<html><head></head><body><div style="font-family: Verdana;font-size: 12.0px;"><div> </div>
<div>Dear all,</div>
<div>th<span style="font-family:Verdana,Geneva,sans-serif;">is is my first time writing a message here but I have already exploited great info from previous mail and threads.</span></div>
<div><span style="font-family:Verdana,Geneva,sans-serif;">I am seeking for advice about a multivariate meta-analysis with metafor.</span></div>
<div><span style="font-family:Verdana,Geneva,sans-serif;">I have a dataset very similar to dat.ishak2007 as nested structure, but the effect size is different (LogOR, related to vaccine effectiveness evalutated at several time points).</span></div>
<div>
<div><span style="font-family:Verdana,Geneva,sans-serif;">Therefore, I guess that a multivariate model with heteroscedastic AR(1) structure for the true effects would serve the purpose, as well explained also in Musekiwa et al. (PlosOne 2016, models 4-6).</span></div>
<div><span style="font-family:Verdana,Geneva,sans-serif;">I found in the Web from the work</span>shop on "Exploring the Limits of Advanced Meta-Analysis" by Wolfgang Viechtbauer the following code:</div>
<div>res <- rma.mv(yi, V, mods = ~ factor(time) - 1, random = ~ time | study, struct="HAR", data=dat, digits=2)</div>
<div>Moreover, nice plotting through:</div>
<div> </div>
</div>
<div><span style="font-size:12px;"><span style="font-family:Verdana,Geneva,sans-serif;"><a id="cb28-2" style="white-space-collapse: preserve;" title="2"><span style="color:#000000;">par(mar=c(4,4,2,2))</span></a></span></span></div>
<div><span style="font-size:12px;"><span style="font-family:Verdana,Geneva,sans-serif;"></span></span>
<div>with(dat, interaction.plot(time, study, yi, type="b", pch=19, col="gray70", lty="solid", xaxt="n", legend=FALSE, bty="l", xlab="Time Point", ylab="Mean Difference")) axis(side=1, at=1:4, lab=c("1 (3 months)", "2 (6 months)", "3 (12 months)", "4 (12+ months)")) points(1:4, coef(res), type="o", pch=19, lwd=4, cex=2)</div>
<span style="font-size:12px;"><span style="font-family:Verdana,Geneva,sans-serif;"></span></span></div>
<div> </div>
<div>The first, more trivial question, is how to derive prediction interval for the 4 factorized timepoints and how to plot them as error bars.</div>
<div> </div>
<div>But more important question: how to derive in an easy way for a non-biostatistician the V matrix choosing the phi values for:</div>
<div>V <- vcalc(vi, cluster=study, time1=time, data=dat, phi=0.8) - 0.8 is in the example related to dat.ishak2007.</div>
<div>Of course I read it is tricky and there are some complex equations and calculations in SAS, but from my point of view an easier alternative would be very precious.</div>
<div> </div>
<div> </div>
<div>As an alternative, could I go opt for a three-level meta-analysis as done by Wu and colleagues (https://pubmed.ncbi.nlm.nih.gov/36780914/)? In the Supplementary they shared this code following the advice in Chapter 10 of "Doing Meta-Analysis with R":</div>
<div> cov.mod2 <- <a data-saferedirecturl="https://www.google.com/url?q=http://rma.mv&source=gmail&ust=1699099697465000&usg=AOvVaw0lI40Uxv5Gz4tS1DI8UJnB" href="http://rma.mv/" target="_blank">rma.mv</a>(Log.RR,<br/>
var.log,<br/>
random = ~ 1 | Study.ID/effect.number, #each different ES across studies are numbered (1,2,3 the first study; 4,5,6 the subsequent and so on) and the variable is called effect.number<br/>
test="t",<br/>
data = COVEND2,<br/>
method = "REML",<br/>
mods = ~ FUP, #FUP stands for the factorized time interval (follow-up)<br/>
control=list(maxiter=10^6))</div>
<div> </div>
<div>In this case as grouping variable, instead of effect.number, could FUP (time as factor) be put?</div>
<div> </div>
<div> </div>
<div>I hope my request for help is clear.</div>
<div> </div>
<div> </div>
<div>Warm regards,</div>
<div>Alberto</div>
<div> </div>
<div class="signature"> </div>
<div class="signature"> </div>
<div class="signature">
<div>
<div>
<p><span style="font-size:10px;">Alberto Enrico Maraolo, MD, MSc (Antimicrobial Stewardship, Evidence Synthesis), FESCMID<br/>
<br/>
Infectious Diseases Specialist, Member of the Steering Committee of SIMIT (ID Italian Society)<br/>
<br/>
Cotugno Hospital, AORN dei Colli, Naples, Italy</span></p>
</div>
<div>
<p><span style="font-size:10px;">mail: <a href="mailto:albertomaraolo@mail.com" onclick="parent.window.phx.iac.notify('mail_compose', {'to':['albertomaraolo@mail.com']}); return false;" target="_blank">albertomaraolo@mail.com</a></span></p>
</div>
<div>
<p> </p>
</div>
</div>
<div>
<p><span style="font-size:10px;">Alberto Enrico Maraolo, MD, MSc (Antimicrobial Stewardship, Evidence Synthesis), FESCMID<br/>
<br/>
Specialista in Malattie Infettive, Consigliere Nazionale Direttivo SIMIT (Società Italiana di Malattie Infettive e Tropicali)<br/>
<br/>
Dirigente Medico, AORN dei Colli - Ospedale Cotugno, Napoli<br/>
<br/>
I Divisione - Malattie Infettive emergenti e ad alta contagiosità (ex Malattie Infettive ad indirizzo neurologico)<br/>
<br/>
mail: <a href="mailto:albertomaraolo@mail.com" onclick="parent.window.phx.iac.notify('mail_compose', {'to':['albertomaraolo@mail.com']}); return false;" target="_blank">albertomaraolo@mail.com</a></span></p>
</div>
</div></div></body></html>