MétaCan
Menu
Back to cohort
Record W4387610550 · doi:10.1016/s2666-5247(23)00222-7

Accuracy of package inserts of SARS-CoV-2 rapid antigen tests: a secondary analysis of manufacturer versus systematic review data

2023· article· en· W4387610550 on OpenAlexaff
Jacob Bigio, Emily MacLean, Rishav Das, Giorgia Sulis, Mikashmi Kohli, Sarah Berhane, Jacqueline Dinnes, Jonathan J Deeks, Lukas E. Brümmer, Claudia M. Denkinger, Madhukar Pai

Bibliographic record

VenueThe Lancet Microbe · 2023
Typearticle
Languageen
FieldMedicine
TopicSARS-CoV-2 detection and testing
Canadian institutionsOttawa HospitalUniversity of OttawaMontreal General HospitalOttawa Public HealthMcGill University Health Centre
Fundersnot available
KeywordsMeta-analysisCoronavirus disease 2019 (COVID-19)MedicineSystematic reviewSevere acute respiratory syndrome coronavirus 2 (SARS-CoV-2)Cochrane collaborationMEDLINECochrane LibraryInternal medicineBiology

Abstract

fetched live from OpenAlex

BACKGROUND: Rapid antigen tests (RATs) were crucial during the COVID-19 pandemic. Information provided by the test manufacturer in product package inserts, also known as instructions for use (IFUs), is often the only data available to clinicians, public health professionals, and individuals on the diagnostic accuracy of these tests. We aimed to assess whether manufacturer IFU accuracy data aligned with evidence from independent research. METHODS: We searched company websites for package inserts for RATs that were included in the July 2022 update of the Cochrane meta-analysis of SARS-CoV-2 RATs, which served as a benchmark for research evidence. We fitted bivariate hierarchical models to obtain absolute differences in sensitivity and specificity between IFU and Cochrane Review estimates for each test, as well as overall combined differences. FINDINGS: We found 22 (100%) of 22 IFUs of the RATs included in the Cochrane Review. IFUs for 12 (55%) of 22 RATs reported statistically significantly higher sensitivity estimates than the Cochrane Review, and none reported lower estimates. The mean difference between IFU and Cochrane Review sensitivity estimates across tests was 12·0% (95% CI 7·5-16·6). IFUs in three (14%) of 22 diagnostic tests had significantly higher specificity estimates than the Cochrane Review and two (9%) of 22 had lower estimates. The mean difference between IFU and Cochrane Review specificity estimates across tests was 0·3% (95% CI 0·1-0·5). If 100 people with SARS-CoV-2 infection were tested with each of the tests in this study, on average 12 fewer people would be correctly diagnosed than is suggested by the package inserts. INTERPRETATION: Health professionals and the public should be aware that package inserts for SARS-CoV-2 RATs might provide an overly optimistic picture of the sensitivity of a test. Regulatory bodies should strengthen their requirements for the reporting of diagnostic accuracy data in package inserts and policy makers should demand independent validation data for decision making. FUNDING: None.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.053
metaresearch head score (Gemma)0.296
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesMetaresearch
Consensus categoriesnone
DomainCandidate signal: Methods · Consensus signal: none
Study designCandidate signal: Meta-analysis · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.947
Threshold uncertainty score0.283

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0530.296
Meta-epidemiology (narrow)0.0020.002
Meta-epidemiology (broad)0.0110.020
Bibliometrics0.0190.017
Science and technology studies0.0010.002
Scholarly communication0.0040.004
Open science0.0030.003
Research integrity0.0020.002
Insufficient payload (model declined to judge)0.0050.001

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.155
GPT teacher head0.383
Teacher spread0.228 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

Study designMeta-analysis
DomainMethods
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations16
Published2023
Admission routes1
Has abstractyes

Explore more

Same venueThe Lancet MicrobeSame topicSARS-CoV-2 detection and testingFrench-language works237,207