Age-Related Differences in Risk Factors for Long-Term Care Certification in Japan: A Decision Tree Analysis Spanning...

Area of Science:

• Gerontology
• Public Health
• Biostatistics

Background:

• Aging societies face challenges in ensuring healthy aging and extending health span.
• Long-term care (LTC) certification is a critical indicator of health status in older populations.
• Identifying early risk factors for LTC certification is essential for timely interventions.

Purpose of the Study:

• To analyze risk patterns for long-term care (LTC) certification within three years.
• To investigate these patterns in a rapidly aging population in Japan.
• To utilize the Kihon Checklist (KCL) and machine learning for risk prediction.

Main Methods:

• Analysis of data from adults aged 65+ in Iiyama City, Japan.
• Application of Exhaustive Chi-squared Automatic Interaction Detector (CHAID) decision trees.
• Independent variables included age, sex, and six KCL domains (physical strength, nutrition, oral function, isolation, memory, mood).

Main Results:

• Age was the strongest predictor of LTC certification.
• In individuals aged 80+, low cognitive function and depression were significant risk factors.
• In younger older adults, physical weakness showed a stronger association with LTC certification.
• Three consistent risk patterns were identified across datasets.

Conclusions:

• Age-specific risk stratification is crucial for effective interventions.
• Targeted strategies can be developed based on age-related risk profiles.
• Preventing LTC certification requires tailored approaches for different age groups within the elderly population.