but sure let's base policy on its reliability and continue to shout as loud as we can at people who are cautious that these results can be tested. What utter cvnts those critics are
Is 82% accuracy good? I guess this isn’t life or limb stuff but I don’t think it’s amazing. I mean I wouldn’t buy something from a shop if they said there was an 8/10 chance of it working. Would you? If Bob said there was an 8/10 chance of your new toilet flushing would you hire him?
shatner’s: only very marginally. The fact that 82% of tests are confirmed means the accuracy of the tests is likely as near as makes no difference to 82%.
This is like the “getting his by a bus within 28 days of a positive test” thing. Yes some people who test positive for covid die of something unrelated within 28 days. But the number is utterly marginal.
Laz the tests are 99.7% accurate (on this measure).
So 3 false positives per thousand.
For 80%ish confirmed by PCR that means you have 12 real positives identified per 1,000 (which they’re only 75%accurate on so incidence is actually 16 per 1,000 in the community).
If the level of Covid in the community falls then the tests will become “less accurate” over time. i.e. if there are 4 people in our thousand sample who have Covid:
You have to test, it's a basic moral obligation. If you don't then you are a social traitor.
Clearly not, but refusing to engage with even the most basic of containment measures, or claims that doing so somehow makes this situation worse, is the mark of a solipsistic qunt.
0
0
Bit weird that they are presenting this as being reassuring when it suggests the test is still the equivalent of blindfolded dart chucking.
0
0
That is so ridiculously unreliable
but sure let's base policy on its reliability and continue to shout as loud as we can at people who are cautious that these results can be tested. What utter cvnts those critics are
0
0
I’m not sure what way this cuts. However, I’m pretty sure it isn’t how the BBC are presenting it; that is to say the tests are very reliable.
I guess they’re better than tossing a coin but certainly not a great test. And false positives are disruptive.
0
0
It's fine. Stop complaining and take the test. It's our only way out of this.
Expected better from a doctor, tbh.
0
0
Sorry I’ll strike myself off.
0
0
well they’re 82% accurate so that’s a good bit better than tossing a coin
and they’re confirmed by a PCR test
0
0
undetected rising infections can also be disruptive crypto as india and brazil are finding out at great cost
0
0
They should vaccinate faster. And use better vaccines.
0
0
Is 82% accuracy good? I guess this isn’t life or limb stuff but I don’t think it’s amazing. I mean I wouldn’t buy something from a shop if they said there was an 8/10 chance of it working. Would you? If Bob said there was an 8/10 chance of your new toilet flushing would you hire him?
0
0
It's confirmed by PCR. People aren't isolating unnecessarily for that long. And it's the best we've got.
Would you rather we did nothing and end up like India and Brazil?
Expected better from a doctor, tbh.
0
0
Ex Dr. I self erased just now. It’s the commercial bar for me now.
0
0
Its not so low as 82% accurate unless the same stat applies to false negatives.
0
0
82% is way more accurate than the howler monkeys were claiming previously and is a decent hit rate if you can quickly clear up oddities with a PCR.
0
0
heh @ howler monkeys
0
0
The issue of false negatives is a bubble burster for the smug ones
bit just shout and shout louder
0
0
80% is a good first.
0
0
Yep. Don’t test. Live your lives
0
0
Better not to know if your aunt dies after you give her a covid kiss.
0
0
You have to test, it's a basic moral obligation. If you don't then you are a social traitor...
0
0
As Rhialto said the fact that 82% of positive tests were confirmed does not mean the test is 82% accurate. It could be higher, could be lower
0
0
this is farcical, though it only applies to the rapid tests
0
0
shatner’s: only very marginally. The fact that 82% of tests are confirmed means the accuracy of the tests is likely as near as makes no difference to 82%.
This is like the “getting his by a bus within 28 days of a positive test” thing. Yes some people who test positive for covid die of something unrelated within 28 days. But the number is utterly marginal.
0
0
Laz the tests are 99.7% accurate (on this measure).
So 3 false positives per thousand.
For 80%ish confirmed by PCR that means you have 12 real positives identified per 1,000 (which they’re only 75%accurate on so incidence is actually 16 per 1,000 in the community).
If the level of Covid in the community falls then the tests will become “less accurate” over time. i.e. if there are 4 people in our thousand sample who have Covid:
3 will test positive correctly
3 will false positive
1 will false negative
and suddenly the tests are down to 50% accuracy.
0
0
Heh arbiter. You’re using facts and worse
maths on laz?!
0
0
Heh at Laz. Without more data the 82% thing tells you absolutely nothing about overall accuracy. It could be very accurate, it could be absolute shit.
0
0
Of course in a no covid environment 100% of positives would be false. Tests are useless in a low covid environment
0
0
You have to test, it's a basic moral obligation. If you don't then you are a social traitor.
Clearly not, but refusing to engage with even the most basic of containment measures, or claims that doing so somehow makes this situation worse, is the mark of a solipsistic qunt.
Join the discussion