pdlamb

I'm too cheap to have read the article, but I'm suspicious. The last couple of years have showed that studies that challenge the accepted facts are even more likely to be wrong (or at least not replicable) than the usual 40-60% of published papers, even with appropriate peer review, that cannot be replicated.

Then there's the fact this is a meta-analysis. No new research went into this publication, it's just statistics. Remember the progression: lies, damned lies, statistics? There are a couple kinds of meta-analyses in general. One tries to include all the relevant studies, one carefully selects studies. It's rarely obvious which kind leads to a given paper. It's an awful lot of work, requiring an awful lot of subject expertise, to tell if the papers that were accepted for meta-analysis were carefully selected to be the ones that will support the desired outcome, while having more-or-less plausible reasons for excluding the rest.

Finally, there's the personal blog page "press release" to publicize the paper. That rings all my alarm bells.

So this result may hold up over time. I'd check back in 5-10 years and see if it's changed the field of cardiology. Until then, I'll regard it as a conclusion to watch to see if it holds up, or if it's like so many other publications that try to make a big splash before they're reduced to something the complete review papers have to include and refute.