MétaCan
Menu
Back to cohort
Record W2910054248 · doi:10.1002/pds.4708

Extreme restriction design as a method for reducing confounding by indication in pharmacoepidemiologic research

2019· article· en· W2910054248 on OpenAlexafffund
Matthew H. Secrest, Robert W. Platt, Colin R. Dormuth, Dan Château, Laura E. Targownik, Rui Nie, Carla M. Doyle, Sophie Dell’Aniello, Kristian B. Filion

Bibliographic record

VenuePharmacoepidemiology and Drug Safety · 2019
Typearticle
Languageen
FieldMathematics
TopicAdvanced Causal Inference Techniques
Canadian institutionsUniversity of ManitobaManitoba HealthUniversity of British ColumbiaMcGill UniversityJewish General Hospital
FundersCanadian Institutes of Health ResearchPfizerMcGill UniversityAmgen
KeywordsMedicineConfoundingPropensity score matchingHazard ratioCohort studyConfidence intervalInternal medicineProportional hazards modelPopulationObservational studyCohortEnvironmental health

Abstract

fetched live from OpenAlex

PURPOSE: Confounding by indication is a concern in observational pharmacoepidemiologic studies, including those that use active comparator, new user (ACNU) designs. Here, we present a method of restriction to an indication, which we call "extreme restriction," to reduce confounding in such studies. METHODS: As a case study, we evaluated the effect of proton pump inhibitors (PPIs) on hospitalization for community-acquired pneumonia (HCAP). PPI use has been associated with increased HCAP risk, but this association likely results from confounding by indication due to gastroesophageal reflux disease (GERD). Using the UK's Clinical Practice Research Datalink, we compared the risk of HCAP within 180 days between PPI users and histamine-2 receptor antagonist (H2RA) users in an ACNU cohort using Cox proportional hazard models with a time-fixed exposure definition adjusted for high-dimensional propensity score deciles. We then performed the same analysis on an "extremely-restricted" cohort of incident nonsteroidal anti-inflammatory drug (NSAID) users, some of whom received PPIs for prophylaxis. Because PPIs were given as prophylaxis in this population, confounding due to GERD should be limited. We compared effect estimates between ACNU and restricted cohorts to evaluate confounding in both analyses. RESULTS: In the ACNU cohort, PPIs were associated with an increased risk of HCAP (hazard ratio [HR]: 1.25; 95% confidence interval [CI]: 1.05, 1.47), but this association was not present in the restricted cohort (HR: 1.06; 95% CI: 0.75, 1.49). CONCLUSIONS: Restriction to a single indication for treatment may reduce confounding by indication in studies conducted in distributed data networks and other large databases.

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)0.339
metaresearch head score (Gemma)0.407
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesMetaresearch
Consensus categoriesMetaresearch
DomainCandidate signal: Methods · Consensus signal: Methods
Study designCandidate signal: Theoretical or conceptual · Consensus signal: none
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.661
Threshold uncertainty score0.816

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.3390.407
Meta-epidemiology (narrow)0.0030.002
Meta-epidemiology (broad)0.0050.008
Bibliometrics0.0030.004
Science and technology studies0.0020.007
Scholarly communication0.0030.003
Open science0.0040.006
Research integrity0.0070.007
Insufficient payload (model declined to judge)0.0080.001

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 head0.456
GPT teacher head0.578
Teacher spread0.122 · 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; the direct Gemma label and the distilled Codex classifier agree on what is shown here.

Study designTheoretical or conceptual
DomainMethods
GenreMethods

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

Citations19
Published2019
Admission routes2
Has abstractyes

Explore more

Same venuePharmacoepidemiology and Drug SafetySame topicAdvanced Causal Inference TechniquesFrench-language works237,207