The antibody and antigen tests are highly accurate. Here is a link to the dozen or so approved serology tests:
https://www.fda.gov/medical-devices/emergency-situations-medical-devices/eua-authorized-serology-test-performance
And a snippet from a Shoreland report:
“FDA has published an assessment of performance results for antibody tests. Per Shoreland’s earlier recommendations, the Architect SARS-CoV-2 IgG (Abbott) and the Elecsys Anti-SARS-CoV-2 (Roche) had both sensitivity and specificity over 99.5%. VITROS Immunodiagnostic Products Anti-SARS-CoV-2 (Ortho-Clinical Diagnostics) and the SARS-COV-2 ELISA (IgG) (EUROIMMUN) also had positive and negative predictive values above 99.5%. All the aforementioned tests should be regarded as accurate and reliable. All other high throughput tests and all rapid point of care antibody tests reviewed had less than optimal performance.”
Up thread someone mentioned that there have been post-market tests of Abbott’s ID Now rapid diagnostic test (for active infection) showing a high level of false negatives, including from a study by Cleveland Clinic.
Those results were based on samples being put in viral transport media, which diluted the sample, which in turn impacted testing accuracy. FDA and Abbott eliminated this option from the product directions/recommendations, now the only approved method is to place the swab directly in the machine. Specificity and sensitivity of that process are in line with data that resulted in the test being approved.
These are the types of issues that can arise when products are rushed to market with emergency use authorization. It’s one thing when it’s a dx or serology test, another thing entirely for a vaccine or treatment.