Identifying Patients With Axial Spondyloarthritis in Large Datasets: Expanding Possibilities for Observational Research
Bibliographic record
Abstract
Objective Observational research of axial spondyloarthritis (axSpA) is limited by a lack of methods for identifying diverse axSpA phenotypes in large datasets. Algorithms were previously designed to identify a broad spectrum of patients with axSpA, including patients not identifiable with diagnosis codes. The study objective was to estimate the performance of axSpA identification methods in the general Veterans Affairs (VA) population. Methods A patient sample with known axSpA status (n = 300) was established with chart review. For feasibility, this sample was enriched with veterans with axSpA risk factors. Algorithm performance outcomes included sensitivities, positive predictive values (PPV), and F1 scores (an overall performance metric combining sensitivity and PPV). Performance was estimated with unweighted outcomes for the axSpA-enriched sample and inverse probability weighted (IPW) outcomes for the general VA population. These outcomes were also assessed for traditional identification methods using diagnosis codes for the ankylosing spondylitis (AS) subtype of axSpA. Results The mean age was 54.7 and 92% were male. Unweighted F1 scores (0.59–0.74) were higher than IPW F1 scores (0.48–0.65). The full algorithm had the best overall performance (F1IPW 0.65). The Early Algorithm was the most inclusive (sensitivityIPW 0.90, PPVIPW 0.38). The traditional method using ≥ 2 AS diagnosis codes from rheumatology had the highest PPV (PPVIPW 0.84, sensitivityIPW 0.34). Conclusion The axSpA identification methods demonstrated a range of performance attributes in the general VA population that may be appropriate for various types of studies. The novel identification algorithms may expand the scope of research by enabling identification of more diverse axSpA populations.
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 | 0.148 | 0.332 |
| Meta-epidemiology (narrow) | 0.002 | 0.002 |
| Meta-epidemiology (broad) | 0.003 | 0.003 |
| Bibliometrics | 0.008 | 0.011 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.005 | 0.006 |
| Open science | 0.004 | 0.009 |
| Research integrity | 0.003 | 0.003 |
| Insufficient payload (model declined to judge) | 0.002 | 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".