1

I need to make a network meta-analysis of diagnostic studies.

Is there a good explanation to do it with R?

Giuseppe Biondi-Zoccai
  • 2,244
  • 3
  • 19
  • 48
Ahmed Mohamed Afifi
  • 11
  • 3
  • You can also look at these related posts: https://stats.stackexchange.com/questions/203745/which-is-the-best-method-for-meta-analysis-of-diagnostic-test-accuracy-studies https://stats.stackexchange.com/questions/337001/how-to-create-interaction-effects-for-diagnostic-network-meta-analysis – Giuseppe Biondi-Zoccai Apr 02 '18 at 08:37

1 Answers1

2

How you do your network meta-analysis in R will depend on the nature of your input data, number of competing diagnostic tests being compared in each study, whether or not you have any covariate information, whether or not you have study arm-level information only or study-level information, etc. All of this will determine what methodology you need to use for your network meta-analysis. Once the methodology is clear, it might be easier to answer your question.

In the meantime, perhaps the following reference might give you a good starting point:

http://journals.sagepub.com/doi/abs/10.1177/0962280216669182

Isabella Ghement
  • 18,164
  • 2
  • 22
  • 46
  • 1
    Other interesting works might be: https://www.ncbi.nlm.nih.gov/pubmed/29548843 https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5784560/ https://repository.library.brown.edu/storage/bdr:674079/PDF/ https://abstracts.cochrane.org/2017-global-evidence-summit/bivariate-network-meta-analysis-diagnostic-test-accuracy-studies – Giuseppe Biondi-Zoccai Apr 02 '18 at 08:42