Determinants and prediction of home nursing utilization among older adults in China: an integration of logistic regression and XGBoost algorithm

Objective: To analyze the factors and their interactions influencing home nursing utilization among older adults in China, and to provide a scientific basis for enhancing home nursing services and improving life quality for older adults. Methods: We utilized nationally representative data from the 2020 wave of the China Longitudinal Aging Social Survey (CLASS), including a total of 2,723 participants. Logistic regression and XGBoost algorithm were employed to examine the determinants of home nursing utilization. Interactions among determinants were also analyzed to uncover deeper insights into home nursing utilization, with XGBoost further applied to generate individual-level predictions. Results: The analysis showed that region type, home visit access, older adults’ subsidy, and family support count were significant enabling resources in home nursing utilization. Activities of Daily Living (ADL) score, chronic disease count, recent hospital duration, and life satisfaction, were strong internal motivators. Interactions, including the influence of region type and chronic disease count, life satisfaction and home visit access, self-rated health and region type, home visit access and family support count, further highlighted the complexity of home nursing utilization. The potential role of online access was also recognized as an emerging factor. Conclusions: This study emphasizes the critical role of both need-based factors and enabling resources in driving home nursing utilization. Identifying high-need populations is essential for improving home nursing utilization and optimizing resource allocation. Targeted interventions are required to address disparities in the distribution of enabling resources, ensuring effective delivery to underserved areas. Community-based systems should prioritize support for older adults with minimal family support while integrating psychological interventions into care models to better align with individual needs. By developing tailored service strategies, this research provides a scientific foundation for optimizing the design and delivery of home nursing services, ultimately improving the quality of care for older adults. Clinical trial number: Not applicable.

Text 124805.pdf - Published Version Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (2MB) Official URL or Download Paper: https://link.springer.com/10.1186/s12912-026-04371...

Actions (login required)

View Item