ABSTRACT
In order to evaluate the effectiveness of an intensive care unit (ICU), the case-mix has to be considered. This was a cohort study. By using Acute Physiology and Chronic Health Evaluation scores (APACHE II score), we evaluated the case-mix and mortality rate of 282 patients who were treated in our postoperative ICU. The overall mortality rate was 10.6 per cent. Higher Acute physiology scores and emergency surgery in the presence of chronic health status were related to higher mortality, but age was not. However, the original APACHE II model could not precisely predict the mortality of Thai patients. We used stepwise logistic regression to determine the predictors of death and found the prediction model to be -7.24 + 0.37 (APACHE II score) + 1.46 (postemergency surgery). The actual mortality for patients with APACHE II score > 15 in our ICU was higher than that predicted by the original APACHE II model. The causes of this difference might be difference in methodology, characteristics of ICU and the quality of care.