data = dat.nomiss,<br />
sm = "DOR", addincr = TRUE)<br />
<br />
## Funnel plot with inverse of square root of effective study size
on y-axis<br />
##<br />
funnel(m, yaxis = "ess", xlim = c(1, 1100), ylim = c(0.15, 0),
studlab = TRUE)<br />
## same result with meta, > 4.15-1:<br />
## funnel(m, xlim = c(1, 1100), studlab = TRUE)<br />
<br />
## Deeks' test for funnel plot asymmetry<br />
##<br />
metabias(m)<br />
</p>
<p>Best wishes, Guido</p>
</div>
</blockquote></div><br></div>