Try to find if for any of Angie's meta-analyses we have the original pubmed search results, to compare apples to apples.