The logistic regression method can be used to predict mortality in alcoholic liver disease by looking at significant predictors for mortality.
Parameters:
(1) hemoglobin as percent of normal
(2) serum albumin in g/dL
(3) encephalopathy score
logistic regression score =
= ((-0.025) * (hemoglobin as percent normal)) - (1.56 * (albumin in g/dL)) + (1.03 * (encephalopathy score)) + 6.0
where:
• Hemoglobin is expressed as a whole number.
• The "normal" hemoglobin used for calculation is lower limit of reference range; an alternative would be the mean of the reference range.
• The encephalopathy score is 1, 2 or 3.
Interpretation
Score |
% mortality |
-5.4 to -4.0 |
4% |
-3.9 to -2.6 |
5% |
-2.5 to -1.2 |
11% |
-1.1 to 0.2 |
33% |
0.3 - 1.6 |
80% |
1.7 - 3.0 |
100% |
where:
• The percent mortality is estimated from Figure 6, page 902.
Purpose: To evaluate a patient with severe alcoholic hepatitis using the logistic regression function of Orrego et al.
Specialty: Gastroenterology, Toxicology
Objective: other testing, severity, prognosis, stage
ICD-10: K70.9,