Identifying demographic and home features influencing older adults' home functioning by using machine learning models

Area of Science:

• Gerontology
• Environmental Health
• Health Informatics

Background:

• Functional difficulties significantly impact older adults' independence and quality of life.
• Understanding the factors contributing to these difficulties is crucial for developing targeted interventions.

Purpose of the Study:

• To identify demographic and housing characteristics associated with functional difficulties in older adults.
• To determine the most effective machine learning model for analyzing these associations.

Main Methods:

• Utilized data from American Housing Survey (AHS) questionnaires for individuals aged 65 and older.
• Employed and compared multiple machine learning models, selecting the best-performing one for analysis.
• Focused on 12 functional challenge items to assess difficulties within the home environment.

Main Results:

• The random forest model demonstrated strong predictive performance (accuracy: 85.8%).
• Key predictors of functional difficulties included walking disability, use of mobility aids (cane/walker), bathroom grab bar presence, go-outside-home disability, and self-care disability.
• These factors were more influential than the householder's general health or age.

Conclusions:

• Home modifications and environmental adaptations are vital for enhancing functional abilities and independence in older adults.
• Findings can guide interventions promoting safe and accessible living environments, improving seniors' quality of life.