- Joined
- Jun 27, 2019
- Messages
- 15,953
- Reaction score
- 201,308
The FDA has opened the floodgates to allow all sorts of unsupported "tests" to be used in the US. They don't have enough personnel to review the many pages of documentation and data that a test usually requires, so right now all they require are some preliminary forms.
I have reviewed some test validation data on some of these rapid serology tests from China. They are very poorly supported and should not be used.
Remember Spain wasted $400 million on such inadequate and poorly designed tests from China. They had a sensitivity rate of 36%.
That means if 100 patients were positive, the test would only detect 36.
You would be better off flipping a coin. That would be more accurate than a test with a sensitivity rate of only 36%. The specificity was also very poor. They made no effort to correct for any of the other coronaviruses we may come in contact with ( there are 6 others that may be found in humans, at this time)
There are a few more very good tests coming on the market that are made by legitimate US or European laboratory test makers. They are far superior in accuracy and sensitivity than these worthless tests from China.
No, you wouldn't be "better off" flipping a coin. The 36% were positive so some people got useful information. I know what you're trying to say, though.
While you make it sound like the Chinese tests are flawed for some reason specific to China, it turns out that many of the rapid response tests made in the US (at major medical centers) have similar problems. Not as bad as the early ones purchased by Spain, but still way less sensitive than we need to them to be.
Even the best tests available at universities (Stanford, MIT, Harvard) are not 100% sensitive and no one yet knows what that rate is. One respected virologist who specializes in building vaccines says she assumes these tests are 88-90% accurate at best. Each iteration is a bit better than the last, but no nation was able to solve the Ab testing problem right away.
And no one knows whether the tests work well in real populations until they are used in real populations.
If your state is building its own Ab tests from scratch, there will be glitches there too. It's true that the tests being built in the US are better than the ones originally sent out of China, but at the same time, there are many private "test makers" here in the US selling test kits to private individuals who may in fact be using tests on par with the Chinese ones.
Caveat emptor.