Due to a “busy” cat-schedule, the regular Friday posting was delayed for one day. If you have a problem with it, the individual with whom you can take it up is indefinitely taking a nap. So good luck with that!
1) First, a post from the past: The Dark Age Mighty Whitey.
2) Weird search query of the week: “kristin kreuk with glasses.”
3) Comment of the week, in response to “Your genes, your rights – FDA’s Jeffrey Shuren misleading testimony under oath”, there were many good comments. But I’ll highlight this one from Dan:
Furthermore, all CLIA certified operations have to pass very stringent proficiency testing to maintain their certification. There is no such oversight for DTC operations,
Joe has already noted that 23andMe, deCODEme and Pathway are CLIA-certified. I can tell you that Navigenics is as well. You’re right that some of the scurrilous operators aren’t CLIA-licensed, and that should be stopped – but that isn’t the responsibility of the FDA.
Even those of us who oppose FDA involvement in DTC typically agree that regulation of analytic validity is important (CLIA is actually a terrible system in many ways, but there currently isn’t a viable alternative). We also believe that companies should be punished by the FTC if they engage in false or unethical advertising. This isn’t about regulation vs anarchy; it’s about paternalistic, unnecessary regulation that will destroy an industry and harm consumers vs sensible regulation that will protect consumers without crippling innovation.
and no way to ensure that the testing is actually accurate and reproducible (I have heard through the grapevine that there is poor inter-lab correlation for many of these DTC SNP arrays).
This is absolutely wrong. Comparisons between 23andMe and deCODEme data, for instance, routinely show concordance of >99.9%. When I compared the 23andMe and deCODEme files of Times journalist Mark Henderson last year I found a discrepancy rate of around 1 in 14,000 – substantially better than the vast majority of routine clinical tests.