I think Jaana's partly correct. While these journals have peer-review panels, that doesn't mean that every article is peer-reviewed. Some studies are printed for interest's sake, and there's usually a letters section where members can submit their thoughts. They are identified as non-peer-reviewed, of course.
The problem that stems from this is that from time to time, a quack writes a letter to the editor, gets his letter published, and then goes off into la-la-land and interprets it as an endorsement. I've seen many quacks who back up their fraudulent claims with 'references' that are merely letters to the editor or human interest articles, but have impressive titles, such as JAMA, CMAJ, JAPA, Lancet... It's deceptive, of course, but why give them this raw material?