Prediction of Liver Steatosis and Fibrosis Based on Clinical Variables Using a Large National Survey Database

Table 2

Result of the multivariable linear regression model including the following covariates: LDL, triglyceride level, age, sex, hemoglobin A1c, total metabolic energy expenditure per week, body mass index, alcohol consumption, average number of cigarettes smoked per day for the last month, and systolic and diastolic blood pressure.

	Estimate	Std. error	t value	Pr (>|t|)
(Intercept)	125.65	29	4.36	0.012
LDL	−1.32	1.357	−0.972	0.39
Triglyceride	2.477	0.545	4.539	0.023
Age	4.455	1.463	3.046	0.034
Sex	−15.361	4.857	−3.162	0.0387
Hemoglobin A1c	7.514	2.733	2.75	0.0514
Metabolic energy expenditure	−13.459	5.36	−2.5	0.066
Body mass index	4.7306	0.3511	13.473	0.000176
Alcohol consumption	−0.923	0.798	−1.157	0.312
Smoking	0.4837	0.3478	1.391	0.24
Systolic blood pressure	2.824	1.852	1.52	0.20
Diastolic blood pressure	1.1	1.785	0.615	0.572

Values are divided by 20. Therefore, the estimated value reflects each 20-unit increase in the predictor value. Values are divided by 10. Therefore, the estimate reflects each 10-unit increase in the predictor value. value is the logarithm scale with base 10; therefore, the estimate reflects each 10-fold increase in the predictor value. Alcohol consumption: 1, drinks alcohol daily; 2, drinks alcohol almost daily; 3, drinks alcohol 3-4 times a week; 4: drinks alcohol 2 times a week; 5, drinks alcohol once a week; 6, drinks alcohol 2-3 times a month; 7, drinks alcohol once a month; 8, drinks alcohol 7–11 times a year; 9, drinks alcohol 3–6 times a year; 10, drinks alcohol 1-2 times a year; and 11 does not drink alcohol.