AI-Enabled Person-Centered Care for Unmet Home- and Community-Based Service Needs Among Chinese Older Adults: Latent Profiles and Risk Prediction.

Authors

Yushan Yu, Dehua Zhu, Mimi Fujitaka, Jun Zhang, Wei-Hong Zhang, Mirko Petrovic, Chun Chen, Liangliang Cheng

Published in

Journal of advanced nursing. Sep 24, 2026. Epub Sep 24, 2026.

Abstract

To identify latent profiles of unmet home- and community-based service needs among Chinese older adults and to develop an interpretable machine learning model to support person-centered care.
Cross-sectional, multisite study.
We recruited 487 community-dwelling adults aged 60 years and older from multiple provinces. Nineteen home- and community-based service items across health services, social services and family caregiver support domains were assessed. A descriptive Jaccard network summarized co-occurrence across services, latent class analysis identified subgroups with distinct patterns of unmet need, and TabNet classified participants into the combined higher-unmet-need group using eight selected demographic and health variables.
Pairwise co-occurrence among unmet service needs was widespread. Three latent class profiles were identified: Generally Met (69.6%), Healthcare-focused Unmet Needs (24.9%) and Extensive Unmet Needs (5.5%). TabNet showed moderate discrimination in the internal cross-validation analysis (AUC 0.767, 95% CI 0.720-0.808).
Latent profiles of unmet service needs were distinct. The interpretable eight-variable TabNet model may support first-stage community nursing screening by prioritizing older adults for subsequent professional assessment. External validation is required before routine implementation.
The model may help nursing teams prioritize older adults for follow-up assessment; referral and care coordination should be determined through professional assessment and shared decision-making.
The study addressed unmet service needs among older adults and may inform resource allocation and person-centered care planning in community-based care.
TRIPOD+AI statement.
No patient or public contribution.

PMID:
42779446
Bibliographic data and abstract were imported from PubMed on 24 Sep 2026.

Read full publication at:
Please sign in to see all details.

Stats

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.