The Impossibility of 100% Specificity in Rapid Diagnostic Test Reporting

Le résumé fourni par la source

Background: In commercial materials for rapid diagnostic tests, it is still common to find claims of “100% specificity”. From a statistical and epistemological point of view, this kind of statement is deeply problematic. Aims: This paper examines how justified those claims really are and what they mean in concrete clinical and public health settings. Methods: We analyze the statistical foundations of specificity estimation, apply the Mayo-Spanos severity framework, and review three commercial rapid antigen test listings to illustrate reporting practices. Results: Claims of 100% specificity are either logically falsified by observed false positives or unsupported by finite validation samples. The rule of three shows that zero false positives in n negatives only rule out a false-positive rate above 3/n. Severity analysis reveals that even with 210 negative samples, the claim that specificity >99.5% is supported with only 65% severity. In low-prevalence settings, assuming 100% specificity inflates positive predictive value (PPV) from 66% to 100%. Conclusion: Reporting 100% specificity is scientifically untenable. Regulatory and editorial standards could realistically require confidence intervals and, at the very least, discourage unqualified point estimates.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.