Keshan Disease and Serum Selenium in China: A National Ecological Study Using Bivariate Spatial Autocorrelation Analysis.

Keshan disease (KD) is an endemic cardiomyopathy associated with selenium deficiency and distributed mainly along China's low-selenium belt. Sporadic cases continue to occur in some historically endemic areas despite elimination efforts. The national-scale ecological spatial association between KD prevalence and serum selenium levels remains unclear. This study examined their spatial association in China to support targeted prevention and long-term surveillance. Using 2020 national surveillance data, we applied ordinary Kriging interpolation with an exponential model to map selenium distribution. Global and local bivariate spatial autocorrelation analyses were conducted in GeoDa to assess spatial associations between serum selenium and KD prevalence. Kriging identified a northeast-to-southwest belt of selenium deficiency overlapping KD-endemic areas. Global bivariate autocorrelation showed weak but statistically significant negative spatial associations between serum selenium and overall KD prevalence (Moran's I = -0.047, P = 0.013) and latent Keshan disease (LKD) (Moran's I = -0.063, P = 0.002), but not chronic Keshan disease (CKD). Given the small Moran's I values, these results should be interpreted cautiously, as national-scale analysis may obscure local heterogeneity. Local analysis identified 67, 67, and 69 low-high clusters for KD, CKD, and LKD, respectively (203 in total across the three analyses). Serum selenium levels were weakly but significantly associated with KD prevalence spatial distribution in China. These findings indicate statistically detectable ecological spatial dependence, while no inference can be made regarding individual-level causality. The identified high-low clusters may be considered priority areas for targeted KD prevention and selenium monitoring.