COIOS
Emergent technology Item new peer-reviewed

Machine learning predicted six-month mortality after Taiwanese care-home admission

An ensemble machine learning model built only from routine intake assessments in 636 Taiwanese long-term care facilities predicted death within 180 days of first admission with an AUROC of 0.90 in a held-out 2024 cohort, where 28.7% of new residents died.

Researchers developed an interpretable gradient-boosting and random-forest ensemble to predict 180-day all-cause mortality among older adults newly admitted to long-term care facilities in Taiwan, using only assessment data available at intake. In 23,901 residents admitted 2020-2023, 22.1% died within six months; in the 2024 temporal validation cohort of 6,216, 28.7% died and the model reached an AUROC of 0.90 (95% CI 0.89-0.91) with calibration weakening at the highest predicted risks. Frequent hospitalisation in the prior six months, impairment in activities of daily living and weight loss were the most influential predictors.

Why it is interesting: A prognostic model that needs no hospital record linkage, validated on a later cohort, with a background mortality rate after care-home entry that is itself notable.

Source
JMIR aging, 2026-09-16
DOI
10.2196/94567
Type
Peer-reviewed article
Design
Retrospective cohort using a nationwide private LTCF registry (636 facilities); development n=23,901 (2020-2023), temporal external validation n=6,216 (2024)
Verdict
New finding
Driver
AI in medicine
Driver
Capacity, funding and care
Driver
Frailty and multimorbidity at the oldest ages