Population-level individualized prospective prediction of opioid overdose using machine learning
Bibliographic record
Abstract
The opioid overdose epidemic has rapidly expanded in North America, with rates accelerating during the COVID-19 pandemic. No existing study has demonstrated prospective opioid overdose at a population level. This study aimed to develop and validate a population-level individualized prospective prediction model of opioid overdose (OpOD) using machine learning (ML) and de-identified provincial administrative health data. The OpOD prediction model was based on a cohort of approximately 4 million people in 2017 to predict OpOD cases in 2018 and was subsequently tested on cohort data from 2018, 2019, and 2020 to predict OpOD cases in 2019, 2020, and 2021, respectively. The model's predictive performance, including balanced accuracy, sensitivity, specificity, and area under the Receiver Operating Characteristics Curve (AUC), was evaluated, achieving a balanced accuracy of 83.7, 81.6, and 85.0% in each respective year. The leading predictors for OpOD, which were derived from health care utilization variables documented by the Canadian Institute for Health Information (CIHI) and physician billing claims, were treatment encounters for drug or alcohol use, depression, neurotic/anxiety/obsessive-compulsive disorder, and superficial skin injury. The main contribution of our study is to demonstrate that ML-based individualized OpOD prediction using existing population-level data can provide accurate prediction of future OpOD cases in the whole population and may have the potential to inform targeted interventions and policy planning.
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 | 0.001 | 0.006 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".