Skip to main content
. 2020 Apr 22;35(11):1069–1085. doi: 10.1007/s10654-020-00627-2

Table 4.

Logistic mixed model for risk of NAFLD with adherence to a posteriori dietary patterns

Risk of NAFLD
Model 1 Model 2
Mean (95% CI) Mean (95% CI)
Follow-up time (years) 0.74 [0.66, 0.84] 0.70 [0.60, 0.81]
Age (years)* 0.93 [0.82, 1.04] 0.88 [0.77, 1.00]
Sex (female) 0.17 [0.04, 0.57] 0.21 [0.05, 0.78]
Energy intake (kcals) 1.00 [1.00, 1.00] 1.00 [1.00, 1.00]
Education level*
Low Ref Ref
Intermediate 0.24 [0.06, 0.84] 0.35 [0.09, 1.33]
High 0.06 [0.01, 0.30] 0.12 [0.02, 0.58]
Physical activity* (metEqhs) 0.99 [0.97, 1.00] 1.00 [0.99, 1.01]
Alcohol intake (units) 0.85 [0.52, 1.40] 1.06 [0.61, 1.85]
BMI (kg/m2) 3.01 [2.28, 4.19]
Diabetes mellitus* 41.5 [8.49, 273]
Hypertension* 7.93 [1.60, 50.8]
A posteriori dietary patterns
Vegetable and fish pattern 1.88 [0.90, 4.05] 1.31 [0.57, 3.01]
Red meat and alcohol pattern 1.40 [0.70, 2.80] 0.79 [0.36, 1.69]
Traditional pattern 0.33 [0.14, 0.77] 0.40 [0.15, 1.00]
Salty snacks and sauces pattern 0.63 [0.28, 1.35] 0.43 [0.17, 1.04]
High fat dairy and refined grains pattern 0.82 [0.55, 1.22] 1.23 [0.79, 1.96]

This table shows results (OR and 95% CIs) of two multivariable logistic mixed models

BMI body mass index, CI credible interval, NAFLD non-alcoholic fatty liver disease, kcal kilocalories, metEqh metabolic equivalent hours

*Depicts baseline variables. Numbers in bold reflect a tail-probability of < 0.05