The Protective Role of Residential Greenness on Diabetes Risk and Insulin Sensitivity: Results from a Nationwide Cohort Studies

Abstract Objective: This study aimed to systematically evaluate the protective effects of residential greenness on diabetes risk and insulin sensitivity in a broad population. Methods: Normalized Difference Vegetation Index (NDVI) was used to assess the level of residential greenness, while estimated Glucose Disposal Rate (eGDR) was employed to assess insulin sensitivity. Based on the nationwide cohort data from the China Health and Retirement Longitudinal Study (CHARLS), multistage statistical analysis methods were applied. First, multivariable logistic regression and restricted cubic splines (RCS) were used to evaluate the association between NDVI and baseline diabetes risk. Generalized additive models (GAM) were utilized to reveal the non-linear relationship and threshold effects between NDVI and baseline eGDR in the general population. A linear mixed-effects model was implemented to analyze the dynamic impact of NDVI on longitudinal changes in eGDR. Results: The findings indicated a significant negative association between residential greenness and diabetes risk. For every 1-unit increase in NDVI, diabetes risk decreased by 64% (OR=0.36, 95% CI: 0.20-0.66, p<0.001). Non-linear analysis revealed a clear threshold for the protective effect. When NDVI = 0.318, diabetes risk significantly decreased. When NDVI = 0.348, eGDR levels showed a significant positive increase. Longitudinal tracking further confirmed that increases in NDVI significantly promoted the annual improvement rate of eGDR, with more pronounced improvements in highly greened areas. Conclusion: This study is the first to quantify the dual protective effects of residential greenness on glucose metabolism health in a national cohort, revealing non-linear threshold characteristics. These findings provide important scientific evidence for optimizing diabetes prevention and control strategies through urban greening.