Comparison of methods to identify individuals prescribed opioid analgesics for pain
Bibliographic record
Abstract
INTRODUCTION: While identifying opioid prescriptions in claims data has been instrumental in informing best practises, studies have not evaluated whether certain methods of identifying opioid prescriptions yield better results. We compared three common approaches to identify opioid prescriptions in large, nationally representative databases. METHODS: We performed a retrospective cohort study, analyzing MarketScan, Optum, and Medicare claims to compare three methods of opioid classification: claims database-specific classifications, National Drug Codes (NDC) from the Centers for Disease Control and Prevention (CDC), or NDC from Overdose Prevention Engagement Network (OPEN). The primary outcome was discrimination by area under the curve (AUC), with secondary outcomes including the number of opioid prescriptions identified by experts but not identified by each method. RESULTS: All methods had high discrimination (AUC>0.99). For MarketScan (n=70,162,157), prescriptions that were not identified totalled 42,068 (0.06%) for the CDC list, 2,067,613 (2.9%) for database-specific categories, and 0 (0%) for the OPEN list. For Optum (n=61,554,852), opioid prescriptions not identified totalled 9,774 (0.02%) for the CDC list, 83,700 (0.14%) for database-specific categories, and 0 (0%) for the OPEN list. In Medicare claims (n=92,781,299), the number of opioid prescriptions not identified totalled 8,694 (0.01%) for the CDC file and 0 (0%) for the OPEN list. DISCUSSION: This analysis found that identifying opioid prescriptions using methods from CDC and OPEN were similar and superior to prespecified database-specific categories. Overall, this study shows the importance of carefully selecting the approach to identify opioid prescriptions when investigating claims data.
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 | 0.048 | 0.106 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.003 |
| Bibliometrics | 0.006 | 0.003 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.002 | 0.001 |
| 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".