Preapproval and postapproval diagnostic test accuracy of food and drug administration-authorized rapid antigen SARS-CoV-2 tests used according to instruction: A systematic review and meta-analysis

BACKGROUND: Manufacturers claim high sensitivity for rapid antigen SARS-CoV-2 tests on product labels, yet systematic reviews report considerably lower sensitivity.

OBJECTIVES: This study aimed to describe study characteristics and compare the sensitivity and specificity of United States Food and Drug Administration (FDA)-approved rapid antigen SARS-CoV-2 tests in preapproval vs. postapproval studies.

METHODS: Methods include systematic review and meta-analysis.

DATA SOURCES: Data sources include FDA website, Medline, Embase, and Google Scholar.

STUDY ELIGIBILITY CRITERIA: Study eligibility criteria include diagnostic test accuracy studies according to the instruction for use.

PARTICIPANTS: Participants include patients with symptoms of COVID-19.

TESTS: Test includes rapid antigen SARS-CoV-2 tests.

REFERENCE STANDARD: Reference standard includes RT-PCR.

ASSESSMENT OF RISK OF BIAS: Assessment of risk of bias was conducted using the Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) tool.

METHODS OF DATA SYNTHESIS: Methods of data synthesis include bivariate binomial-normal restricted maximum likelihood random-effect meta-analysis and meta-regressions applying the delta method and likelihood-ratio tests.

RESULTS: We identified postapproval studies for 13 of 61 (21%) rapid antigen tests, of which nine tests had eligible studies. The analysis incorporated 13 preapproval studies (591 patients with COVID-19, 3155 participants) and 26 postapproval studies (2765 patients with COVID-19, 12 444 participants). The pooled sensitivity for preapproval and postapproval studies was 86.5% (95% CI: 83.3-89.1%) and 84.5% (95% CI: 81.2-87.3%), respectively. The absolute difference was 2.0% (95% CI: -1.9% to 6.2%) and (0%, 95% CI: -0.6% to 0.6%) for sensitivity and specificity, respectively. Two of the nine tests had lower sensitivity in postapproval studies.

DISCUSSION: Our study found that sensitivity estimates from postapproval studies on FDA-approved rapid antigen tests are largely consistent with manufacturers' estimates. However, for two of the nine tests, postapproval sensitivity was lower than the manufacturers' high estimates. Differences in sensitivity observed in prior systematic reviews likely result from variations in study populations, not bias in study conduct. Given that 79% of FDA-approved rapid antigen tests lacked postapproval studies, ongoing evaluations are needed to ensure alignment with clinical expectations.