*3.2. Comparison of DII Score by Baseline Glycemic Status*

Compared with the normoglycemia group (0.883, 95% CI: 0.793, 0.973), participants with prediabetes (1.081, 95% CI: 0.981, 1.181) and type 2 diabetes (1.249, 95% CI: 1.151, 1.346) had higher DII scores (overall weighted *p* < 0.001). Figure 2 presents the distribution of DII scores among three groups. The proportion of high DII scores was higher in participants with prediabetes or type 2 diabetes. Moreover, we compared the component of DII scores among the three groups to find the main cause of the difference.


**Table 1.** Basic characteristics of participants by baseline glycemic

 status.

*Nutrients* **2022** , *14*, 2556

lipoprotein cholesterol; and SBP: systolic blood pressure.

Participants with type 2 diabetes had higher scores in alcohol, fiber, MUFA, PUFA, niacin, thiamin, riboflavin, vitamin B6, vitamin C, vitamin E, Mg, Zinc, Selenium, folic acid, N-3 fatty acids, and N-6 fatty acids (Figure 3, Table 2). We also noticed lower scores of participants with type 2 diabetes in certain components, such as carbohydrates, protein, total fat, saturated fat, vitamin B12, Fe, and energy. When compared to the normoglycemia group, the DII component scores remained consistent between participants with prediabetes and type 2 diabetes but to a lesser extent in the former.

**Figure 3.** Comparison of the DII component scores by baseline glycemic status. Data are presented as the weighted mean value and 95%CI. DII: dietary inflammatory index; MUFA: monounsaturated fatty acids; and PUFA: polyunsaturated fatty acids.



#### *3.3. Association between Dietary Inflammation and Long-Term Mortality*

The overall weighted 5-year all-cause mortality was 4.56%, and the weighted 5-year cardiovascular mortality was 1.17%. The Cox regression models revealed that higher DII scores were associated with higher long-term all-cause mortality (HR per 1 score increase: 1.105, 95% CI: 1.065, 1.147; *p* < 0.001) and cardiovascular mortality (HR per 1 score increase: 1.172, 95% CI: 1.092, 1.258; *p* < 0.001) (Table S1 in the Supplementary Materials).

The association was stable after adjusting for age, sex, educational level, BMI, smoke, hypertension, hyperlipidemia, glycemic status, recreational activity, and alcohol consumption. Compared with participants with low DII scores, participants with mediate or high DII scores had higher risk of all-cause death (Mediate DII: adjusted HR: 1.181, 95% CI: 1.009, 1.381; *p* = 0.038; high DII: adjusted HR: 1.240, 95% CI: 1.053, 1.459; *p* = 0.010) and cardiovascular death (adjusted HR: 1.442, 95% CI: 1.051, 1.979; *p* = 0.023; high DII: adjusted HR: 1.423, 95% CI: 1.006, 2.013; *p* = 0.046) (Figure 4, Table S1 in the Supplementary Materials).

**Figure 4.** Association between DII scores and long-term (**a**) all-cause mortality and (**b**) cardiovascular mortality. HR was adjusted for age, sex, educational level, BMI, smoke, hypertension, hyperlipidemia, glycemic status, recreational activity, and alcohol consumption. CI: confidence interval; DII: dietary inflammatory index; HR: hazard ratio; and Ref: reference.
