MACHINE LEARNING APPROACHES FOR FALL PREDICTION IN KOREAN COMMUNITY-DWELLING OLDER ADULTS
2024

Machine Learning for Predicting Falls in Older Adults

Sample size: 9884 publication Evidence: high

Author Information

Author(s): Cho Jungwon, Yang Minhee, Cho Eunhee

Primary Institution: Yonsei University

Hypothesis

Can machine learning models effectively predict falls among community-dwelling older adults?

Conclusion

Machine learning methods can accurately identify older adults at high risk for falls, enabling early interventions.

Supporting Evidence

  • Falls are the second leading cause of death worldwide.
  • Approximately 27% of older adults fall annually.
  • The random forest model had the highest AUC value of 0.98.

Takeaway

This study used computers to help figure out which older people might fall, so they can get help before it happens.

Methodology

The study used various machine learning models on data from the 2020 Korean National Survey of Older Adults, including logistic regression and random forest.

Participant Demographics

Community-dwelling older adults in Korea.

Digital Object Identifier (DOI)

10.1093/geroni/igae098.3368

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication