Machine Learning Approaches to Identify Social Determinants of Frailty Among Middle-Aged and Older Adults in the USA, England, and China (76897)

Session Information: Interdisciplinary, Multidisciplinary Research
Session Chair: James W. McNally

Wednesday, 27 March 2024 11:35
Session: Session 2
Room: Room 603
Presentation Type: Oral Presentation

All presentation times are UTC + 9 (Asia/Tokyo)

Frailty has become a global health concern and is associated with social determinants of health (SDoH). However, the relative importance and cumulative contribution of multidomain SDoH to frailty, and whether these relationships differ across different national settings, remain unclear. We included participants aged ≥45 years from the Health and Retirement Study (HRS, N=5,792), the English Longitudinal Study of Ageing (ELSA, N=3,773), and the China Health and Retirement Longitudinal Study (CHARLS, N=5,016). SDoH (n=121 for HRS, n=125 for ELSA, and n=94 for CHARLS) were selected across seven domains, including adverse childhood experiences, socioeconomic status, material circumstances, social connections, social stressors, health behaviors, and healthcare systems. Frailty was assessed by the frailty index (FI). We developed Extreme Gradient Boosting to predict frailty at the 4-year follow-up and used SHapley Additive exPlanations to quantify the variable-wise and domain-wise contributions of SDoH. Our models explained 0.242 (95% confidence interval [CI]: 0.203–0.281), 0.258 (95% CI: 0.191–0.325), and 0.173 (95% CI: 0.126–0.215) of the variance in FI among all participants from HRS, ELSA, and CHARLS. Health behaviors and social connections or stressors were the most important domains in HRS and ELSA, while material circumstances contributed largely in CHARLS. Several important SDoH predictors, such as body mass index, were consistent across countries, while country-specific risk factors, such as engagement in maintenance or gardening in HRS, were also identified. Our findings reveal the prioritization of SDoH domains and factors for addressing aging disparities and promoting healthy aging, especially region-specific risk factors for tailored public health prevention strategies.

Yan Luo, City University of Hong Kong, Hong Kong
Mengzhuo Guo, Sichuan University, China
Qingpeng Zhang, The University of Hong Kong, Hong Kong

About the Presenter(s)
Yan Luo is a PhD candidate at the School of Data Science, City University of Hong Kong. His research focuses on using machine learning approaches to identify social determinants and biological markers of diseases and aging.

Connect on Linkedin

Connect on ResearchGate

Additional website of interest

See this presentation on the full scheduleWednesday Schedule

Conference Comments & Feedback

Place a comment using your LinkedIn profile


Share on activity feed

Powered by WP LinkPress

Share this Presentation

Posted by Clive Staples Lewis

Last updated: 2023-02-23 23:45:00