[bane]

Theory: They had some early tech that looked promising in some specific clinical circumstances that through dumb luck were the ones that were tested early on. Holmes, as an almost engineer with no particular experience in anything, thinks medicine is just like engineering. It didn't work in the general case or was less accurate than acceptable for medicine, and so now Theranos is basically just doing normal lab tests, but subsidizing the costs with VC money and trying to make it up at scale or a breakthrough in the original technological approach at some point in the future. Holmes is spinning a convoluted web of misdirection, nonspecific, fractured answers and non-answers until either or both of those things happen.

The questions are easy to answer and wouldn't give away any trade secrets, but she pretty systematically refuses to give simple direct answers to very simply questions. If answering these simple questions, like how many tests they can run using their tech, would give away trade secrets then their tech isn't likely very interesting. The copy coming out of the company insists they are able to do the things they claim, but it doesn't appear they have any particularly commercial ready technology to go to market with.

Holmes is caught in a bit of a bind, she used her family's powerful connections to raise absurd amounts of money on nothing at all and now has to deliver or face the wrath of those same powerful connections. She's also managed to fill the board up with big names through her family connections who have absolutely no medical background at all (except for Dr. Foege) in an effort to give some kind of legitimacy.

Here's an example of her answering simple questions to get a feel:

https://www.youtube.com/watch?v=A8qgmGtRMsY

https://www.youtube.com/watch?v=MBs-oj7U-bo

Here's an example of how she answers things, (exact quote) "So we have developed hundreds of tests over the course of the last 12 years that can run on a tiny sample using proprietary Theranos technology..." The critical missing components to this statement:

a) To what accuracy compared to traditional tests?

b) All of the tests on one sample, or one test per nanotainer sample?

c) Are all of those tests current tests that Theranos would run on a sample? (R&D could be generated 10 tests a week, but 90% of those tests could be poor, or could not be clinically useful)

She then goes on about taking the data they've generated for their FDA submission and putting it into the public domain and being a leader in transparency. She goes on to talk about all the various technologies Theranos has developed, but it's only the nanotainer tube that's being tested and blah blah blah it's all very complicated and none of the answers clarify or complete any other answer - it's like getting a good image from a shattered mirror.

Simple yes/no answers turn into minutes of non-answers. This isn't difficult, but she makes it absolutely excruciating.

Here's an exceprt from her bio from theranos.com

"Theranos' breakthrough advancements have made it possible to quickly process the full range of laboratory tests from a few drops of blood - instead of numerous tubes - and at unprecedented low costs, and are now directly accessible to people and their physicians through Theranos Wellness Centers opening nationwide."

Except not a single part of this has been demonstrated to be true in any way.

[adevine]

It's interesting that the only test cleared is one for herpes. Conceptually, a finger prick sample would very clearly work for binary, "you've got it or you don't" tests, like STD tests (e.g. you've been able to buy home HIV tests for years that just require a few drops of blood from a finger).

Most blood tests, though, are about detecting whether levels of a substance are in a particular range - milligrams of cholesterol per deciliter of blood, mg glucose per deciliter, etc. Taking a much smaller sample is pretty much guaranteed to make the plus/minus error range larger, and capillary blood from a finger prick will have a greater percentage of contamination from intracellular fluid.

Even glucose meters you can use at home are much less accurate than glucose readings taken with a venous draw in a lab. In that case, the benefit to a diabetic of being able to do frequent home testing outweighs the loss of accuracy. If you're going to get work done by a lab, though, you are going to want it to be as accurate as possible, and in that scenario I don't see how a finger prick will be able to compare to a venous draw.