A Dynamic Health Status-Based Framework for Personalized Older Adult Welfare Services and Policy: A Nursing-Informed Illustrative Application
ABSTRACT
Objective:
This study proposed a conceptual and computationally structured health status-based framework as a preliminary foundation for future artificial intelligence (AI)-supported older adult welfare decision-making based on multidimensional health status rather than chronological age alone.
Materials and Methods:
Diabetes mellitus, hypertension, and dementia were represented as component-level state vectors. Disease burden, interaction risk, functional impairment, and recent health changes were integrated into an Elderly Health Status Score, which was subsequently converted into a Health-Equivalent Age. The framework was illustrated using two hypothetical cases.
Results:
The illustrative cases demonstrated that chronological age alone may not adequately reflect individual care needs. Using the predefined parameters for illustration, the framework estimated the Health-Equivalent Age of a healthy 70-year-old to be approximately 65 years, whereas a vulnerable 65-year-old was identified as potentially requiring more intensive health and welfare services. These results illustrate the operation of the proposed framework rather than clinically validated estimates. No statistical testing was performed because the cases were hypothetical.
Conclusion:
The proposed framework may complement age-based eligibility criteria by reflecting individual health vulnerability and care needs in nursing, community care, and welfare-service prioritization. Validation of the model weights, state-transition parameters, and predictive performance using longitudinal clinical data is required before practical implementation.
Keywords:
Multimorbidity
geriatric nursing
clinical decision support systems
health policy
VOLUME
,
ISSUE
ADVANCE KNOWLEDGE. SHARE YOUR RESEARCH. PUBLISH WITH PURPOSE—FREE OF CHARGE
Yeditepe Journal of Health Sciences is an international, open-access journal in health sciences with double-blind peer review and no publication fees.