Bayes Covid
TO UNDERSTAND THE ISSUE AT A HIGH LEVEL, TWO SIMPLISTIC EXAMPLES...
EX1: Of 1000 ASYMPTOMATIC...
if rapidTest *sensitivity* is 80%... 20% will have false negatives
assume 1% of pop are positive ( @ 2% - false positives)
if rapidTest *specificity* is 100% all infection-free would test negative)
from the pre-test probability (1%)
10 in 1000 will have it, 990 will not
considering the 80% sensitivity
8 will test positive,
2 will test negative (false negatives)
Of the 990, 100% test negative ( *specificity* is 100%)
Positive predictive value = 100%
Negative predictive value = 99.8% (990/992 = not having Covid-19 if the test is negative)
EX2: Of the 1000 SYMPTOMATIC with PRE-TEST probablity at 30% (Sensitivity and Specificity remaining the same)
300 will have the disease
240 (80%) testing positive,
60 (20%) testing negative (falsely)
https://www.statnews.com/2020/08/20/covid-19-test-accuracy-supplement-the-math-of-bayes-theorem/
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7269418/
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7315940/