Monitoring treatment of alcoholic liver disease: evaluation of various severity indices.

Monitoring the effectiveness of treatment sfor alcoholic liver disease involves the use of variables that have prognostic significance and are unaffected by the treatment in unspecific ways. Here we review the value of histological and functional variables for this purpose. We conclude that histological variables, although important in defining the characteristics of the sample, have several practical problems. The functional variables are most effective when used in combinations, e.g., in global indices such as the Combined Clinical and Laboratory Index, the Child-Turcotte-Pugh Index, or the Cox model. In situations involving mortality and dropouts, functional indices cannot be used to measure changes in severity; in such cases, mortality might be the only measure for assessing the effectiveness of a treatment. In clinical trials, it is essential to determine the risk of a Type II error, to monitor compliance and drinking, and to trace appropriately all the patients who were not compliant or who dropped out of the trial.