Table 2.
Prospective association of the Mediterranean diet score with the risk of hepatic steatosis, CoLaus Study, Switzerland (n=2288)
| Risk ratio (95% CI) across quintiles of Mediterranean diet score* | P trend | Risk ratio (95% CI) per SD increase* | |||||
| Q1 | Q2 | Q3 | Q4 | Q5 | |||
| Range | 1.83–7.63 | 7.64–8.35 | 8.36–8.92 | 8.93–9.59 | 9.60–12.1 | ||
| N total | 458 | 458 | 457 | 458 | 457 | ||
| Fatty liver index, median (IQR)† | 21.8 (10.4 to 37.7) | 20.4 (9.0 to 39.4) | 18.5 (8.9 to 34.6) | 18.1 (7.7 to 33.1) | 14.2 (6.6 to 28.5) | ||
| N cases (score ≥60) | 36 | 43 | 35 | 22 | 17 | ||
| Unadjusted | 1.00 (ref) | 1.19 (0.78 to 1.82) | 0.97 (0.62 to 1.52) | 0.61 (0.37 to 1.02) | 0.47 (0.27 to 0.83) | 0.001 | 0.79 (0.70 to 0.90) |
| Multivariable‡ | 1.00 (ref) | 1.11 (0.72 to 1.72) | 1.07 (0.68 to 1.66) | 0.71 (0.42 to 1.22) | 0.50 (0.28 to 0.91) | 0.006 | 0.84 (0.73 to 0.96) |
| Multivariable+BMI | 1.00 (ref) | 1.12 (0.73 to 1.71) | 0.90 (0.58 to 1.39) | 0.74 (0.44 to 1.24) | 0.61 (0.34 to 1.09) | 0.031 | 0.85 (0.73 to 0.99) |
| Multivariable+BMI+WC | 1.00 (ref) | 1.07 (0.70 to 1.64) | 0.90 (0.59 to 1.37) | 0.74 (0.44 to 1.26) | 0.60 (0.34 to 1.08) | 0.034 | 0.85 (0.71 to 1.02) |
| NAFLD score, median (IQR)§ | −2.1 (−2.4 to -1.5) | −2.0 (−2.4 to −1.4) | −2.0 (−2.4 to -1.3) | −2.1 (−2.5 to -1.5) | −2.1 (−2.5 to -1.6) | ||
| N cases (score ≥−0.640) | 41 | 46 | 51 | 38 | 32 | ||
| Unadjusted | 1.00 (ref) | 1.12 (0.75 to 1.67) | 1.25 (0.84 to 1.84) | 0.93 (0.61 to 1.41) | 0.78 (0.50 to 1.22) | 0.17 | 0.93 (0.82 to 1.05) |
| Multivariable‡ | 1.00 (ref) | 1.13 (0.75 to 1.70) | 1.26 (0.85 to 1.87) | 1.01 (0.65 to 1.56) | 0.80 (0.50 to 1.28) | 0.28 | 0.95 (0.83 to 1.09) |
| Multivariable+BMI | 1.00 (ref) | 1.17 (0.78 to 1.75) | 1.17 (0.80 to 1.71) | 1.07 (0.69 to 1.65) | 0.95 (0.60 to 1.52) | 0.71 | 0.99 (0.86 to 1.15) |
| Multivariable+BMI+WC | 1.00 (ref) | 1.12 (0.75 to 1.67) | 1.17 (0.80 to 1.72) | 1.07 (0.70 to 1.65) | 0.96 (0.60 to 1.53) | 0.80 | 1.00 (0.86 to 1.17) |
Statistical analysis using Poisson regression with robust SEs; results are expressed as risk ratios and 95% CIs.
*In categorical analysis, the population was divided into five groups by quintiles (Q1–Q5) of the Mediterranean diet score; SD was 1.20 for the multivariable analyses.
†Calculated based on an algorithm including BMI, WC, triglycerides and gamma-glutamyl transferase.
‡Adjusted for age (years), sex, marital status (single, married/cohabiting and widowed/separated/divorced), occupational status (working and not working), education level (university, high school, apprenticeship and mandatory education), smoking status (never, former and current), energy intake (kcal/day), total energy expenditure (kcal/day) and date of dietary assessment.
§Calculated based on an algorithm including presence of the metabolic syndrome and type 2 diabetes, and concentrations of fasting serum insulin, fasting serum aspartate aminotransferase (AST) and the AST/alanine aminotransferase ratio.
BMI, body mass index; NAFLD, non-alcoholic fatty liver disease; WC, waist circumference.