Fig. 2

The association between SAF and max IMT by sex, and by the presence of diabetes or chronic kidney disease. Plot A: forest plot of coefficients and confidence intervals for the association between SAF and max IMT in the total population and subgroups in four models. M0 was the crude association not adjusted for covariates. M1 was adjusted for age, sex, and RS subcohorts. Model 2 was additionally adjusted for body mass index, dyslipidemia, and hypertension. Model 3 was additionally adjusted for smoking status, diabetes, and eGFR on top of model 2. Sex and diabetes were not adjusted in the models when stratified by sex or diabetes status. Plots B–D: effect plots of SAF by sex and the presence of diabetes or CKD. Fitted values of max IMT along SAF values in subgroups (R package “effects”). The solid line represents the predicted values of max IMT and the shade represents 95% confidence intervals with the other predictors held at the mean, deriving from the following models: B max IMT ~ SAF + age + sex + RS subcohorts + BMI + dyslipidemia + hypertension + smoking status + diabetes + eGFR + SAF*sex. C max IMT ~ SAF + age + sex + RS subcohorts + BMI + dyslipidemia + hypertension + smoking status + diabetes + eGFR + SAF*diabetes. D max IMT ~ SAF + age + sex + RS subcohorts + BMI + dyslipidemia + hypertension + smoking status + diabetes + eGFR + CKD + SAF*CKD