Is PubMed a credible source?

So... I wrote an OIA that lead to an article in The Dominion Post which has comments on the article enabled.

There was one acupuncturist who listed a source that just screams "bias". That being the Acupuncture Research Resource Center (ARRC). The do have a Latest Research section though so I checked that out.

Given the sources are from PubMed I took the time to check the first 3 on the list.

These were;

The first thing I noticed was that the list hadn't been updated since May 2011 so either research has ceased or the site just isn't being maintained. Not a great start for the promoter of the site.

Of the first 3 I found the following Red Flags in the methodology;

  • None had a placebo arm that used sham acupuncture.
  • They start from the assumption that acupuncture is a real thing.
  • Two of the three were focused on comparing one form of pseudoscience to another.
  • Two of the three were Chinese papers. There's a well documented bias that shows amazingly few papers on acupuncture that come out of China show a negative (or even neutral (from memory)) outcome to the study. I can dig that out if people want to see the reference.

My question about this isn't about the studies. My question is about PubMed.

What are the criteria on getting published in PubMed? In the past I've considered PubMed to be a credible source. Looking at the quality of these studies I'm finding myself less likely to use it as a primary source.

Trackback URL for this post:

Darcy Cowan has just pointed

Darcy Cowan has just pointed out to me on G+ that PubMed is just a clearing house for papers. It appears my previous impression of the site was unjustified. :/