A novel mortality model for acute alcoholic hepatitis including variables recorded after admission to hospital

作者：Mazzocco Thomas^*; Hussain Amir; Hussain Sajid; Shah Amir A

来源：Computers in Biology and Medicine, 2014, 44: 132-135.

DOI：10.1016/j.compbiomed.2013.11.005

• Alcoholic hepatitis
• Alcoholic liver disease
• Clinical decision support system
• Logistic regression
• Mortality model

Severe forms of alcoholic hepatitis in patients with alcoholic liver disease are associated with high mortality; it is therefore vital to identify those patients at greatest risk of mortality in 28 days as they may benefit from aggressive intervention. The aim of this study is to propose a new predictive model that can be used in clinical practice to identify such patients and to monitor their progress while in hospital. %26lt;br%26gt;A cohort of 82 patients was selected and for each of them, a number of clinical findings and standard laboratory tests at the time of admission to hospital were recorded. Also, some variables were collected up to 7 days after admission. %26lt;br%26gt;The proposed logistic regression model selected four statistically significant predictors (namely, the level of creatinine on and after admission, the presence of encephalopathy and prothrombin time evaluated after admission). A comparison with the available mortality predictive scores showed an increase by 25% in predictive power, demonstrating increased accuracy in identifying these sick patients with alcoholic hepatitis in clinical practice.

• 出版日期2014-1-1