Data utilized in this population-based study was from NHANES, a nationwide survey conducted by the National Center for Health Statistics of the Centers for Disease Control and Prevention, employing a randomized, stratified, multi-stage approach (12). Participants underwent a series of evaluations including physical examinations, health and nutrition questionnaires, and laboratory assessments. The NHANES study protocol was approved by the Ethics Review Board of the National Center for Health Statistics. Comprehensive detailed design and data can be accessed at https://www.cdc.gov/nchs/nhanes/. Six NHANES cycles data were obtained and merged: 2007–2008, 2009–2010, 2011–2012, 2013-2014, 2015-2016, and 2017-2018 (59,842 participants). Participants under the age of 20, pregnant, as well as those without data on WWI index, urinary albumin to urinary creatinine ratio (UACR, mg/g), eGFR, and T2DM were excluded. Finally, a total of 5,028 eligible T2DM patients were included in the study.

The WWI index, as an exposure variable, was defined as WC divided by the square root of weight. Diabetes was defined as either self-reported doctor diagnosis of diabetes, or fasting plasma glucose (FPG, mmol/L) ≥7.0 mmol/L, or glycohemoglobin≥6.5%, or the use of diabetes medication. The urinary albumin to urinary creatinine ratio was used to calculate the UACR, while the eGFR was calculated using the CKD-EPI creatinine equation, considering factors such as age, gender, race, and serum creatinine (Scr, μmol/L) (13). DKD in T2DM patients was diagnosed if UACR was >30 mg/g and/or eGFR was <60 mL/min/1.73m².

We collected demographic data including age, gender, and race, along with potential covariates such as annual household income, educational level, physical activity, smoking status, hypertension, cardiovascular disease, BMI, alanine transaminase (ALT, U/L), aspartate transaminase (AST, U/L), gamma-glutamyl transferase (GGT, U/L), FPG, glycohemoglobin, triglycerides (TG, mmol/L), total cholesterol (TC, mmol/L), high-density lipoprotein cholesterol (HDL-c, mmol/L), low-density lipoprotein cholesterol (LDL-c, mmol/L), blood urea nitrogen (BUN, mmol/L), serum uric acid (SUA, umol/L) and Scr. BMI was categorized as <25, 25-29.9, and ≥30 kg/m² representing normal weight, overweight, and obesity. Smokers were identified as current and former. Self-reported hypertension was also defined. The presence of cardiovascular disease was identified based on self-reported history of specific conditions such as heart attack, stroke, congestive heart failure, coronary artery disease, or angina. Detailed measurement procedures for all variables are available in the NHANES database at https://www.cdc.gov/nchs/nhanes/.

Population-weighted descriptive statistics was calculated in this study. Continuous variables were reported as mean with standard error (SE), and categorical variables as percentage with SE. Weighted Student’s t-test and chi-squared test were used to compare variables among groups. Weighted Pearson correlation analysis assessed the association of WWI index with other covariates. Logistic regression models were used to investigate associations between weight, WC, WWI index and DKD, low-eGFR (eGFR<60 mL/min/1.73m²), and albuminuria (UACR>30mg/g) risk, with three models used for different levels of covariate adjustment (Model 1, without covariates adjusted; Model 2, with adjustments for age, gender, and race; Model 3, with adjustments for multiple covariates including age, gender, race, annual household income, education level, physical activity, smokers, hypertension, cardiovascular disease, BMI, ALT, AST, GGT, FPG, glycohemoglobin, TG, TC, HDL-c, LDL-c, BUN, SUA, and Scr). Smooth curve fitting analysis was used based on a generalized additive model for DKD, low-eGFR, and albuminuria, with WWI index as the independent variable. ROC analysis was used to evaluate the impact of WWI index on DKD, low-eGFR, and albuminuria risk. Subgroup analyses were performed based on age (<60/≥60 years), gender (female/male), race (white/non-white), smokers (yes/no), BMI (normal weight/overweight/obese), hypertension (yes/no), and cardiovascular disease (yes/no). All statistical analyses followed Centers for Disease Control and Prevention guidelines, incorporating a complex multistage cluster survey design, and combining subsample weights from six cycles. The statistical analyses were performed by the Empower software (http://www.empowerstats.com) and R 4.2.1 (http://www.R-project.org). A two-side P value <0.05 was considered statistically significant.

The study involved 5,028 T2DM participants meeting the inclusion criteria, with an average age of 58.89 ± 0.29 years and DKD composition of 38.60%. We analyzed and compared general information and clinical indicators between the non-DKD and DKD groups ( Table 1 ). Our results revealed that the DKD group had significantly higher values for age, annual household income under $20,000, hypertension, cardiovascular disease, WC, GGT, glycohemoglobin, FPG, TG, BUN, SUA, Scr, urinary albumin and UACR compared to the non-DKD group (P<0.05). However, the DKD group showed significant decreases in education level above high school, moderate physical activity, LDL, and eGFR (P<0.05). Notably, the DKD group exhibited higher WWI levels than the non-DKD group (11.75 ± 0.03 vs. 11.53 ± 0.02, P<0.001).

The T2DM participants were divided into four groups based on their WWI levels: quartile I, quartile II, quartile III, and quartile IV, as detailed in Table 2 . When compared to the quartile I-WWI group, the quartile II-WWI, quartile III-WWI, and quartile IV-WWI groups showed significant increases in age, annual household income under $20,000, prevalence of hypertension, cardiovascular disease, BMI, Weight, WC, BUN, SUA, urinary albumin, and UACR (P<0.01). Additionally, male, education level above high school, moderate physical activity, FPG, HDL-c, eGFR, and urinary creatinine exhibited significant decreases (P<0.05). There were also significant differences in the distribution of race among the various quartile groups (P<0.001). No differences in smokers, ALT, AST, GGT, Glycohemoglobin, TG, TC, LDL-c, and Scr were observed between groups. Importantly, as the WWI levels increased progressively, both the prevalence of albuminuria, low-eGFR, and DKD also showed a gradual increase (albuminuria: 20.14% vs. 26.00% vs. 26.58% vs. 30.77%, P<0.001; low-eGFR: 10.29% vs. 12.26% vs. 20.52% vs. 22.76%, P<0.001; DKD: 26.76% vs. 32.63% vs. 39.06% vs. 42.96%, P<0.001).

In the weighted Pearson correlation analysis, WWI index showed significant associations with several variables ( Table 3 ). Specifically, WWI index was positively correlated with age, annual household income under $20,000, hypertension, cardiovascular disease, BMI, weight, WC, BUN, SUA, albuminuria, low-eGFR, DKD (P<0.05). Conversely, WWI index was negatively correlated with male, education level above high school, moderate physical activity, TC, HDL-c, LDL-c, eGFR, and urinary creatinine (P<0.05). After adjusting for age, gender, and race, the WWI index was still positively correlated with albuminuria, low-eGFR, DKD (P<0.05).

The association between WWI index and DKD risk was presented in Table 4 . The unadjusted model revealed a positive association between WWI index and an increased risk of DKD (OR=1.44, 95%CI:1.33-1.56, P< 0.001). This association remained statistically significant after adjusting for age, gender, and race (OR=1.36, 95%CI:1.24-1.48, P< 0.001). After additional adjustment for multiple covariates, the results indicated a 32% increased risk of DKD per unit increase in WWI index (OR=1.32, 95%CI:1.12-1.56, P<0.001). Similarly, we also observed a significant association between the WWI index and low-eGFR, and albuminuria after adjusting covariates (low-eGFR: OR=1.24, 95%CI:1.00-1.52, P=0.049; albuminuria: OR=1.37, 95%CI:1.17-1.60, P< 0.001) ( Data Sheet 1 ). When classifying WWI index into quartiles, we found that higher WWI quartiles were associated with a higher prevalence of DKD compared to the lowest quartiles in the fully adjusted model (OR=1.47, 95%CI:1.06-2.04, P=0.021). Additionally, the results of the smooth curve fitting analysis confirmed a positive correlation between WWI index and the risk of DKD, low-eGFR, and albuminuria risk ( Figure 1 ). No significant threshold effect was observed in DKD.

A multivariate logistic regression model was performed using WWI index, age, gender, race, annual household income, education level, moderate physical activity, smokers, hypertension, cardiovascular disease, BMI, ALT, AST, GGT, FPG, glycohemoglobin, TG, TC, HDL-c, LDL-c, BUN, SUA, and Scr as independent variables, and DKD as dependent variables. The results showed that WWI index, age, gender, race, smokers, hypertension, GGT, glycohemoglobin, BUN, SUA, and Scr were more strongly associated with DKD risk (P< 0.05) ( Table 5 ).

Figure 2 illustrates the evaluation of WWI index, Weight, WC, and BMI, in terms of their predictive power for low-eGFR, albuminuria, and DKD risk based on the ROC analysis. In terms of DKD risk, the corresponding AUC values are WWI index (57.19%), Weight (51.23%), WC (52.23%), and BMI (49.96%). The corresponding AUC values in low-eGFR risk are as follows: WWI index (60.01%), Weight (52.89%), WC (51.97%), and BMI (51.40%). As for albuminuria risk, the corresponding AUC values are WWI index (54.79%), Weight (51.26%), WC (51.18%), and BMI (49.35%).

We conducted further investigations into the relationship between WWI index and DKD risk in different subgroups based on age, gender, race, smoking status, BMI, hypertension, and cardiovascular disease ( Figure 3 ). Our findings showed a significant interaction between age, cardiovascular disease, and WWI index in relation to DKD risk (P for interaction<0.05). Specifically, WWI index had a stronger association with the risk of DKD in the elderly (age≥60 years) (OR=1.43, 95%CI:1.16-1.77, P<​0.001 vs. OR=1.01, 95%CI:0.79-1.29, P=0.941) and cardiovascular disease (OR=1.83, 95%CI:1.34-2.49, P<​0.001 vs. OR=1.17, 95%CI:0.98-1.40, P= 0.076) population. No significant interactions were observed in the other subgroups (P for interaction> 0.05).