[R-meta] Meta package - adhoc CI correction

[Guido Schwarzer]

Dear Tobias,

Ouch, this is a bug in meta, version 4.16-0. I just submitted version 
4.16-1 to CRAN which should be available later today.

Please note, the following changes concerning the Hartung-Knapp ad hoc 
methods in version 4.16-0:

- argument 'adhoc.hakn = "iqwig6"' instead of 'adhoc.hakn = "ci"' uses 
the ad hoc method for Hartung-Knapp method described in General Methods 
6.0 (IQWiG, 2020)

- argument 'adhoc.hakn = "ci"' uses the ad hoc method described in 
Jackson et al. (2017)

I guess your friend still has version 4.15-1 of meta and thus argument 
'adhoc.hakn = "ci"' worked for him / her as this is the IQWiG method.

Being a co-author of Jackson et al. (2017), it is quite embarrassing to 
provide a buggy version of this ad hoc method (my only justification is 
that I misread the description in van Aert and Jackson, 2019, p. 525).

See the attached R code for a summary of the various methods for your 
example and two variants. The R code only works with meta, version 
4.16-1 due to a minor change in metabind().

Best wishes, Guido

References:

IQWiG. General Methods: Draft of Version 6.0 [Internet]. 2020. Available 
from: https://www.iqwig.de/en/methods/methods-paper.3020.html 
<https://www.iqwig.de/en/methods/methods-paper.3020.html>

Jackson D, Law M, Rücker G, Schwarzer G. The Hartung-Knapp modification 
for random-effects meta-analysis: A useful refinement but are there any 
residual concerns? Stat Med. 2017;36:3923–34.
Aert RCM van, Jackson D. A new justification of the Hartung-Knapp method 
for random-effects meta-analysis based on weighted least squares 
regression. Research Synthesis Methods. 2019;10(4):515–27.
1.
IQWiG. General Methods: Draft of Version 6.0 [Internet]. 2020. Available 
from: https://www.iqwig.de/en/methods/methods-paper.3020.html 
<https://www.iqwig.de/en/methods/methods-paper.3020.html>

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: DN_Adhoc_ex.R
URL: <https://stat.ethz.ch/pipermail/r-sig-meta-analysis/attachments/20210119/32ee6c74/attachment.ksh>