Using Diagnoses to Estimate Health Care Cost Risk in Canada
Bibliographic record
Abstract
OBJECTIVE: Until recently, the options for summarizing Canadian patient complexity were limited to health risk predictive modeling tools developed outside of Canada. This study aims to validate a new model created by the Canadian Institute for Health Information (CIHI) for Canada's health care environment. RESEARCH DESIGN: This was a cohort study. SUBJECTS: The rolling population eligible for coverage under Ontario's Universal Provincial Health Insurance Program in the fiscal years (FYs) 2006/2007-2016/2017 (12-13 million annually) comprised the subjects. MEASURES: To evaluate model performance, we compared predicted cost risk at the individual level, on the basis of diagnosis history, with estimates of actual patient-level cost using "out-of-the-box" cost weights created by running the CIHI software "as is." We next considered whether performance could be improved by recalibrating the model weights, censoring outliers, or adding prior cost. RESULTS: We were able to closely match model performance reported by CIHI for their 2010-2012 development sample (concurrent R=48.0%; prospective R=8.9%) and show that performance improved over time (concurrent R=51.9%; prospective R=9.7% in 2014-2016). Recalibrating the model did not substantively affect prospective period performance, even with the addition of prior cost and censoring of cost outliers. However, censoring substantively improved concurrent period explanatory power (from R=53.6% to 66.7%). CONCLUSIONS: We validated the CIHI model for 2 periods, FYs 2010/2011-2012/2013 and FYs 2014/2015-2016/2017. Out-of-the-box model performance for Ontario was as good as that reported by CIHI for the development sample based on 3-province data (British Columbia, Alberta, and Ontario). We found that performance was robust to variations in model specification, data sources, and time.
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 | 0.003 | 0.019 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.003 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".