Referee report. For: Derivation and validation of a mortality risk prediction model in older adults needing home care: Updating the RESPECT (Risk Evaluation for Support: Predictions for Elder-Life in their Communities Tool) algorithm for use with data from the interRAI Home Care Assessment System [version 1; peer review: 1 approved with reservations]
Bibliographic record
Abstract
Background Despite an increasing number of risk prediction models being developed within the healthcare space, few have been widely adopted and evaluated in clinical practice. RESPECT, a mortality risk communication tool powered by a prediction algorithm, has been implemented in the home care setting in Ontario, Canada, to support the identification of palliative care needs among older adults. We sought to re-estimate and validate the RESPECT algorithm in contemporary data. Methods The study and derivation cohort comprised adults living in Ontario aged 50 years and older with at least 1 interRAI Home Care (interRAI HC) record between April 1, 2018 and September 30, 2019. Algorithm validation used 500 bootstrapped samples, each containing a 5% random selection from the total cohort. The primary outcome was mortality within 6 months following an interRAI HC assessment. We used proportional hazards regression with robust standard errors to account for clustering by the individual. Kaplan–Meier survival curves were estimated to derive the observed risk of death at 6 months for assessment of calibration and median survival. Finally, 61 risk groups were constructed based on incremental increases in the observed median survival. Results The study cohort included 247,377 adults and 35,497 deaths (14.3%). The mean predicted 6-month mortality risk was 18.0% and ranged from 1.5% (95% CI 1.0%–1.542%) in the lowest to 96.0 % (95% CI 95.8%–96.2%) in the highest risk group. Estimated median survival spanned from 36 days in the highest risk group to over 3.5 years in the lowest risk group. The algorithm had a c-statistic of 0.76 (95% CI 0.75-0.77) in our validation cohort. Conclusions RESPECT demonstrates good discrimination and calibration. The algorithm, which leverages routinely-collected information, may be useful in home care settings for earlier identification of individuals who might be nearing the end of life.
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 | 0.011 | 0.186 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.004 | 0.003 |
| Science and technology studies | 0.003 | 0.001 |
| Scholarly communication | 0.003 | 0.003 |
| Open science | 0.004 | 0.004 |
| Research integrity | 0.008 | 0.004 |
| Insufficient payload (model declined to judge) | 0.497 | 0.240 |
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".