But my thoughts were the same as Dr. Cohen's: it can't be that bad: 4 of 5 supplements lacking the principle ingredient?
Then the agency analyzed the products using DNA bar coding, a type of genetic fingerprinting that the agency has used to root out labeling fraud in the seafood industry.
... The technology can single out which plants a supplement contains by identifying its unique DNA.
Dr. Cohen at Harvard said that the attorney general’s test results were so extreme that he found them hard to accept. He said it was possible that the tests had failed to detect some plants even when they were present because the manufacturing process had destroyed their DNA.
But that does not explain why the tests found so many supplements with no DNA from the herbs on their labels but plenty of DNA from unlisted ingredients, said Marty Mack,
Actually, maybe it does. There's been a fair bit of complaint that "herbal medicine" doses are relatively random since the potency of a herb will vary quite a bit. If they've starting using extracts to get a relatively uniform chemical and then bulking it up with fillers, there might not be much DNA from the original plant, which is consistent with the lab's findings. The story doesn't mention whether the lab looked for the chemical signature of active ingredients. (And, unlike science news reporting, I have no way to look up the lab's actual results or rationales.)
That doesn't excuse using fillers not mentioned on the label, of course.
This doesn't lower my confidence in herbal medicine pills because I don't have any. Maybe fresh herbs (chamomile or the like), but even there I've not found much requirement for them.