139. UTILIZATION OF CANADIAN HEALTHCARE ADMINISTRATIVE DATABASES TO ACCURATELY IDENTIFY CASES OF ANCA-ASSOCIATED VASCULITIS
Bibliographic record
Abstract
Background: ANCA-associated vasculitis (AAV) is a group of diseases that include granulomatosis with polyangiitis (GPA), eosinophilic granulomatosis with polyangiitis (EGPA) and microscopic polyangiitis (MPA). Given the rarity of these multisystem autoimmune diseases they are difficult to study. Large healthcare administrative databases provide a unique opportunity to address this. In Ontario, Canada, a province with over 14 million people, there is a single provincial healthcare database. Local hospitals assign an ICD10 code to discharges and upload data to this provincial system. To date there has been no validation of the ICD10. Our objective was to validate local coding to determine if searching the provincial system by ICD10 code would result in patients being identified correctly. Methods: Validation at hospital level was done by searching Hamilton Health Sciences and St. Joseph’s Healthcare administrative databases from 2011 to 2017. The databases were searched for: AAV (GPA, MPA or eGPA), arteritis, polyarteritis nodosa, giant cell arteritits, and nephritis. When patients had multiple visits one was selected randomly for review. Two algorithms were developed to identify patients. The first (algorithm one) used ICD10 coding alone. The second (algorithm two) combined ICD10 codes with ANCA serology. Chart review was completed to confirm or refute the diagnosis of AAV. Results: A search of the hospital databases led to 417 unique patients being identified. In the sample 52.0% had AAV (GPA n = 112, MPA n = 88, and eGPA n = 15). Using algorithm one, 113 patients (89%) were coded correctly as having AAV with the algorithm having a sensitivity of 52.6% (CI 46.0-59.2%), specificity of 93.1% (CI 88.6-96.2%), and a positive predictive value (PPV) of 89.0% (CI 82.7-93.2%). Patients who were incorrectly coded were most commonly coded as arteritis (30.7%). Table 1 details the accuracy of algorithm one for each type of AAV. Algorithm two, which included ANCA status, improved sensitivity to 98.6% (CI 96.0-99.7) at the expense of reducing specificity to 85.6% (CI 80.0-90.2%). Conclusion: Our results show that the specificity of the ICD10 coding for AAV is quite high. The sensitivity however is quite low, particularly for identifying patients with MPA. Combining ANCA serology with ICD10 coding improved this. Disclosures: None Accuracy of Coding Using ICD10 CI (Confidence interval)
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 | 0.003 | 0.029 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.011 | 0.016 |
| Science and technology studies | 0.002 | 0.000 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| 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".