Hacker News Reader

Comment by Robin_Message | Hacker News Reader

A 6% false positive rate? You don't want anyone taking this without probable cause then, else you're going to get a metric crap-ton of people going to their doctor's and having to have a full workup done.

Edit to clarify: My point was that just because it was cheap doesn't mean it'll ever be sold to the public. Apart from anything else, consider the psychological impact of discovering you (may) have HIV from a test kit you picked up from next to the Oreos on a whim.

View on HN

This is true of most tests. When he likelihood of a false positive is significant in relation to the expected incidence, you need corroboration such as symptoms developing before choosing to run the tests.

Sidebar: This is misunderstood by people criticizing doctors who refuse to run tests without corroborating evidence. It's not just the money/scarce resources, it's the problem of false positives dominating actual positives.

If you do some searching around you will find 3% to 5% false positive for blood test is not uncommon.

Of course at $1/15m/test, a battery of tests is practically built in to the price. If the instructions are to "take three cards and put a drop of blood on each," and that it isn't a positive result unless all three agree, you've solved that problem.

Works for medical professionals, but I think to sell this to the general public, those three cards should be baked in together to avoid confusion.

Highly unlikely that three runs of the same test are fully independant, i.e. they might match all on the same antibody which is similar to an HIV one but not due to a HIV. You likely only prevent operation goof-ups like "oops, spilled a bit of blood from the needle of my bro here ..." (or something less fancy and more likely).

Which is why in practice you run a different test to confirm results.

Wikipedia says that the false positive rate of the standard ELISA test is only 1.5% (thats < 6%) http://en.wikipedia.org/wiki/HIV_test#Accuracy_of_HIV_testin...

and a "positive" result should be confirmed by a Western Blot test. http://en.wikipedia.org/wiki/HIV_test#Interpreting_antibody_...

Reply on news.ycombinator.com