<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<style type="text/css" style="display:none;"> P {margin-top:0;margin-bottom:0;} </style>
</head>
<body dir="ltr">
<div class="elementToProof" style="font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
Oh, Gerta, I think you solved my problem. I ran it with sm set to "OR" and it changed, solving my issue. See below the output:</div>
<div class="elementToProof" style="font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<br>
</div>
<div class="elementToProof" style="font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<img size="27650" contenttype="image/png" style="max-width: 1250px;" data-outlook-trace="F:1|T:1" src="cid:33f397c5-ff99-4507-9eeb-c266f3c1b04a"></div>
<div class="elementToProof" style="font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 12pt; color: rgb(0, 0, 0);">
<br>
</div>
<div id="Signature" style="color: inherit;">
<div style="font-size: 10pt;"><b>Sancho Xavier</b></div>
<div style="font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 10pt;">
Candidate for Postgraduate Programme in Epidemiology in Public Health</div>
<div style="font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 10pt; color: rgb(0, 0, 0);">
MBA in Data Science and Analytics at USP (ongoing)</div>
<div style="font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 10pt;">
<b>Linkedin:</b> linkedin.com/in/sancho-xavier-59565a18a</div>
<div style="font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 10pt;">
<b>Orcid iD:</b> https://orcid.org/0000-0001-9493-4098</div>
<div style="font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 10pt; color: rgb(0, 0, 0);">
<b>Web of Science (Clarivate):</b> <a href="https://www.webofscience.com/wos/author/record/GON-5675-2022">https://www.webofscience.com/wos/author/record/GON-5675-2022</a></div>
<div style="font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 10pt; color: rgb(0, 0, 0);">
<b>Reference Citation Analysis (RCA): </b><a href="https://referencecitationanalysis.com/00001471">https://referencecitationanalysis.com/00001471</a></div>
<div style="font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 10pt;">
<b>Whatsapp: </b>+258 852255846 </div>
<span style="font-family: Aptos, Aptos_EmbeddedFont, Aptos_MSFontService, Calibri, Helvetica, sans-serif; font-size: 10pt;">------------------------------------------------------------------------------</span></div>
<div id="appendonsend"></div>
<hr style="display:inline-block;width:98%" tabindex="-1">
<div id="divRplyFwdMsg" dir="ltr"><font face="Calibri, sans-serif" style="font-size:11pt" color="#000000"><b>De:</b> Dr. Gerta Rücker <gerta.ruecker@uniklinik-freiburg.de><br>
<b>Enviado:</b> quarta-feira, 31 de julho de 2024 14:06<br>
<b>Para:</b> R Special Interest Group for Meta-Analysis <r-sig-meta-analysis@r-project.org>; Sancho Xavier <sanchoxavierxavier@gmail.com><br>
<b>Cc:</b> Michael Dewey <lists@dewey.myzen.co.uk><br>
<b>Assunto:</b> AW: [SPAM](FN) Re: [R-meta] Is this R script correct for conducting a meta-analysis of AOR for hypertension prevalence?</font>
<div> </div>
</div>
<div class="BodyFragment"><font size="2"><span style="font-size:11pt;">
<div class="PlainText">Hi Sancho and Michael,<br>
<br>
There is clearly an error in Sancho's code. The first screenshot shows that he declared sm = "AOR" - which doesn't make sense. The possible values of sm (summary measure) are from a fixed list, e.g., "RD", "RR", "OR", "ASD", "HR", "MD", "SMD", or "ROM". Consequently,
as "AOR" is not a possible value, it is ignored. <br>
<br>
Now, as no summary measure is specified, metagen() assumes the mean difference (MD) to be the effect measure. This also explains why the confidence intervals (that include 1) correspond to highly significant p-values (p < 0.0001) in the second screenshot: As
implicitly mean differences are assumed, the null hypothesis assumes a mean of 0, not 1. And the CIs don't include 0.<br>
<br>
Solution: You have to set sm = "OR".<br>
<br>
Best,<br>
Gerta<br>
<br>
<br>
<br>
<br>
UNIVERSITÄTSKLINIKUM FREIBURG<br>
Institute for Medical Biometry and Statistics<br>
<br>
Dr. Gerta Rücker<br>
Guest Scientist<br>
<br>
Stefan-Meier-Straße 26 · 79104 Freiburg<br>
gerta.ruecker@uniklinik-freiburg.de<br>
<br>
<a href="https://www.uniklinik-freiburg.de/imbi-en/employees.html?imbiuser=ruecker">https://www.uniklinik-freiburg.de/imbi-en/employees.html?imbiuser=ruecker</a><br>
<br>
-----Ursprüngliche Nachricht-----<br>
Von: R-sig-meta-analysis <r-sig-meta-analysis-bounces@r-project.org> Im Auftrag von Michael Dewey via R-sig-meta-analysis<br>
Gesendet: Mittwoch, 31. Juli 2024 18:30<br>
An: Sancho Xavier <sanchoxavierxavier@gmail.com>; R Special Interest Group for Meta-Analysis <r-sig-meta-analysis@r-project.org><br>
Cc: Michael Dewey <lists@dewey.myzen.co.uk><br>
Betreff: [SPAM](FN) Re: [R-meta] Is this R script correct for conducting a meta-analysis of AOR for hypertension prevalence?<br>
<br>
Dear Sancho<br>
<br>
Responses in line<br>
<br>
On 31/07/2024 15:27, Sancho Xavier wrote:<br>
> Dear Michael,<br>
> <br>
> Thank you for your response. I did not apply any moderator for the <br>
> pooled OR in the meta-analysis. Could you clarify something for me?<br>
> In a meta-analysis, is it always necessary to calculate the pooled OR <br>
> including a moderator variable?<br>
<br>
No, you include a moderator if your scientific hypothesis demands it.<br>
<br>
> *Another question:*<br>
> Was the pooled OR supposed to be significant in the following conde (image):<br>
<br>
Well the confidence interval clearly includes the null. I am not sure <br>
what you mean by supposed to be here.<br>
<br>
If you want to include output it is better to paste it in text format as <br>
including screenshots mangles the post.<br>
<br>
Michael<br>
<br>
> <br>
> <br>
> *Results:*<br>
> <br>
> *Dataset:*<br>
> *<br>
> *<br>
> **<br>
> *<br>
> *<br>
> Do you have any suggestions, ideas, or observations about the scripts <br>
> and results based on the dataset?<br>
> <br>
> Best wishes,<br>
> <br>
> *Sancho Xavier*<br>
> Candidate for Postgraduate Programme in Epidemiology in Public Health<br>
> MBA in Data Science and Analytics at USP (ongoing)<br>
> *Linkedin:* linkedin.com/in/sancho-xavier-59565a18a<br>
> *Orcid iD:* <a href="https://orcid.org/0000-0001-9493-4098">https://orcid.org/0000-0001-9493-4098</a><br>
> *Web of Science (Clarivate):* <br>
> <a href="https://www.webofscience.com/wos/author/record/GON-5675-2022">https://www.webofscience.com/wos/author/record/GON-5675-2022</a>
<br>
> <<a href="https://www.webofscience.com/wos/author/record/GON-5675-2022">https://www.webofscience.com/wos/author/record/GON-5675-2022</a>><br>
> *Reference Citation Analysis (RCA): <br>
> *https://referencecitationanalysis.com/00001471 <br>
> <<a href="https://referencecitationanalysis.com/00001471">https://referencecitationanalysis.com/00001471</a>><br>
> *Whatsapp: *+258 852255846<br>
> ------------------------------------------------------------------------------<br>
> ------------------------------------------------------------------------<br>
> *De:* Michael Dewey <lists@dewey.myzen.co.uk><br>
> *Enviado:* quarta-feira, 31 de julho de 2024 05:05<br>
> *Para:* R Special Interest Group for Meta-Analysis <br>
> <r-sig-meta-analysis@r-project.org><br>
> *Cc:* Sancho Xavier <sanchoxavierxavier@gmail.com><br>
> *Assunto:* Re: [R-meta] Is this R script correct for conducting a <br>
> meta-analysis of AOR for hypertension prevalence?<br>
> Dear Sancho<br>
> <br>
> Are you intending to use the variable you have named variable as a<br>
> moderator? If not the code looks fine to me although I do not usually<br>
> use that package (I assume it is meta).<br>
> <br>
> Michael<br>
> <br>
> On 27/07/2024 00:18, Sancho Xavier via R-sig-meta-analysis wrote:<br>
>> *Hi, community!*<br>
>> I am conducting a meta-analysis and meta-regression to calculate the <br>
>> */pooled adjusted odds ratio (AOR)/* and */I² with the p-value/* for <br>
>> variables associated with hypertension prevalence. I have a dataset with <br>
>> the following columns:<br>
>> <br>
>> * *study_id:* Study identifier<br>
>> * *AOR:* Adjusted odds ratio (each independent variable)<br>
>> * *lower_CI:* Lower confidence interval (each independent variable)<br>
>> * *upper_CI:* Upper confidence interval (each independent variable)<br>
>> * *variable:* Independent variable<br>
>> <br>
>> Here is the script I am using in *R*:<br>
>> <br>
>> *# Calculate the standard error (seTE)*<br>
>> *<br>
>> *<br>
>> data$seTE <- (log(data$upper_CI) - log(data$lower_CI)) / (2 * 1.96)<br>
>> <br>
>> *# Meta-analysis*<br>
>> *<br>
>> *<br>
>> meta_analysis <- metagen(TE = log(data$AOR),<br>
>> seTE = data$seTE,<br>
>> data = data,<br>
>> studlab = data$study_id,<br>
>> comb.fixed = TRUE,<br>
>> comb.random = TRUE)<br>
>> <br>
>> *I would like to know if this script is correct for achieving my goals of:*<br>
>> <br>
>> * Calculating the pooled AOR and the 95% confidence interval.<br>
>> *<br>
>> Calculating the I² and the p-value to assess heterogeneity.<br>
>> <br>
>> Look at the dataset:<br>
>> <br>
>> Any feedback or suggestions for improvement would be greatly appreciated!<br>
>> <br>
>> Thank you!<br>
>> <br>
>> *Sancho Xavier*<br>
>> Candidate for Postgraduate Programme in Epidemiology in Public Health<br>
>> MBA in Data Science and Analytics at USP (ongoing)<br>
>> *Linkedin:* linkedin.com/in/sancho-xavier-59565a18a<br>
>> *Orcid iD:* <a href="https://orcid.org/0000-0001-9493-4098">https://orcid.org/0000-0001-9493-4098</a>
<br>
> <<a href="https://orcid.org/0000-0001-9493-4098">https://orcid.org/0000-0001-9493-4098</a>><br>
>> *Web of Science (Clarivate):* <br>
>> <a href="https://www.webofscience.com/wos/author/record/GON-5675-2022">https://www.webofscience.com/wos/author/record/GON-5675-2022</a>
<br>
> <<a href="https://www.webofscience.com/wos/author/record/GON-5675-2022">https://www.webofscience.com/wos/author/record/GON-5675-2022</a>><br>
>> <<a href=""></a>https://www.webofscience.com/wos/author/record/GON-5675-2022 <br>
> <<a href="https://www.webofscience.com/wos/author/record/GON-5675-2022">https://www.webofscience.com/wos/author/record/GON-5675-2022</a>>><br>
>> *Reference Citation Analysis (RCA): <br>
>> *https://referencecitationanalysis.com/00001471 <br>
>> <<a href=""></a>https://referencecitationanalysis.com/00001471 <br>
> <<a href="https://referencecitationanalysis.com/00001471">https://referencecitationanalysis.com/00001471</a>>><br>
>> *Whatsapp: *+258 852255846<br>
>> ------------------------------------------------------------------------------<br>
>> <br>
>> <<a href=""></a>http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=emailclient <<a href="http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=emailclient">http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=emailclient</a>>>
Virus-free.www.avg.com <<a href=""></a>http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=emailclient <<a href="http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=emailclient">http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=emailclient</a>>><br>
>> <br>
>> <#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2><br>
>> <br>
>> _______________________________________________<br>
>> R-sig-meta-analysis mailing list @ R-sig-meta-analysis@r-project.org<br>
>> To manage your subscription to this mailing list, go to:<br>
>> <a href="https://stat.ethz.ch/mailman/listinfo/r-sig-meta-analysis">https://stat.ethz.ch/mailman/listinfo/r-sig-meta-analysis</a>
<br>
> <<a href="https://stat.ethz.ch/mailman/listinfo/r-sig-meta-analysis">https://stat.ethz.ch/mailman/listinfo/r-sig-meta-analysis</a>><br>
> <br>
> -- <br>
> Michael<br>
<br>
-- <br>
Michael<br>
<br>
_______________________________________________<br>
R-sig-meta-analysis mailing list @ R-sig-meta-analysis@r-project.org<br>
To manage your subscription to this mailing list, go to:<br>
<a href="https://stat.ethz.ch/mailman/listinfo/r-sig-meta-analysis">https://stat.ethz.ch/mailman/listinfo/r-sig-meta-analysis</a><br>
</div>
</span></font></div>
</body>
</html>