The association between serum cadmium and diabetes in the general population: A cross-sectional study from NHANES (1999-2020).

Background: Associations between serum cadmium and diabetes had been reported in previous studies, however there was still considerable controversy regarding associations. Studies in general population that investigated the effects of serum cadmium on diabetes were currently lacking. We designed this cross-sectional study among U.S. adults under high and low cadmium exposure to assess associations between serum cadmium and diabetes. Methods: This cross-sectional study analyzed 52,593 adults who aged more than 20 years and participated in the National Health and Nutrition Examination Survey (NHANES), 1999-2020. The missing values and extreme values in the covariables were filled by multiple interpolation. Univariate logistics regression, multivariate logistics regression and smooth fitting curves were used to analyze the association between serum cadmium and diabetes. Simultaneously, sensitivity analysis was carried out by converting the serum cadmium from continuous variable to categorical variable. The stratification logistics regression model was used to analyze whether there were special groups in each subgroup to test the stability of the results. Results: In this cross-sectional study, serum cadmium levels were negatively correlated with the occurrence of diabetes in the low serum cadmium exposure group (OR = 0.811, 95% CI 0.698, 0.943; Conclusion: Serum cadmium was negatively associated diabetes among the representative samples of the whole population in the United States under the normal level of serum cadmium exposure. However, there was no association between serum cadmium level and the occurrence of diabetes in the high serum cadmium exposure group. This study promoted an update of new preventative strategy targeting environment for the prevention and control of diabetes in the future.