Administrative Hospitalization Database Validation of Cardiac Procedure Codes
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.
Bibliographic record
Abstract
BACKGROUND: Although cardiac procedures are commonly used to treat cardiovascular disease, they are costly. Administrative data sources could be used to track cardiac procedures, but sources of such data have not been validated against clinical registries. OBJECTIVES: To examine accuracy of cardiac procedure coding in administrative databases versus a prospective clinical registry. SAMPLE: We examined a total of 182,018 common cardiac procedures including percutaneous coronary intervention (PCI), coronary artery bypass graft (CABG) surgery, valve surgery, and cardiac catheterization procedures during fiscal years 2005 and 2006 across 18 cardiac centers in Ontario, Canada. RESEARCH DESIGN: Accuracy of codes in the Canadian Institute for Health Information (CIHI) administrative databases were compared with the clinical registry of the Cardiac Care Network. RESULTS: Comparing 17,511 CIHI and 17,404 registry procedures for CABG surgery, the positive predictive value (PPV) of CIHI-coded CABG surgery was 97%. In 6229 CIHI-coded and 5885 registry-coded valve surgery procedures, the PPV of the administrative data source was 96%. Comparing 38,527 PCI procedures in CIHI to 38,601 in the registry, the PPV of CIHI was 94%. Among 119,751 CIHI-coded and 111,725 registry-coded cardiac catheterization procedures, the PPV of administrative data was 94%. When the procedure date window was expanded from the same day to ±1 days, the PPV was 96% (PCI) and exceeded 98% (CABG surgery), 97% (valve surgery), and 95% (cardiac catheterization). CONCLUSIONS: Using a clinical registry as the gold standard, the coding accuracy of common cardiac procedures in the CIHI administrative database was high.
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 imitationNot 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.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
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.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it