EUGENIZATION.

From today’s NY Times:

Another Study Finds Heart Risks in a Diabetes Drug.

Too bad “Just eat less sugar” isn’t easily patentable or easily commodified and sold.

In the interest of fair and balanced reporting, it is true that meta-analyses suck pretty badly as far as quality goes (when speaking of scientific studies).  The main problem with meta-analyses is that all the data points are thrown together and treated with equal weight, with complete disregard of the quality of the study that it came from.  In other words, if you picked poorly constructed studies, then your conclusions from the meta-analysis will be flawed (you see this effect prominently with the Rhea et al “Multiple vs. single set” meta-analysis).  So in all fairness to Glaxosmith Kline, it is possible that Dr. Nissen’s meta-analysis is highly flawed.

Of course, you could just eat less sugar and not become insulin resistant, develop diabetes, and have to go on the drug…

[I'm all for better living through chemistry, by the way.  But if you're taking an asprin everyday because you hit your head with a mallet each morning, I think you're probably barking up the wrong tree.]