Table 2.
Adjusted linear mixed model of liver stiffness (measured in kPa, n = 676 )
| Fixed effects | Effect estimate baseline | Time trend (per year) |
|---|---|---|
| Estimate (95% CI) | Estimate (95% CI) | |
| Yearly LSM change | - | 2.81 (0.56; 5.0)* |
| Age per 10 years increase | 1.0 (0.68; 1.3)* | -0.18 (-0.45; 0.088) |
| Regular alcohol usea | 1.3 (0.46; 2.0)* | -0.046 (-0.67; 0.58) |
| High substance useb | 0.66 (-0.96; 2.3) | -0.48 (-1.9; 0.94) |
| Body mass index (kg/m2) | 0.25 (0.17; 0.32)* | -0.079 (-0.14; -0.013)* |
| Hepatitis C RNA positivec | 1.2 (0.54; 1.9)* | - |
| Low HDL Cholesterold | 1.4 (0.64; 2.2)* | -0.55 (-1.2; 0.08) |
| Elevated HbA1ce | 3.1 (0.68; 5.5)* | 4.6 (2.3; 6.9)* |
| Hepatitis C status changef | - | -0.73 (-1.3; -0.21)* |
The constant term (β0) was -5.4 (CI -8.0; -2.8). Except for HCV status, the time trend indicates the effect of the variable remaining at baseline levels over time. 274 participants had two or more liver stiffness measurements. Significantly associated (p < 0.05) variables are labelled with bold text and an asterisk
aUsing alcohol on one or more days per week during the past 12 months
bThe iSUSI is a continuous variable ranging from 0–1, were (1) indicates the effect of maximum substance use and (0) indicates the effect of no substance use
cHepatitis C virus RNA positive at baseline
dBelow 1.3 mmol/L for women and 1.0 mmol/L for men
eAbove 48 mmol/mol, indicating type 2 diabetes
fResolved hepatitis C infection compared with no change in HCV infection status from baseline to the following liver stiffness measurements