Scholarlink

Background/Objective <p>With rapid urbanization and population aging in China, diabetes has become a major public health challenge for middle-aged and older adults. Understanding its spatiotemporal patterns and key drivers is essential for targeted prevention. However, geographical insights into the nonlinear relationships, threshold effects, and interactions between multidimensional drivers and diabetes prevalence are still lacking.</p> Methods <p>Using national longitudinal data from the China Health and Retirement Longitudinal Study (CHARLS, 2011–2020), we developed a framework to capture the spatiotemporal dynamics of diabetes prevalence and to analyze its linear, nonlinear, threshold, and interaction effects with multidimensional drivers, including lifestyle, social, and ecological environmental factors. Spatial autocorrelation analysis identified the clustering patterns, while a Random Forest–SHAP approach quantified and interpreted the contributions of these drivers.</p> Results <p>Prevalence exhibited a “rise-then-decline” pattern, peaking in 2018. Significant spatial clustering was observed, with northeast and central-eastern China as hotspots and the southwest as a coldspot. Body mass index (BMI) was the most influential driver, and its impact on diabetes prevalence increased sharply once it exceeded 24&#xa0;kg/m². Among all multidimensional drivers, BMI also displayed the strongest synergistic effects with other factors, while ecosystem quality was inversely related to prevalence, providing quantifiable protective effects, most pronounced within the 0.38–0.62 index range.</p> Conclusions <p>This integrated geographical framework effectively captures spatiotemporal trends and complex driver relationships, constituting a transferable model for other chronic disease studies. The findings highlight pronounced spatial disparities and identify actionable risk thresholds for BMI and ecosystem quality, providing a robust evidence base for precision prevention, targeted regional interventions, and advancing the “Healthy China” initiative and active ageing.</p>

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

Multidimensional drivers of diabetes prevalence among middle-aged and elderly in China: from a geographical perspective

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月24日

查看全文
  • Yiming Song,
  • Hongzhou Wang,
  • Xin Li,
  • Hanwen Pan,
  • Zhuyan Huang,
  • Lin Chen,
  • Gengchen Cao,
  • Chao Zhao

摘要

Background/Objective

With rapid urbanization and population aging in China, diabetes has become a major public health challenge for middle-aged and older adults. Understanding its spatiotemporal patterns and key drivers is essential for targeted prevention. However, geographical insights into the nonlinear relationships, threshold effects, and interactions between multidimensional drivers and diabetes prevalence are still lacking.

Methods

Using national longitudinal data from the China Health and Retirement Longitudinal Study (CHARLS, 2011–2020), we developed a framework to capture the spatiotemporal dynamics of diabetes prevalence and to analyze its linear, nonlinear, threshold, and interaction effects with multidimensional drivers, including lifestyle, social, and ecological environmental factors. Spatial autocorrelation analysis identified the clustering patterns, while a Random Forest–SHAP approach quantified and interpreted the contributions of these drivers.

Results

Prevalence exhibited a “rise-then-decline” pattern, peaking in 2018. Significant spatial clustering was observed, with northeast and central-eastern China as hotspots and the southwest as a coldspot. Body mass index (BMI) was the most influential driver, and its impact on diabetes prevalence increased sharply once it exceeded 24 kg/m². Among all multidimensional drivers, BMI also displayed the strongest synergistic effects with other factors, while ecosystem quality was inversely related to prevalence, providing quantifiable protective effects, most pronounced within the 0.38–0.62 index range.

Conclusions

This integrated geographical framework effectively captures spatiotemporal trends and complex driver relationships, constituting a transferable model for other chronic disease studies. The findings highlight pronounced spatial disparities and identify actionable risk thresholds for BMI and ecosystem quality, providing a robust evidence base for precision prevention, targeted regional interventions, and advancing the “Healthy China” initiative and active ageing.