Scholarlink

Objective. To develop a predictive model for promoting rational medication use among the elderly in Changchun City, providing insights for enhancing their medication management. Methods. A survey questionnaire was conducted with 1,300 older adults living in Changchun City. The survey used a whole-population approach, and variables were identified and screened using a LASSO algorithm. This led to the creation of two prediction models: a decision-tree-based model and a logistic regression-based model. Results. The logistic regression algorithm was the most effective predictive model for this study when evaluated regarding classification ability, fitting effectiveness, and calibration ability.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A LASSO Algorithm-Based Predictive Model for Appropriate Medication Use: Development and Comparison

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Yufang He,
  • Kan Lyu,
  • Shan Huo,
  • Jie Ren,
  • Qian Zhao,
  • Xinyu Wang,
  • Zheng Xie

摘要

Objective. To develop a predictive model for promoting rational medication use among the elderly in Changchun City, providing insights for enhancing their medication management. Methods. A survey questionnaire was conducted with 1,300 older adults living in Changchun City. The survey used a whole-population approach, and variables were identified and screened using a LASSO algorithm. This led to the creation of two prediction models: a decision-tree-based model and a logistic regression-based model. Results. The logistic regression algorithm was the most effective predictive model for this study when evaluated regarding classification ability, fitting effectiveness, and calibration ability.