Warning Systems for Undesirable Events during Cancer Treatment: Applying Machine Learning to Administrative Data
Bibliographic record
Abstract
Patients frequently experience undesirable events during cancer treatment. Identifying those at risk of undesirable events could facilitate research on prevention and mitigation. We linked multiple population-level administrative databases in Ontario to identify patients who received systemic therapy for cancer between July 1, 2014, and June 30, 2020. We developed longitudinal machine learning systems to predict undesirable events during cancer treatments from demographic, cancer, treatment, symptom, and laboratory features. We focused on three categories of undesirableevents: 1) acute care use, specifically emergency department visits and hospitalizations; 2) cytopenias, including anemia, thrombocytopenia, and neutropenia; and 3) nephrotoxicity from cisplatin, both acute kidney injury and chronic kidney disease. Machine learning generally outperformed simpler regression-based models and provided risk estimates beyond the initial period after treatment initiation. Sensitivity analyses revealed performance deterioration within some subgroups, most concerningly in some underserved populations. Together, these studies demonstrate the potential of warning systems based on longitudinal machine learning models applied to administrative data to predict undesirable events in cancer care. They also highlight potential challenges to address when the systems are deployed in clinical practice.
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 | 0.007 | 0.041 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.002 | 0.001 |
| 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".