4 ms·
What numbers are you using for sensitivity and specificity?
by DanBC 7y ago
What numbers are you using for sensitivity and specificity?
- marcell 7y agoIt depends on the study, I've been using their reported confidence intervals. The two California studies (Santa Clara and Los Angeles) used the same kit, which has 2 false positives of 371 tests, and 10-20% false negative rate.
- nodamage 7y agoIt's worth noting that is the manufacturer claim but has not held up to independent validation. Specifically, the Premier Biotech/Hangzhou Biotest Biotech test was validated by a Chinese provincial CDC and found 4 false positives out of 150. [1] It was also validated by the COVID-19 Testing project and found 3 false positives out of 108. [2] The Biomedomics test used in the Miami Dade study was also validated by the COVID-19 Testing Project and found 14 false positives out of 107. [2] Hence I would recommend taking the results of the California and Florida studies with a huge grain of salt as the prevalence rates they found were within the false positive rates of the tests used. [1] https://imgcdn.mckesson.com/CumulusWeb/Click_and_learn/COVID19_CDC_Evaluation_Report.pdf https://imgcdn.mckesson.com/CumulusWeb/Click_and_learn/COVID... [2] https://covidtestingproject.org/ https://covidtestingproject.org/