TABLE 5.
Regression model β coefficient (95% CI) n = 166 |
P-value | |
Demographic factors | ||
Age (Years) | −0.25 (−0.80 to 0.30) | 0.377 |
Sex | 0.01 (−0.25 to 0.27) | 0.929 |
BMI | −0.02 (−0.07 to 0.02) | 0.341 |
Lifestyle factors | ||
Physical activity | −0.09 (−0.16 to −0.01) | 0.030 |
DFS (as score) | 0.69 (0.32 to 1.05) | <0.001 |
Low Group¤ (score < 54) | Reference | |
Mid Group¤ (score 54–65) | 0.51 (0.22–0.80) | 0.001 |
High Group¤ (score ≥ 66) | 0.56 (0.26–0.87) | <0.001 |
Immune marker | ||
IL-6, fg/mL per mmol/L Cr | −0.03 (−0.06 to 0.00) | 0.077 |
KP marker | ||
KA, nmol/L per mmol/L Cr | −0.21 (−0.37 to−0.05) | 0.009 |
The selected model was analysed using Mixed-effect multilevel regression to control for random effects from different studies. BMI, Body Mass Index; DFS, dietary fat and free sugar screener score; IL-6, interleukin-6; KA, kynurenic acid; Cr, creatinine. All variables except BMI were log2-transformed prior to analysis. ¤ denotes variable was chosen to be analysed primarily as a categorical variable rather than continuous variable. Only variables with p-values < 0.05 were considered important in predicting the severity of the depressive symptoms. Significant p-value (< 0.05) are denoted in bold.