Table 4.
Mediating the role of the LLDS on the associations between SES and vitamin deficiencies and insufficiencies a.
| Predictor: Low SES b Mediator: LLDS | a Path | b Path | Mediation Effect | Total Effect | Prop. Mediated c | ||||
|---|---|---|---|---|---|---|---|---|---|
| a (SE) | p | b (SE) | p | m (SE) | p | t (SE) | p | ||
| Single-vitamin deficiency | −2.26 (0.37) | <0.001 | −0.007 (0.007) | 0.3 | 0.017 (0.016) | 0.3 | 0.22 (0.09) | 0.01 | - |
| Multi-vitamin deficiency | −2.26 (0.40) | <0.001 | −0.019 (0.011) | 0.09 | 0.043 (0.027) | 0.1 | 0.16 (0.13) | 0.3 | - |
| Folic acid deficiency | −2.35 (0.35) | <0.001 | −0.022 (0.007) | 0.002 | 0.051 (0.018) | 0.004 | 0.51 (0.09) | <0.001 | 10.1% |
| Folic acid insufficiency | −2.35 (0.36) | <0.001 | −0.018 (0.007) | 0.01 | 0.043 (0.020) | 0.03 | 0.11 (0.091) | 0.2 | 39.1% |
| Vit B12 deficiency | −2.35 (0.36) | <0.001 | −0.006 (0.013) | 0.6 | 0.015 (0.030) | 0.6 | 0.12 (0.14) | 0.4 | - |
| Vit B12 insufficiency | −2.34 (0.35) | <0.001 | −0.018 (0.008) | 0.02 | 0.041 (0.019) | 0.03 | 0.11 (0.09) | 0.2 | 37.2% |
| Vit B6 deficiency | −2.39 (0.35) | <0.001 | −0.019 (0.013) | 0.2 | 0.045 (0.033) | 0.2 | 0.22 (0.18) | 0.2 | - |
| Vit B6 insufficiency | −2.39 (0.35) | <0.001 | −0.022 (0.008) | 0.006 | 0.052 (0.021) | 0.01 | 0.14 (0.10) | 0.1 | 37.1% |
| Vit D deficiency | −2.31 (0.36) | <0.001 | 0.009 (0.012) | 0.5 | −0.021 (0.030) | 0.5 | −0.32 (0.14) | 0.03 | - |
| Vit D insufficiency | −2.31 (0.36) | <0.001 | −0.011 (0.007) | 0.1 | 0.025 (0.017) | 0.1 | −0.07 (0.09) | 0.5 | - |
| Vit E insufficiency | −2.39 (0.36) | <0.001 | 0.004 (0.007) | 0.5 | −0.010 (0.016) | 0.5 | 0.15 (0.08) | 0.07 | - |
| Vit K deficiency | −2.36 (0.33) | <0.001 | −0.000 (0.009) | 0.9 | 0.000 (0.022) | 0.9 | −0.14 (0.12) | 0.2 | - |
Abbreviations: LLDS: Lifelines diet score, SE: standard error, SES: socioeconomic status, Prop.: mediated proportion mediated. a all analyses were adjusted for age, gender, BMI and smoking behavior, a, b, m, and t represented the estimates of different paths in the mediation model, standard error of the estimates and p-value were also shown. p < 0.05 was considered significant. vitamin A was not included because there were too few (N = 7) individuals who had an insufficient vitamin A status. The cutoff values applied for folic acid status were serum folic acid < 10.2 nmol/L, 10.2–13.6 nmol/L, and > 13.6 nmol/L for deficiency, insufficiency, and sufficient status, respectively. Vitamin K insufficiency was defined as desphospho-uncarboxylated matrix Gla protein (dp-ucMGP) level higher than 500 pmol/L. The cutoff values employed for vitamin B12 status were serum vitamin B12 < 150 pmol/L, 150–220 pmol/L, and > 220pmol/L for deficiency, insufficiency, and sufficient status, respectively. The cutoff values used for serum vitamin B6 status were serum vitamin B6 < 20 nmol/L, 20-30 nmol/L, and > 30 nmol/L for deficiency, insufficiency, and sufficient status, respectively. Serum alpha-tocopherol levels below 30 μmol/L were considered indicating vitamin E insufficiency. Vitamin A insufficiency was defined as serum retinol concentration below 0.7 μmol/L. The cutoff values applied for vitamin D status were serum 25(OH)D < 30 nmol/L, 30–50 nmol/L, and > 50nmol/L for deficiency, insufficiency, and sufficient status, respectively. b High SES was treated as a reference group. c proportion mediated was not calculated if the mediation effect was not significant.