Meta-Analysis Resources

Tools for Those Who Summarize the Evidence Base

Resources and networking for those who conduct or interpret meta-analyses related to any phenomenon that is gauged in multiple studies.

My aim is to review important discussion on meta-analysis. It includes discussions of missing article,  missed sub-group analyses agreement/disagremment between meta-analyses, and between meta-analyses and big RCT:s. I know there is a literature, but need suggestions on where to start. Am I right when I suggestb that Ioannides is one of the analysts taht have added to this topic?

Why? I teach MA and I have put togerther technical compendoum. One of my students suggested I add a number of good examples to look forward to as they plow the compendium.

Any suggestions?

Views: 53

Replies to This Discussion

Yes, John Ioannidis is a good place to start. I was particularly stunned by his group's demonstration of how dramatically different excluded literature results were from included literature. He compared gene-association results from the West with those searched in Mandarin Chinese and found a reversed association.

The 2005 book on publication bias (led by Rothstein) covers many of aspects such as you list in intimate detail.

Hi and thanks! I have the Rothstein et al. book on my shelf but I did not screen it.  Thanks for the reminder. Would you think it worthwhile to try to review the discussion on metaanalysis validity so far? I guess one man can do the work (myself) but I feel I need help with input. And psychological  support. The iioannides work I was thinking about was published, if I am correct,  in the NEJM, some years ago. But there might be much more today.

Hi back!

I believe that meta-analysis is no longer seriously challenged, except perhaps in camps that doubt science as a means to the truth.  Instead, the hallmarks of strong systematic reviews (including meta-analysis) are beginning to become conventional.

One indication of this progress is the development of methodological quality scales for meta-analysis. For example, Shea, Grimshaw et al. developed the AMSTAR scale, which is seeing broad use. Merely reading through the final items on the scale is enough to educate someone what procedures good reviews should follow. Here's an open link to the original scholarly work on this scale.


© 2018   Created by Blair T. Johnson.   Powered by

Badges  |  Report an Issue  |  Terms of Service