Browsing by Subject "Predictive Model"

Now showing 1 - 1 of 1
  • No Thumbnail Available
    Item
    Behavioral Health Needs of Older Adults Living in Poverty: Machine Learning-Based Predictive Models
    (2023-01-13) Hong, Saahoon; Yi, Eun-Hye G.; Walton, Betty; Kim, Hea-Won
    To develop contextually sensitive and effective services for older adults in poverty, this study aimed to identify the characteristics and patterns of older adults’ BH service needs, compared to those of middle-aged adults. The findings suggest that employment is the most important predictor for classifying older adults with behavioral health needs, followed by adjustment to trauma, independent living, legal system involvement, sleep, disability, transportation, social skills, and self-care. Interestingly, gender and race were not significantly important in classifying behavioral health needs between middle-aged and older adult groups. The older adults who had non-actionable ratings on employment and actionable ratings on the legal system (current JS involvement), middle-aged adults were more likely to struggle with anxiety than older adults. The older adults with non-actionable ratings on employment, legal system, and adjustment to trauma, non-disabled older adults were more likely to present behavioral health needs on medical/physical, anxiety, independent living, recreational, and sleep.