However, if one is going to make comparisons, I think it's better to use the same draw of blood for both lab companies. Or at least within an hour of each other.
There are too many variables at play, and numbers can change between morning and afternoon even on the same day depending on the test.
Repeated tests and obtaining standard deviations would also be a good idea.
Respectfully, you are making a broad, inaccurate statement and assumedly without a medical background.
While it's true that some numbers can change even on the same day (Gassée specifically mentioned glucose as an example of a lab value that can fluctuate quite a bit), the concentration of red blood cells and platelets is not affected by what you had for lunch and definitely should not vary this much. Keep in mind that the difference Gassée saw was was almost enough to trigger unnecessary therapeutic phlebotomy.
Sure. Except I never said or meant to imply RBC's and platelets change on the same day. Part of the context of the comment, were my own tests.
So the question is, why wouldn't one want to remove as many confounding factors as possible?
"Taken one hour apart, the Stanford and Theranos HCT numbers differ by about 7%: 44.1 Theranos vs. Stanford 41.1."
He was comparing the draws of the same day mainly
To get around this, they supposedly dilute the samples to a point where the concentrations of analytes are well below the machines minimum operating specifications.
2) Theranos invents a new low volume blood draw technology and test mechanism (which contains many subtests)
3) The test they use to augment their limited test is a commercial product that works on larger volumes of blood
4) because their test doesn't actually work all that well (I actually don't know enough about this part, as their work is mostly secret) they have to pivot a bunch of of their tests to using a different company's test machine
5) Those other commercial products actually do work with smaller volumes of blood, diluted to larger volumes, in case you can't get a large volume from a patient
There isn't anything wrong with Theranos diluting blood to make their tests work with another commercial product (this is a legitimate use). However, it does seem to suggest that the technology that hyped Theranos up to a multibillion value isn't as good as was claimed; instead, Theranos is mostly just another conventional provider of diagnostic tests. If they're just another conventional provider, it sort of kills the whole startup story and investment, because why would you invest in a disruptor if you can make money from the establishment?
That said, it's common to see startups start with a really ambitious goal, fail, then pivot to become more conventional companies with decent but not spectacular revenue streams.
This is like if every time you fetched an Uber, a yellow cab showed up! Except for one market out of 200, where they run five black cars, in a small SF neighborhood.
Since you 99% sure that other labs have problems, are you suggesting that all the other labs rely on technology that also have a dearth of peer-review published results?
[1] http://jama.jamanetwork.com/article.aspx?articleid=2110977
> As of January 5, 2015, a search in PubMed using Theranos as a search term identified affiliations for only 2 unrelated articles coauthored by Theranos Inc employees, although these 2 reports do not offer insights about their company.
[2] http://www.newyorker.com/magazine/2014/12/15/blood-simpler
> Clarke argues that finger-stick blood tests aren’t reliable for clinical diagnostic tests; because the blood isn’t drawn from a vein, the sample can be contaminated by lanced capillaries or damaged tissue. Holmes strongly disagrees: “We have data that show you can get a perfect correlation between a finger stick and a venipuncture for every test that we run.” When I asked for evidence, I was sent a document by Daniel P. Edlin, Theranos’s senior product manager, titled “Select Data.” It purported to show favorable results from numerous comparison tests. I asked Edlin if the tests had been conducted by an independent third party. He replied by e-mail: “The clinical tests were conducted by a combination of Theranos and external labs,” but he wouldn’t say which ones.
> When I asked Holmes for evidence that her tests were independently audited, she said that there have been “tens” of audits and “external third-party comparisons” of Theranos’s tests, including those done by the hospital groups that are adopting its finger-prick tests and the pharmaceutical companies that have contracted with Theranos for testing their products. Holmes says that Theranos is certified in forty-eight states, with two more applications pending, under the federal Clinical Laboratory Improvement Amendments of 1988. Under the amendments, laboratories like hers must be certified before they can perform tests for the general public, and their performance is evaluated three times a year by the College of American Pathologists. Holmes also pointed me to a pilot study published by Hematology Reports, an online-only peer-reviewed journal; she is listed as a co-author. The report, released in April, concluded that Theranos tests “correlated highly with values obtained” from standard lab tests.
What I am suggesting is that other labs do have variances in results, on certain lab tests. The OP did say that there was a 7% variation in Platelet results in successive days, using the Stanford lab.
Quest Diagnostics have had previous settlements with DoJ for "inaccurate and unreliable" lab test kits. [0]
[0] http://www.justice.gov/opa/pr/quest-diagnostics-pay-us-302-m...