Validation of an algorithm based on administrative data to detect new onset of atrial fibrillation after cardiac surgery
Bibliographic record
Abstract
INTRODUCTION: Postoperative atrial fibrillation (POAF) is a frequent complication of cardiac surgery associated with important morbidity, mortality, and costs. To assess the effectiveness of preventive interventions, an important prerequisite is to have access to accurate measures of POAF incidence. The aim of this study was to develop and validate such a measure. METHODS: A validation study was conducted at two large Canadian university health centers. First, a random sample of 976 (10.4%) patients who had cardiac surgery at these sites between 2010 and 2016 was generated. Then, a reference standard assessment of their medical records was performed to determine their true POAF status on discharge (positive/negative). The accuracy of various algorithms combining diagnostic and procedure codes from: 1) the current hospitalization, and 2) hospitalizations up to 6 years before the current hospitalization was assessed in comparison with the reference standard. Overall and site-specific estimates of sensitivity, specificity, positive (PPV), and negative (NPV) predictive values were generated, along with their 95%CIs. RESULTS: Upon manual review, 324 (33.2%) patients were POAF-positive. Our best-performing algorithm combining data from both sites used a look-back window of 6 years to exclude patients previously known for AF. This algorithm achieved 70.4% sensitivity (95%CI: 65.1-75.3), 86.0% specificity (95%CI: 83.1-88.6), 71.5% PPV (95%CI: 66.2-76.4), and 85.4% NPV (95%CI: 82.5-88.0). However, significant site-specific differences in sensitivity and NPV were observed. CONCLUSION: An algorithm based on administrative data can identify POAF patients with moderate accuracy. However, site-specific variations in coding practices have significant impact on accuracy.
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 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 | 0.016 | 0.132 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| 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) | 0.001 | 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, 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".