Abstract 10878: New Bleeding Risk Prediction Model for Patients with Atrial Fibrillation on Direct Oral Anticoagulants
Bibliographic record
Abstract
Introduction: Although several risk scores exist to predict major bleeding events (MBEs) in patients with atrial fibrillation (AF) on anticoagulation therapy, most were derived from warfarin treated patients. Therefore, we sought to develop a new bleeding risk score specifically for AF patients on direct oral anticoagulants (DOAC). Hypothesis: A DOAC-specific risk prediction model will have good discrimination and calibration for 1-year and 3-year MBEs. Methods: AF patients aged ≥ 18 years on DOAC therapy who had at least 12-month follow-up between 2010 and 2017 were extracted from the Northwestern Medicine EHR. The primary endpoint was MBE defined by the International Society on Thrombosis and Hemostasis (ISTH) criteria. Seventy percent of patients were randomly assigned to a derivation cohort and the rest of 30% was used as an internal validation cohort. In the derivation cohort, all significant univariate predictors (p < 0.05) were incorporated into a multivariate Cox proportional hazard model. Harrel’s c-statistics was used for discrimination and Hosmer-Lemeshow test for calibration. Results: A total of 7,642 patients with AF on DOAC (Apixaban 43.9%, Rivaroxaban 36.4%, Edoxaban 0.1%, Dabigatran 19.5%) were analyzed (mean age, 69 ± 12 years: men, 59.5%: non-Hispanic White, 85.8%). During a median follow-up period of 3.3 years (interquartile range 2.0-5.0), 1,283 MBEs (922 in derivation cohort and 361 in validation cohort) were observed. Final model involved age, race, body mass index in addition to all components of CHA 2 DS 2 -Vasc score. Calibration plots at 1 year and 3 years looked good fit and Hosmer-Lemeshow test was not statistically significant. C-statistics of the new model at 1 year and 3 years were 0.70 (95% CI: 0.67 - 0.72) and 0.67 (95% CI: 0.65 - 0.69), both of which were higher than the reported c-statistics from HAS-BLED score. In the validation cohort, we observed the same significant differences in c-statistics at both years. Conclusions: A DOAC-specific risk prediction score had good model performance to quantify bleeding risk in patients with AF on DOAC.
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 imitationNot 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.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
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.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
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".