When will this shit end?

[zahidf]

Interesting

 

 

 

[steviewevie]

we really need those antibody tests.

[Toilet Duck]

8 minutes ago, steviewevie said:

we really need those antibody tests.

The problem is we have them and they are about 95% sensitive and 95% specific. For most tests, this is pretty decent accuracy and once the % of the population infected goes over about 15%, they start to become useful. They just give false positives more often than true positives when the % infected is under that rate. It's explained better than I can do it Here

Edited May 6, 2020 by Toilet Duck

[steviewevie]

3 minutes ago, Toilet Duck said:

The problem is we have them and they are about 95% sensitive and 95% specific. For most tests, this is pretty decent accuracy and once the % of the population infected goes over about 15%, they start to become useful. They just give false positives more often than true positives when the % infected is under that rate. It's explained better than I can do it Here

Ok...so...excuse my stupidity...but if fairly accurate why isn't the govt pushing out mass antibody testing now? It won't help people make decisions for themselves as each could be wrong, but would give an overall picture?

[Wooderson]

10 minutes ago, Toilet Duck said:

The problem is we have them and they are about 95% sensitive and 95% specific. For most tests, this is pretty decent accuracy and once the % of the population infected goes over about 15%, they start to become useful. They just give false positives more often than true positives when the % infected is under that rate. It's explained better than I can do it Here

Glasters (and any big event) dependent upon these numbers IMO. Can't see gates opening next June unless theres been a huge national testing programme/people are passported in as immune.

[zahidf]

 

 

[zahidf]

I wonder if record store gigs will be first allowed back? Relatively small, can limit attendance, easier to distance in theory with markings.

[Toilet Duck]

11 minutes ago, steviewevie said:

Ok...so...excuse my stupidity...but if fairly accurate why isn't the govt pushing out mass antibody testing now? It won't help people make decisions for themselves as each could be wrong, but would give an overall picture?

Unfortunately no, it won't help people make decisions, because the positive test result has more chance of being wrong than right at the moment. When more of the population has been infected, then the accuracy of the test means that it will pick up true positives more frequently. If we are at about 4-5% of the population infected, then at the current accuracy at of the test., a positive result will be correct 53% of the time. At 1% infected, the test result will be correct about 18% of the time. The accuracy changes if you target specific populations (i.e., those who have been own close contact with a known case), so they can be used for some situations with a better chance of picking up positive cases, but random population testing requires a certain % of the population to be positive before a test with that level of accuracy is actually useful. The PCR test is far more sensitive (maybe too sensitive, there are issues discharging people form COVID wards as they keep testing positive, but they are surrounded by virus and the test may be picking that up rather than virus from the patient). 

[Toilet Duck]

10 minutes ago, Wooderson said:

Glasters (and any big event) dependent upon these numbers IMO. Can't see gates opening next June unless theres been a huge national testing programme/people are passported in as immune.

Ah sure Nal's asteroid will have wiped us out by then anyway! More seriously, I think we'll have learned how to live with it by then. 

[steviewevie]

2 minutes ago, Toilet Duck said:

Unfortunately no, it won't help people make decisions, because the positive test result has more chance of being wrong than right at the moment. When more of the population has been infected, then the accuracy of the test means that it will pick up true positives more frequently. If we are at about 4-5% of the population infected, then at the current accuracy at of the test., a positive result will be correct 53% of the time. At 1% infected, the test result will be correct about 18% of the time. The accuracy changes if you target specific populations (i.e., those who have been own close contact with a known case), so they can be used for some situations with a better chance of picking up positive cases, but random population testing requires a certain % of the population to be positive before a test with that level of accuracy is actually useful. The PCR test is far more sensitive (maybe too sensitive, there are issues discharging people form COVID wards as they keep testing positive, but they are surrounded by virus and the test may be picking that up rather than virus from the patient). 

but we can't know the % of population infected without the antibody test...?

[Toilet Duck]

2 minutes ago, steviewevie said:

but we can't know the % of population infected without the antibody test...?

We can, but via the PCR test. Expanding the testing criteria will identify more cases. Once we reach a threshold of % infection, then the antibody tests become useful (or we generate more accurate antibody tests!). 

[steviewevie]

ok...and PCR gives us current infection rate?

[Toilet Duck]

Just now, steviewevie said:

ok...and PCR gives us current infection rate?

That's where the case numbers are coming from at the moment alright. 

[ace56blaa]

40 minutes ago, zahidf said:

 

 

Work in a cinema and this can only be bad news if it goes through, most films have been delayed so there'll be next to nothing to show and all zero hour employees will have to return to work with little or no hours because of a lack of business. 

[Toilet Duck]

12 minutes ago, Cream Soda said:

I am confused.  How does the percentage of correct positive results change depending on the percentage of population infected?  

So, if you know somebody is positive and you use a test that is 95% sensitive and 95% specific, it will correctly diagnose the disease 95% of the time (sensitivity) and will classify you as positive 95% of the time (specificity). However, if you don't know whether somebody is positive or not, then it does not give the correct result 95% of the time as a different question is being asked (what's the probability that the test will give the correct result in a random samples of people). This is where the prevalence of the disease becomes important (because at low levels of infection, more people have not been infected, so the false positive rate increases). 

[stuartbert two hats]

They've worked out how to do massages safely!

 

 

[Crazyfool01]

Some lockdown measures changed from Monday ... maybe ... according to boris 

[zahidf]

 

 

[zahidf]

10 minutes ago, ace56blaa said:

Work in a cinema and this can only be bad news if it goes through, most films have been delayed so there'll be next to nothing to show and all zero hour employees will have to return to work with little or no hours because of a lack of business. 

Tenent is still due for middle of july and they may screen some more indie films? Theres definitely a backlog 

[Simsy]

26 minutes ago, Cream Soda said:

I am confused.  How does the percentage of correct positive results change depending on the percentage of population infected?  

The Twitter thread @Toilet Duck linked to earlier does a great job of explaining it. Basically if a smaller percentage of people are infected, most of the positive results will be false positives.

[steviewevie]

39 minutes ago, crazyfool1 said:

Some lockdown measures changed from Monday ... maybe ... according to boris 

we'll probably be allowed out twice a day...and that will be it.

[Crazyfool01]

Just now, steviewevie said:

we'll probably be allowed out twice a day...and that will be it.

So a completely pointless announcement ... just one for the sake of it ... in my eyes 

[zahidf]

Track and trace apps ain't going well

 

 

 

 

[stuie]

6 minutes ago, crazyfool1 said:

So a completely pointless announcement ... just one for the sake of it ... in my eyes 

They are going to lay out the next steps so probably more than just an announcement for announcement sake!

[steviewevie]

9 minutes ago, crazyfool1 said:

So a completely pointless announcement ... just one for the sake of it ... in my eyes 

maybe you can get 50cm closer to other people...maybe you can meet a few other people...maybe garden centres can open...I don't know, but I don't think it will be much yet.