Dynamic classification for administrative health data case definitions: application to juvenile arthritis
Bibliographic record
Abstract
Introduction: Administrative health databases are important resources for population-based chronic disease research and surveillance, but case definitions to identify disease cases can have low sensitivity and specificity. Case definitions constructed using repeated measurements of diagnoses over time have potential for improved accuracy. In particular, dynamic classification, which uses probabilistic models to classify individuals into disease, non-disease, and indeterminate categories and then updates these classifications as new information becomes available, are promising alternatives to conventional static classification methods for case ascertainment. Purpose & Objectives: The research purpose was to develop and evaluate methods that use longitudinal diagnostic information from population-based administrative databases to improve the accuracy of chronic disease case definitions. The objectives were to: (1) develop dynamic classification methods for retrospective longitudinal administrative data, (2) apply and validate dynamic classification to identify cases of juvenile arthritis (JA), and (3) compare the performance of case definitions developed using dynamic classification with case definitions developed using static approaches for classification. Methods: Dynamic longitudinal discriminant analysis (LoDA) was adapted for retrospective longitudinal administrative data and applied to administrative health data from Manitoba to identify cases of JA. The Pediatric Rheumatology Clinical Database was used for validation. Performance of the JA case definitions constructed with dynamic LoDA was measured using sensitivity, specificity, positive predictive value (PPV), and mean time to classification. Classification accuracy was compared for dynamic LoDA, deterministic case definitions and static LoDA models. Results: The study cohort included 797 children from the clinical database who could be linked to administrative health data from birth to age 16; 386 (48.4%) were JA cases and 411 (51.6%) were non-cases. The dynamic LoDA model with the best fit used a longitudinal binary variable for any-JA related physician visit or hospitalization. It had sensitivity of 0.70, specificity of 0.81, PPV of 0.82, and left 2% of the cohort unclassified after all data were used. On average, it took 9.21 years of data to classify individuals as a JA case or non-case. Both the deterministic case definition and static LoDA model outperformed the best dynamic LoDA model. Conclusion: The results suggest that dynamic classification can produce accurate case definitions using longitudinal information from administrative health data, although the choice of methods and their comparative performance will depend on the characteristics of the disease.
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 | 0.042 | 0.173 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.006 | 0.005 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.001 | 0.002 |
| 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".