Prognostic scoring for critically ill hospitalized patients.

The objective of the study was to validate and refine the APACHE II (acute physiology and chronic health evaluation II) prognostic system in the Indian context. We prospectively collected data on 79 patients admitted in the medical intensive care unit. We have studied APACHE II and 11 other physiological variables and sought to improve the risk prediction by developing a new score to be governed by the rule of thumb at the bed side. The new score included the following five variables: pH and serum albumin at admission and heart rate, bilirubin and Glasgow coma scale at 48 hours. A score below 3.5 was independently associated with a statistically significant increase in the risk of hospital death. This model resulted in a pseudo r2 of 0.43 in comparison to pseudo r2 of 0.02 and 0.12 for APACHE II scores on the day of admission and after 48 hours, respectively.