Logistic regression results for categorical and continuous variables for the four models: (A) logistic regression results of micronutrient intake (continuous variable); (B) logistic regression results of intake inadequate (categorical variable). RDA: recommended dietary allowance; AI: adequate intake. Model 0 had no adjustment of data; Model 1 was a multifactor logistic regression adjusted for all eleven covariates; Model 2 used propensity score matching to be balanced covariates (including smoke, alcohol, and physical activity) and then used multifactorial logistic regression adjusting for age, gender, educational level, income, BMI, diabetes, HPL, and HTN; Model 3 used PSM to balance all eleven covariates.