Mortality task

Expired or hospice disposition Daily at 07:00

Potential use: goals-of-care conversations. We prioritize performance over time and the workload of identifying higher-risk hospitalizations. This endpoint includes hospice disposition, so it does not measure death alone.

Retrospective benchmark results do not establish benefit in clinical care.

Models

External AUROC is the mean AUROC across eligible external sites, weighted by test hospitalizations. MIMIC-CLIF remains a public reference. Full-report selections do not change this summary.

Task definition and ranking methods

Endpoint: discharge disposition recorded as expired or hospice; a missing disposition, or one recorded as still admitted, is labelled negative. Prediction time: daily at 07:00 during the first ICU stay.

Endpoint

Expired or hospice disposition

Prediction time

Daily at 07:00

Reporting mode

Pooled landmarks continuous

Task threshold

0.50

High-risk alarm rule

Score ≥0.50

Decision-curve action

Not calculated on maximum risk

Potential action

goals-of-care conversations evaluated retrospectively; not a clinical recommendation

Task ID

icu_daily_mortality

Task class

outcome

Temporal pattern

continuous

Prediction unit

one-per-window

Task type

binary

Label column

label_mortality

The external summary weights each eligible site by test hospitalizations, never daily prediction rows. Scores and intervals in Summary use two decimals; ordering uses the available source precision before display rounding. Full report retains more precision. Formal ranks are pending verification of identical task versions, evaluation protocols, eligible site sets and per-site test partitions across submissions. A change requires a new ranking version and reference-model reevaluation. Exact ties must share a rank once comparability is verified. The current exports do not establish all these conditions. Only per-site confidence intervals are shown in Summary; averaging site confidence limits does not establish an interval for the ranking score.

Summary policy: flair-manuscript-2026-10-18. A missing required external estimate leaves the overall summary unavailable; remaining site reports stay accessible.

Each site's own cell is the pooled-across-landmarks AUROC it published, with a stay-level bootstrap interval computed at the site and not re-derived here. Full report retains its landmark analysis, operating point and cohort tables.

Which result each site publishes: the site a model was trained on publishes its local arm; every other site publishes external validation. A site's own-data result is never presented as external validation.

Intervals: 95% percentile-bootstrap intervals (500 replicates specified) resample hospitalizations at each site. They are not adjusted for multiplicity across tasks, sites, or thresholds, no formal hypothesis tests are performed, and they should not be used to infer statistically significant differences between sites.

Validation tags: external validation means a model is evaluated on participating external site data. Local validation means the model is evaluated on the held-out test split of the site it was trained on.

Want your model evaluated? Request a run →