[R] question including crossover trials in meta-analysis

[Marc Schwartz]

Hi,

Bert has pointed you to some R specific packages for meta-analyses via the Task View.

It sounds like you may need to first address some underlying conceptual issues, which strictly speaking, is off-topic for this list.

That being said, a quick Google search came up with some possible resources, beyond the Cochrane reference:

  https://academic.oup.com/ije/article/31/1/140/655940

  https://onlinelibrary.wiley.com/doi/abs/10.1002/jrsm.1236

  https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0133023


Perhaps with additional conceptual background, that might assist in enabling you to make use of the R packages that provide relevant functionality.

Another option, if you want to stay with the dmetar package, would be to contact the package maintainer for some guidance relative to how to use the functionality in the package given your specific use case.

Regards,

Marc Schwartz


> On Sep 14, 2020, at 3:14 PM, Bert Gunter <bgunter.4567 using gmail.com> wrote:
> 
> Did you first try a web search? -- you should always do this before posting
> here.
> 
> "meta-analysis in R" brought up this:
> 
> https://CRAN.R-project.org/view=MetaAnalysis
> 
> Have you looked at this task view yet?
> 
> 
> Bert Gunter
> 
> "The trouble with having an open mind is that people keep coming along and
> sticking things into it."
> -- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )
> 
> 
> On Mon, Sep 14, 2020 at 11:50 AM Belgers, V. (Vera) <
> v.belgers using amsterdamumc.nl> wrote:
> 
>> Dear sir/madam,
>> Thank you in advance for taking the time to read my question. I am
>> currently trying to conduct a meta-analysis combining parallel and
>> crossover trials. According to the Cochrane Handbook, I can include
>> crossover trials by using t-paired statistics. So far, I have managed to
>> conduct a meta-analysis and forest plot of the parallel trials using the
>> dmetar package, but I did not succeed in including the crossover trials. I
>> do have the raw data of most of these crossover trials.
>> Does anybody know how to add crossover trials to the meta-analysis?
>> With kind regards,
>> Vera Belgers