Death ORACL: an algorithm to predict death using insurance claims data
Bibliographic record
Abstract
The inability to identify dates of death in insurance claims data is the United States is a major limitation to retrospective claims-based research. While deaths result in disenrollment, disenrollment can also occur due to changes in insurance providers. We created an algorithm to differentiate between disenrollment from health plans due to death and disenrollment for other reasons. We identified 5,259,735 adults who disenrolled from private insurance between 2007 and 2018. Using death dates ascertained from the Social Security Death Index, inpatient discharge status, and death indicators in the administrative data, 7.6% of all disenrollments were classified as resulting from death. We used elastic net regression to build an algorithm using claims data in the year prior to disenrollment; candidate predictors included medical conditions, individual demographic characteristics, treatment utilization, and structural factors related to health insurance eligibility and coding. Using a predicted probability threshold of 0.9 (selected to reflect the corresponding known prevalence of mortality), internal validation found that the algorithm classified death at disenrollment with a positive predictive value of 0.815, sensitivity of 0.721 and specificity of 0.986 (AUC=0.97). Independent data sources were used for external validation and for an applied example. Code for implementation is publicly available.
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 | 0.005 | 0.015 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.005 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.007 | 0.004 |
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".