<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8">
</head>
<body>
<p>Dear Sean,</p>
<p>You might try R package netmeta, functions netmeta() and
netcomb() that have been developed perfectly for your situation
<a class="moz-txt-link-freetext" href="https://cran.r-project.org/web/packages/netmeta/index.html">https://cran.r-project.org/web/packages/netmeta/index.html</a>. For
the theory see
<a class="moz-txt-link-freetext" href="https://onlinelibrary.wiley.com/doi/10.1002/bimj.201800167">https://onlinelibrary.wiley.com/doi/10.1002/bimj.201800167</a> and
also the seminal paper
<a class="moz-txt-link-freetext" href="https://academic.oup.com/aje/article/169/9/1158/125216">https://academic.oup.com/aje/article/169/9/1158/125216</a> .</p>
<p>Attached you find a presentation from our teaching in Freiburg,
also including some R code. Hope this is helpful.<br>
</p>
<p>Best,</p>
<p>Gerta<br>
</p>
<p><br>
</p>
<div class="moz-cite-prefix">Am 09.03.2021 um 23:57 schrieb Sean:<br>
</div>
<blockquote type="cite"
cite="mid:CAP7kma5mzw5QqXg_+H3uO1GExagJzOOZpJr-m+ScO-yqo-VgaA@mail.gmail.com">
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
<div dir="ltr">Hi Dr. R<span style="color:rgb(32,33,36)"><font
face="arial, sans-serif">ü</font></span>cker,
<div><br>
- Let A, B, C, D be some fungicides, do I understand it
correctly that<br>
treatments are combinations in the sense of A+B, A+C+D, A+B+C
etc.?
<div><br>
</div>
<div><b>Yes, that is correct. </b><br>
<br>
- In each trial, several of these combinations are compared?</div>
<div><br>
</div>
<div><b>Yes. </b><br>
<br>
- "Design" is meant to be the notion in network
meta-analysis, thus "A<br>
vs A+B vs A+B+C" would be a possible design?</div>
<div><br>
</div>
<div><b>Yes I think you have it, for example, trial 1 had A+B,
A+C+D, and A+B+C as its treatments and trial 2 had A+C,
A+D, and A+C+D, and so on. </b><br>
<br>
Which R package do you use?</div>
<div><br>
</div>
<div><b>metafor</b></div>
</div>
<div><b><br>
</b></div>
<div>Best,</div>
<div><br>
</div>
<div>Sean</div>
</div>
<br>
<br>
<div class="gmail_quote">
<div dir="ltr" class="gmail_attr">On Tue, Mar 9, 2021 at 5:47 PM
Dr. Gerta Rücker <<a
href="mailto:ruecker@imbi.uni-freiburg.de"
moz-do-not-send="true">ruecker@imbi.uni-freiburg.de</a>>
wrote:<br>
</div>
<blockquote class="gmail_quote" style="margin:0px 0px 0px
0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Dear
Sean,<br>
<br>
At first, I have to understand your problem. You mention
treatments that <br>
are combinations of fungicides, and you mention designs. To
clarify <br>
these notions, my questions are:<br>
<br>
- Let A, B, C, D be some fungicides, do I understand it
correctly that <br>
treatments are combinations in the sense of A+B, A+C+D, A+B+C
etc.?<br>
<br>
- In each trial, several of these combinations are compared?<br>
<br>
- "Design" is meant to be the notion in network meta-analysis,
thus "A <br>
vs A+B vs A+B+C" would be a possible design?<br>
<br>
(A different possibility is that you do not have combinations
of <br>
treatments in the sense above, but - simpler - different
designs such as <br>
A, B, C in one trial and A, C, D in another trial.)<br>
<br>
Could you clarify these notions?<br>
<br>
Which R package do you use?<br>
<br>
Best,<br>
<br>
Gerta<br>
<br>
Am 09.03.2021 um 23:22 schrieb Sean:<br>
> Hello everyone,<br>
><br>
><br>
> I’ve been stuck on a question about inconsistency testing
for quite<br>
> some time, but first a little simplified background:<br>
><br>
><br>
> I’ve calculated effect sizes for all treatments from 50
independent<br>
> trials conducted over the past 10 years. These treatments
are<br>
> different fungicides applied to a plant to control a
foliar pathogen.<br>
> Throughout those 10 years, researchers tested 20
different products,<br>
> and a treatment (4-15 per trial) is different
combinations of usually<br>
> 1-6 of those fungicides. There was no coordination over
those 10 years<br>
> in experimental design, so no treatment was truly
replicated. Instead,<br>
> what I’ve done is reduce treatments into larger
categories based on<br>
> the modes of action of those fungicides. This has allowed
me to have<br>
> enough similarly coded treatments to perform a network
meta-analysis.<br>
> That went all well and good, however, when it comes to
inconsistency<br>
> testing, I have as many study designs as I have studies.
50<br>
> independent trials, 50 designs.<br>
><br>
><br>
> Can I even technically perform inconsistency testing?
What I've read<br>
> in the literature doesn't seem to account for my
situation. If not,<br>
> what does this mean for my meta-analysis? Do I truly need
to perform<br>
> inconsistency testing?<br>
><br>
><br>
> Thank you all for your time, hope your week is going
well!<br>
><br>
><br>
> Sean<br>
><br>
> _______________________________________________<br>
> R-sig-meta-analysis mailing list<br>
> <a href="mailto:R-sig-meta-analysis@r-project.org"
target="_blank" moz-do-not-send="true">R-sig-meta-analysis@r-project.org</a><br>
> <a
href="https://stat.ethz.ch/mailman/listinfo/r-sig-meta-analysis"
rel="noreferrer" target="_blank" moz-do-not-send="true">https://stat.ethz.ch/mailman/listinfo/r-sig-meta-analysis</a><br>
</blockquote>
</div>
</blockquote>
<pre class="moz-signature" cols="72">--
Dr. rer. nat. Gerta Rücker, Dipl.-Math.
Institute of Medical Biometry and Statistics,
Faculty of Medicine and Medical Center - University of Freiburg
Stefan-Meier-Str. 26, D-79104 Freiburg, Germany
Phone: +49/761/203-6673
Fax: +49/761/203-6680
Mail: <a class="moz-txt-link-abbreviated" href="mailto:ruecker@imbi.uni-freiburg.de">ruecker@imbi.uni-freiburg.de</a>
Homepage: <a class="moz-txt-link-freetext" href="https://www.uniklinik-freiburg.de/imbi-en/employees.html?imbiuser=ruecker">https://www.uniklinik-freiburg.de/imbi-en/employees.html?imbiuser=ruecker</a>
</pre>
</body>
</html>