[R-meta] Covariance-variance matrix when studies share multiple treatment x control comparison
Viechtbauer, Wolfgang (SP)
wo||g@ng@v|echtb@uer @end|ng |rom m@@@tr|chtun|ver@|ty@n|
Thu Sep 26 16:52:14 CEST 2019
1) As has been discussed on this mailing list on a few occasions: In addition to adding 'Study' as random effects, you should also add random effects for the estimates within studies. So, your 'base' model should be:
MHF$Id <- 1:nrow(MHF)
rma.mv(hedged, var, method="REML", random = ~ 1 | Study/Id, data=MHF)
(otherwise, you are assuming no heterogeneity within studies, which is a very strong assumption)
2) Your 'Egger-like multilevel regression test' model would simply be:
rma.mv(hedged, var, mods = ~ sqrt(var), method="REML", random = ~ 1 | Study/Id, data=MHF)
You then test if sqrt(var) is a significant predictor or not.
3) But 'var' should be a matrix -- after all, that was the point of the earlier discussion. So, if it is, then it would be:
rma.mv(hedged, var, mods = ~ sqrt(diag(var)), method="REML", random = ~ 1 | Study/Id, data=MHF)
From: Ju Lee [mailto:juhyung2 using stanford.edu]
Sent: Thursday, 26 September, 2019 14:58
To: Viechtbauer, Wolfgang (SP); r-sig-meta-analysis using r-project.org; James Pustejovsky (jepusto using gmail.com)
Subject: Re: Covariance-variance matrix when studies share multiple treatment x control comparison
Thank you for your response and sorry I forgot to CC the mailing list!
I am currently running my egger's regression test as shown below. My previous understanding was that I should look at the p-value of intercept term (following a previously published R code) if I run a "mixed" model using precision as moderator variable against residuals, but according to your comments I should be looking at the precision coefficients instead? So based on my outputs below, significance testing of plot asymmetry is at p=0.09 and not p=0.3823?
Also, if I find significant violation of plot asymmetry in such case what additional options do I have to test these issues? I am currently calculating FSN which are extremely higher than proposed thresholds and removing influential outliers and re-fitting the model. But because rma.mv does not allow me to use other methods like trim and fill I wonder if these two other methods would be enough in case we detect plot asymmetry.
Thank you for your time to answer these many questions.
>Full.egger.es<-rma.mv(hedged,var, method="REML", random = ~ 1 | Study, data=MHF)
>egger.full<-rma.mv(residuals~precision,var,data=MHF,random = ~ 1 | Study)
Multivariate Meta-Analysis Model (k = 857; method: REML)
estim sqrt nlvls fixed factor
sigma^2 0.9929 0.9964 182 no Study
Test for Residual Heterogeneity:
QE(df = 855) = 4106.3487, p-val < .0001
Test of Moderators (coefficient(s) 2):
QM(df = 1) = 2.7267, p-val = 0.0987
estimate se zval pval ci.lb ci.ub
intrcpt 0.0817 0.0936 0.8727 0.3828 -0.1017 0.2651
precision -0.0392 0.0238 -1.6513 0.0987 -0.0858 0.0073 .
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
More information about the R-sig-meta-analysis