<div dir="ltr"><div><div><div>Dear Guido, <br><br></div>Thank you so much for your prompt reply and kind explanation. Indeed I was a bit confused because the error message I get states that:</div></div><div>1. The comparisons for the treatment with 0 standard error were not being considered in the network meta-analysis</div><div>2. As a result of removing those comparisons, the study ends up having the wrong number of comparisons and, consequently, <br></div><div>3. I should consider removing the study from the analysis. <br></div><div></div><div>In your example this doesn't happen because when the rows with a NA (or 0) standard error are removed, the study where they are ends up with just one comparison (between treatments B and C) which can still be included in the analysis. In my case, when the study with the 0 standard error is removed the links between the remainder comparisons are lost so the study can't be included in the analysis (eg. we loose comparison AB and end up with AC and BC). I was thus wondering if there was any way to work around this or whether 0 variances (& consequently standard errors) were simply not allowed in a meta-analysis at all?<br></div><div><br></div><div>For the time being I have excluded the study from the analysis and, coincidentally,I have also encountered a second problem with the checks for the netmeta version 0.9 - 6. For another of the studies in my analysis we have 3 arms and the situation depicted in the network below:</div><div><br></div><div><img src="cid:ii_15ea739c571b76f9" alt="Inline image 1" width="561" height="346"></div><div><br></div><div>Each node of the image above is a treatment in the study with its respective mean and, in parenthesis, approximate variance. The arrows of the edges indicate the direction of the comparison such that the arrow originates at the numerator treatment and points to the denominator treatment for the log response ratio.The numbers in the edges correspond to the log response ratios for each treatment comparison and, in parenthesis, the corresponding variance. You can also find a copy of this data (along with another dummy study) and the code that generates the error in the attached folder. In this case the error message I initially get suggests that I activate the details.chkmultiarm for more details. When I do so, I get the following error: <br></div><div><br></div><div><br></div><div>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(234,234,234)">Multi-arm studies with
inconsistent treatment effects:</span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(234,234,234)"> </span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(234,234,234)"><span> </span>studlab<span> </span>treat1<span> </span>treat2<span> </span>TE<span>
</span>resid</span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(234,234,234)"><span> </span>S14172<span>
</span>exotic.forest<span>
</span>forest.harvested<span> </span>0.7929675<span> </span>0.528645</span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(234,234,234)"><span> </span>S14172<span>
</span>exotic.forest indigenous.forest 16.0182753 -0.528645</span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(234,234,234)"><span> </span>S14172 forest.harvested indigenous.forest
16.8112428<span> </span>0.528645</span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(234,234,234)"> </span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(234,234,234)"> </span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(234,234,234)">Multi-arm studies with zero
treatment arm variance:</span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(234,234,234)"> </span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(234,234,234)"><span> </span>studlab<span> </span>treat<span> </span>var.treat</span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(234,234,234)"><span> </span>S14172<span>
</span>exotic.forest 0.02237507</span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(234,234,234)"><span> </span>S14172<span>
</span>forest.harvested 0.01261770</span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(234,234,234)"><span> </span>S14172 indigenous.forest 0.00000000</span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(234,234,234)"> </span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(234,234,234)">Legend:</span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(234,234,234)"><span> </span>resid - residual deviation (observed minus
expected)</span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(234,234,234)"><span> </span>TE<span> </span>-
treatment estimate</span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(234,234,234)"><span> </span>var.treat - treatment arm variance</span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(222,142,48)">Error: Problems in
multi-arm studies!</span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(222,142,48)"><span> </span>- Study with inconsistent treatment
estimates: 'S14172'</span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(222,142,48)"><span> </span>- Study with zero treatment arm variance:
'S14172'</span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(222,142,48)"><span> </span>- Please check original data used as input to
netmeta().</span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(222,142,48)"><span> </span>- Argument tol.multiarm in netmeta() can be
used to relax consistency</span></p>
<p class="MsoNormal" style="line-height:11.25pt;background:rgb(44,40,40) none repeat scroll 0% 0%;word-break:break-all"><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(222,142,48)"><span> </span>assumption for multi-arm studies (if
appropriate).</span><span style="font-size:10pt;font-family:"Lucida Console";color:rgb(234,234,234)"></span></p>
<br></div><div>I was wondering where I could find out more information about what is causing the treatments to be inconsistent (is it the 0 variance for one of the treatments) and what are the consequences of relaxing the consistency assumption in this case.</div><div><br></div><div>Again, thank you so much for your kind attention and help! <br></div><div><br></div><div>With best wishes, <br></div><div><br></div><div>Carla <br></div></div><div class="gmail_extra"><br><div class="gmail_quote">On Wed, Sep 20, 2017 at 10:57 PM, Guido Schwarzer <span dir="ltr"><<a href="mailto:sc@imbi.uni-freiburg.de" target="_blank">sc@imbi.uni-freiburg.de</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div text="#000000" bgcolor="#FFFFFF">
<p>Dear Carla,</p>
<p>I think this is a misunderstanding.</p>
<p>R function netmeta() works perfectly with zero standard errors
(by excluding pairwise comparisons with zero or missing standard
errors from the network meta-analysis). This is the same behaviour
as in "basic" meta-analysis of pairwise comparisons (see metagen()
in R package <b>meta</b>). I attached a corresponding fictitious
example for netmeta() to this email.<br>
</p>
<p>The new checks in <b>netmeta</b>, version 0.9-6, you are
referring to, related for <i>treatment arm variances in multi-arm
studies</i> which are calculated internally. Sometimes, users
enter standard errors for pairwise comparisons in multi-arm
studies that simply do not "add up". Accordingly, treatment arm
variances can be negative.</p>
<p>Maybe somebody else can also comment on your idea to use a (very)
small standard error for a pairwise comparison with zero standard
error. In my view, the problem with such an approach is that this
study gets a very large weight in the (network) meta-analysis.</p>
Best wishes,<br>
Guido<span class="HOEnZb"><font color="#888888"><br>
<pre class="m_6494916945777013038moz-signature" cols="72">--
Dr. Guido Schwarzer (<a class="m_6494916945777013038moz-txt-link-abbreviated" href="mailto:sc@imbi.uni-freiburg.de" target="_blank">sc@imbi.uni-freiburg.de</a>)
Institute for Medical Biometry and Statistics
Stefan-Meier-Str. 26, D-79104 Freiburg | Phone: <a href="tel:+49%20761%202036668" value="+497612036668" target="_blank">+49 (0)761 203 6668</a>
<a class="m_6494916945777013038moz-txt-link-freetext" href="http://www.imbi.uni-freiburg.de" target="_blank">http://www.imbi.uni-freiburg.<wbr>de</a> | Fax: <a href="tel:+49%20761%202036680" value="+497612036680" target="_blank">+49 (0)761 203 6680</a>
</pre>
</font></span></div>
</blockquote></div><br><br clear="all"><br>-- <br><div class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><b>Carla Gómez Creutzberg</b><br>
PhD. Candidate - <a href="http://www.tylianakislab.org/the-group.html" target="_blank">Tylianakis Lab</a><br>
University of Canterbury - <span lang="en-NZ"><font color="black" size="2" face="Tahoma"><span dir="ltr" style="font-size:10pt"><font size="2" face="Times New Roman"><span style="font-size:16px"><font size="2" face="Tahoma"><span style="font-size:10pt"><font size="1"><span style="font-size:13px"><font size="2"><i>Te
Whare Wānanga o Waitaha</i></font><font size="2"><i><br>
</i></font></span></font></span></font></span></font></span></font></span>Christchurch, New Zealand<a href="http://www.tylianakislab.org/the-group.html" target="_blank"></a> <br></div><div><a href="mailto:cgomezcre@gmail.com" target="_blank">cgomezcre@gmail.com</a> <br></div></div></div></div></div></div></div></div></div></div></div></div></div></div>
</div>