|
|
|
|
|
by ImaCake
1758 days ago
|
|
A quick word on test attributes and why we still don't have a more rapid test than PCR. Tests for disease rely on two numbers: 1. "sensitivity" - the proportion of people with COVID who get a positive test 2. "specificity" - the proportion of people without COVID who get a negative test. COVID, despite the media attention, is a rare disease compared to the number of people tested. Say we have 10,000 people tested for COVID at the airport. We have 99% sensitivity and 99% specificity. And we know 100 people have COVID (1% prevalence) in this group. Our test would find 99/100 of the positive cases. But it would also find an additional 99 false positives! It also misses one true positive case. Which would be disasterous for the quarantine measures in place in Australia and NZ. In short, even a gold standard rapid test is not enough, although at 99% specificity and sensitivity it would be somewhat useful. This is a lot less intuitive than it looks. The companies pushing rapid antigen tests at the start of the pandemic would have know better, but they chose to lie to the public about this. https://en.wikipedia.org/wiki/Sensitivity_and_specificity |
|