You Can't Believe ANYTHING !

The Atlantic, a respected mainstream literary magazine, says you can't believe anything! (November 2010 issue)

They are not talking about political adverts, but about peer-reviewed MEDICAL RESEARCH as well as Internet sites. On this Blog we've recently discussed Elite Opposition to Online Information and compared it to peer-reviewed journals and books, so these items caught my eye.

Truth Lies Here by Michael Hirschorn, is a hit piece against right-leaning web sites. It starts with the alleged efforts of the "Digg Patriots" to drive down the readership of left-leaning web items by coordinated use of the Digg "bury" option. The reader is lead to believe that left-leaning groups have not use similar tactics. Digg, a website that allows users to recommend web items has since discontinued the "bury" option so the point is moot in any case. Hirschorn goes on to misreport the Sherrod incident (which I discussed here) as well as the Acorn pimp and prostitute caper. He claims the videos were "heavily doctored" when in fact they were simply edited.

The Acorn sting video speaks for itself. According to the NY Times "...two conservative activists pretending to be a pimp and a prostitute used a hidden camera and recorded Acorn employees advising them on how to conceal the source of illegal income and manage 14-year-old Salvadoran prostitutes in the country illegally: 'Train them to keep their mouth shut.'" Perhaps the activists had to visit several Acorn sites before they got that damning video, but it is clear at least one Acorn worker had no problem helping a pimp exploit underage illegal female immigrants. In the Sherrod case the editing was misleading, but the real story was how the Agriculture Department and the NAACP "bit" and fired and condemned Sherrod, despite the fact she had informed her superiors of the true situation and the NAACP had the complete video that proved Sherrod was not a racist but was reporting on a redemptive moment in her career.

Hirschorn blasts the usual suspect, Sarah Palin, for using Twitter shorthand, including "Ground Zero mosque" (it is a cultural center and two blocks away).

Lies, Damned Lies, and Medical Science, by David Freedman, is a longer and much more serious piece that calls into question nearly all medical research. Freedman begins with the fact that Albanian immigrants to Greece have their "perfectly healthy" appendixes removed at a rate three times higher than Greeks, apparently because surgery residents are over-eager to rack up scalpel time. The researchers who uncovered the situation had trouble getting their study published, which led them to do some further investigations of medical research journals.

Many peer-reviewed medical findings are later refuted. This fact may be interpreted in two ways: 1) The system is working and correcting itself, or 2) Why are so many medical studies wrong in the first place?

Well, according to the researcher Freedman interviewed, the problem is the need for researchers to get grants and publish, and that may be accomplished only by getting new and surprising results. This leads them to come up with new theories and then construct research projects that are biased to prove those theories. Even in apparently properly set up randomized trials, results are exaggerated. For example, of 49 most widely used cited research articles over the past 13 years, 34 were retested and 41% of those were shown to be wrong or exaggerated! "Drug studies have the added corruptive force of financial conflict of interest." They hardly ever study the effect of not prescribing any medication. And, when it comes to nutritional studies, "ignore them all" is the best advice! Clearly, this information should be taken into account as we consider government involvement in health care and end-of-life issues, as I discussed here.

But medical research is not especially fact-free, "a remarkably consistent paucity of strong evidence in published economics studies made it unlikely that any of them were right."

Ira Glickstein