Many in SV find it hard to grapple with traditional startups not penetrating health care to a greater degree; in Rise of the Robots (one of the Financial Times and The Economist books of the year) it's lamented for an entire chapter.
Traditionally the level of fundamental research and development displayed by Theranos - in areas of health or national defence - are done for years behind closed doors and then slowly rolled out to the public to mitigate any questioning of the underlying tech since the standards are understandably higher (i.e - it has to work from the start). This stands in contrast to the "move fast, break things" attitude pervasive in start-up culture.
I don't think there's anything necessarily fraudulent about Theranos, from what I've read, which seems to be the not-so-subtle undercurrent in much of the commentary. Rather they flipped the model startups should use in health, which is establish a business model that works and then work on preparatory tech in the background until it's ready to be rolled out. This seems to be largely what Theranos has done, but not what it's purposely chosen to articulate to the public and investors as it surely would have garnered less attention/funding. At this point Theranos seems to be playing the waiting game; waiting for their technology to reach a point where they can make a more transparent case for their business model and change the narrative once they've reached a point that's more aligned with the aspirations that have been articulated by Holmes for the past decade.
Once again, Peter Thiel's approach with Palantir looks to be extremely well executed and one that perhaps Theranos should have emulated.