MétaCan
Menu
Back to cohort
Record W2785817299 · doi:10.1136/bmj.j5779

Concerns about composite reference standards in diagnostic research

2018· article· en· W2785817299 on OpenAlex

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.
fundA Canadian funder is recorded on the work.

Bibliographic record

VenueBMJ · 2018
Typearticle
Languageen
FieldMathematics
TopicStatistical Methods and Bayesian Inference
Canadian institutionsMcGill University Health Centre
FundersCanadian Institutes of Health Research
KeywordsComposite numberImperfectReference modelReference dataComputer scienceTest (biology)StatisticsClass (philosophy)Reference valuesEconometricsData miningArtificial intelligenceMathematicsAlgorithmMedicine

Abstract

fetched live from OpenAlex

Composite reference standards are used to evaluate the accuracy of a new test in the absence of a perfect reference test. A composite reference standard defines a fixed, transparent rule to classify subjects into disease positive and disease negative groups based on existing imperfect tests. The accuracy of the composite reference standard itself has received limited attention. We show that increasing the number of tests used to define a composite reference standard can worsen its accuracy, leading to underestimation or overestimation of the new test’s accuracy. Further, estimates based on composite reference standards vary with disease prevalence, indicating that they may not be comparable across studies. These problems can be attributed to the fact that composite reference standards make a simplistic classification and then ignore the uncertainty in this classification. Latent class models that adjust for the accuracy of the different imperfect tests and the dependence between them should be pursued to make better use of data

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.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.003
metaresearch head score (Gemma)0.027
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesMetaresearch
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Theoretical or conceptual · Consensus signal: Theoretical or conceptual
GenreCandidate signal: Methods · Consensus signal: none
Teacher disagreement score0.540
Threshold uncertainty score0.999

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0030.027
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0010.000

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 head0.269
GPT teacher head0.550
Teacher spread0.281 · 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