Showing posts with label statistics. Show all posts
Showing posts with label statistics. Show all posts

Monday, January 10, 2011

Three Whole-Grain Portions Daily May Lower Cardiovascular Risk

Yeah--Right!
    This one's a beaut.

" Daily consumption of 3 portions of whole-grain foods (WGF) is linked to lower cardiovascular disease (CVD) risk in healthy, middle-aged people mainly by lowering blood pressure (BP), according to the results of a randomized controlled dietary trial reported in the American Journal of Clinical Nutrition. (Am J Clin Nutr. 2010;92:733-740)


Given the role that fiber plays in slowing down the digestion of carbohydrates, everything these folk found can be explained by lowered blood glucose levels. Here's where it gets really entertaining:


"Although the mechanisms by which WGFs decrease BP remain unclear, our findings have important public health implications and provide a sound scientific basis for advising the daily consumption of 3 servings of WGF to combat CVD."


In fact, what they found was a statistical clue to do some science. They admit that the science eludes them and they can't figure out why WGF loading would lower BP.  Nor do they seem disposed to look for the science or consider the possibility that their assumptions are flawed.
    Another indicator of the shortage of three-digit IQs in the nutritional sciences.

Tuesday, December 7, 2010

The Amazing Dr. Tai

If this weren't so sad it would be funny. Check out this American Diabetes Association journal article:
http://care.diabetesjournals.org/content/17/2/152.abstract.
    On the way to "a mathematical model for the determination of total areas under curves," Dr. Tai believes he's invented what is actually a centuries-old numerical integration technique--freshman mathematics material. What's sad is that none of the journal's reviewers or editors caught it.
    It says something about the education needed to get into medicine. The lack of fundamental math skills also explains why they're so bad at interpreting statistical data. It also helps explain why they don't get the simple math of (Glucose-in-blood ~ Carbohydrates-ingested).

Dr. Tai and "The Tai Model" get this month's Delusional award.