Key takeaways from this new article, for people blocked by the WSJ paywall:
"That letter cited “deficient practices” in five categories. In one of them, hematology, the problems found by inspectors posed “immediate jeopardy to patient health and safety,” regulators said.
The test that inspectors found Theranos kept doing despite the erratic quality-control results was a hematology test that measures how long it takes blood to clot, the people familiar with the matter said.
The test is an important part of the treatment regimen for patients who are at risk of strokes or have blood disorders...
People who have seen the Theranos inspection report said it seems to indicate the company ignored quality-control results for the blood-clotting test that fell short of its own criteria.
The results produced in the quality-control checks repeatedly deviated from the lab’s typical result by more than two standard deviations, those people said...
At Theranos, there were seven quality-control failures in a single day, said a person familiar with the report...
Dr. Hamill said any doctors who received prothrombin time test results from the Theranos lab during the six-month period shouldn’t rely on them and should have their patients retested as soon as possible. “Those results are not worth anything,” he said...
The federal inspectors also concluded that the Theranos lab in California used expired reagents, or substances added to blood samples to elicit chemical reactions, for some prothrombin time tests, according to the people familiar with the inspection report.
Those people said the report also found that the lab’s director didn’t have responsibility for the lab’s quality-control program and the lab’s quality-control manager was unqualified..."
EDITED TO ADD: Actually, the really key sentence is this one: "Theranos spokeswoman Brooke Buchanan said: “The PT/INR issues identified by CMS related to tests run on conventional equipment using venipuncture samples.” A CMS spokeswoman declined to comment." Because that means that these quality control problems were not due to Theranos' "revolutionary" finger-prick tests, they were from regular old venipuncture tests running on a standard machine that were giving wildly incorrect results, expired reagents, no QC in the lab, etc. Bad data is worse than no data. If they couldn't get old-fashioned labs right, no wonder they were having trouble with the finger-prick stuff.
That is rather scary. Quality control is a serious requirement to any competent lab. With physicians making decisions by results labs transmit, there is an inherent need of verifiable trustworthiness to the results.
A patient may be subjected to life altering medication or treatments, which is not a trivial responsibility that can be shrugged off with erratic processes or lack thereof.
A lab director's responsibility includes ownership of quality processes, reviewing abnormal results after the fact to see if proper processes were followed (such as communicating with the respective provider), reviewing financial success, and much more. To shrug off quality control and operate as if reputable is not just unethical, but incompetent. Such operations should cease immediately.
Disclaimer: I have experience in the medical lab industry.
"That letter cited “deficient practices” in five categories. In one of them, hematology, the problems found by inspectors posed “immediate jeopardy to patient health and safety,” regulators said.
The test that inspectors found Theranos kept doing despite the erratic quality-control results was a hematology test that measures how long it takes blood to clot, the people familiar with the matter said.
The test is an important part of the treatment regimen for patients who are at risk of strokes or have blood disorders...
People who have seen the Theranos inspection report said it seems to indicate the company ignored quality-control results for the blood-clotting test that fell short of its own criteria.
The results produced in the quality-control checks repeatedly deviated from the lab’s typical result by more than two standard deviations, those people said...
At Theranos, there were seven quality-control failures in a single day, said a person familiar with the report...
Dr. Hamill said any doctors who received prothrombin time test results from the Theranos lab during the six-month period shouldn’t rely on them and should have their patients retested as soon as possible. “Those results are not worth anything,” he said...
The federal inspectors also concluded that the Theranos lab in California used expired reagents, or substances added to blood samples to elicit chemical reactions, for some prothrombin time tests, according to the people familiar with the inspection report.
Those people said the report also found that the lab’s director didn’t have responsibility for the lab’s quality-control program and the lab’s quality-control manager was unqualified..."
EDITED TO ADD: Actually, the really key sentence is this one: "Theranos spokeswoman Brooke Buchanan said: “The PT/INR issues identified by CMS related to tests run on conventional equipment using venipuncture samples.” A CMS spokeswoman declined to comment." Because that means that these quality control problems were not due to Theranos' "revolutionary" finger-prick tests, they were from regular old venipuncture tests running on a standard machine that were giving wildly incorrect results, expired reagents, no QC in the lab, etc. Bad data is worse than no data. If they couldn't get old-fashioned labs right, no wonder they were having trouble with the finger-prick stuff.