4 ms·
> The downside to this is that it deprives their classmates Assume probability of false positive is 1%. If you are tested weekly, there is a 41% chance you'll
by nanis 5y ago
> The downside to this is that it deprives their classmates
Assume probability of false positive is 1%. If you are tested weekly, there is a 41% chance you'll falsely test positive at least once in a year. If you are tested every workday, 93% chance. This is assuming there is no malfeasance.
If you assume a false positive probability of 5%, the probabilities go up to 93% and 100%.
The negative effects of false positives are the same whether kids are using soda or something else is causing them. At some point, the costs have to be weighed against the benefits.
Here is a false positive/false negative simulator I put together last year (thrice silently banned -- so I stopped trying to do more after restoring it one last time) helps you explore what happens with various scenarios of infection rates and false positive/negative probabilities.
Linking to the discussion to provide context for the app:
https://www.covid2020.icu/false-positive-false-negative-simulator/ https://www.covid2020.icu/false-positive-false-negative-simu...
- BenjiWiebe 5y agoMy workspace immediately retests if you get a positive. While you wait.
- nanis 5y agoPCR or something else? If PCR, is the number of cycles pre-set or is that a choice variable?
- dtech 5y agoThat seems like a good way to miss infections (and get people back to work, which is probably the point). The tests seem to have a high false-negative rate. Assuming independence and 25% FN rate, a second test will miss an additional 20% of infections of all people tested.
- yarcob 5y agoSame at my kids school. Whenever they have a positive antibody test, they just do it again. I think that the antibody tests are sketchy in the first place, but if they don't even stick to the protocol and don't follow up positive antibody tests with PCR tests, it's almost pointless.
- makomk 5y agoPart of the reason it took so long to roll out these lateral flow tests is that it was hard to find ones with an acceptably low false positive rate - it needed to be well below 1% for exactly the reason you outline. (Apparently PHE estimates it at less than 1 in 1000: https://publichealthmatters.blog.gov.uk/2021/03/30/covid-19-reintroducing-confirmatory-pcr-testing/ https://publichealthmatters.blog.gov.uk/2021/03/30/covid-19-... ) Even when the rate of Covid infections in the UK was much lower than it currently is, retesting of people using PCR suggested that the large majority of positives were in fact true positives, at least amongst the adults retested. It's possible that the story is different for kids, but on the other hand they're not vaccinated in the UK whereas many adults are...
- nanis 5y agohttps://www.bmj.com/content/371/bmj.m4469 https://www.bmj.com/content/371/bmj.m4469 and https://www.bmj.com/content/372/bmj.n706 https://www.bmj.com/content/372/bmj.n706 both make interesting reading.