[BostonCard]

Stanford did test the performance of the tests, and did the analysis with both the manufacturers specifications and its own numbers.  I think my beef is that it is hard to be sure of the specificity when they only tested 300 patients without disease; it produces a 95% confidence interval of about 1%, which might not seem like much, but is the difference between the results they claim and pretty much all the positive tests being false positives.

BC

[BostonCard]

Stanford did test the performance of the tests, and did the analysis with both the manufacturers specifications and its own numbers.  I think my beef is that it is hard to be sure of the specificity when they only tested 300 patients without disease; it produces a 95% confidence interval of about 1%, which might not seem like much, but is the difference between the results they claim and pretty much all the positive tests being false positives.

BC