# [R-meta] Meta-analysis of incidence data with double arcsine transformation

Joao Afonso jot@|on@o @end|ng |rom gm@||@com
Tue Feb 4 18:16:38 CET 2020

```Dear all,

I am conducting a meta-analysis on incidence data. I have basically adapted
a previous model that I had used for conducting an analysis on prevalence
data, producing the following model:

*ies.da=escalc(xi=nototal_stand, ti=ssizeanimal, data=inc_all_nomv,
measure="IRLN", add=0)pes.da=rma(yi, vi, data=ies.da, method =
"DL")pes=predict(pes.da, transf=transf.ipft.hm <http://transf.ipft.hm>,
targ=list(ti=inc_all_nomv\$ssizeanimal))print(pes.da,
digits=3)confint(pes.da, digits=3)print(pes, digits=3)*

The model runs fine and there doesn't seem to be an error. The output is as
follows:

*Random-Effects Model (k = 26; tau^2 estimator: DL)tau^2 (estimated amount
of total heterogeneity): 0.896 (SE = 0.590)tau (square root of estimated
tau^2 value):      0.947I^2 (total heterogeneity / total variability):
99.96%H^2 (total variability / sampling variability):  2704.01Test for
Heterogeneity:Q(df = 25) = 67600.255, p-val < .001*

*Model Results:estimate     se    zval   pval   ci.lb <http://ci.lb>
ci.ub   -1.075  0.188  -5.729  <.001  -1.443  -0.707  ****

*print(pes, digits=3)  pred ci.lb <http://ci.lb> ci.ub cr.lb <http://cr.lb>
cr.ub  0.000 0.000 0.000 0.000 0.531 *

The last bit of the output ( *print(pes, digits=3)*) is what's concerning
me as it seems that the back transformation hasn't been made correctly. Any
idea of what I could be doing wrong?

Many thanks for all the help!

Wishing all a great day,

--
João Afonso
DVM, MSc Veterinary Epidemiology
PhD Student
Department of Infection and Global Health
University of Liverpool
+351914812305

[[alternative HTML version deleted]]

```