Scoring Systems and Outcome
R. MORENO,P.METNITZ
Introduction
The evaluation of severity of illness in the critically ill patient is made
through the use of severity scores and prognostic models. Severity scores are
instruments that aim at stratifying patients based on the severity of illness,
assigning to each patient an increasing score as their severity of illness
increases. Prognostic models, apart from their ability to stratify patients
according to their severity, predict a certain outcome (usually the vital status
at hospital discharge) based on a given set of prognostic variables and a cer-
tain modeling equation.
The development of these kinds of systems, applicable to heterogeneous
groups of critically ill patients, started in the 1980s.
The first general severity of illness score applicable to most critically ill
patients was the Acute Physiology and Chronic Health Evaluation (APACHE)
[1]. Developed at the George Washington University Medical Centre in 1981
by William Knaus et al., the APACHE system demonstrated the ability to eval-
uate, in an accurate and reproducible form, the severity of disease in this pop-
ulation [2–4].
Two years later, Jean-Roger Le Gall and co-workers published a simplified
version of this model, the Simplified Acute Physiology Score (SAPS) [5]. This
model soon became very popular in Europe, especially in France.
Another simplification of the original APACHE system, the APACHE II,
was published in 1985 by the same authors of the original model [6]. This sys-
tem introduced the possibility to predict mortality, needing for this purpose
the selection of a major reason for intensive care unit (ICU) admission from
a list comprising 50 operative and non-operative diagnoses. Additional con-
tributions for the prediction of prognosis comprise the Mortality Probability
Models (MPM) [7], developed by Stanley Lemeshow using logistic regression
techniques.
The last developments in this field include the third version of the APACHE

system (APACHE III) [8] and the second versions of the SAPS (SAPS II) [9]
and MPM (MPM II) [10]. All of them used multiple logistic regression to select
and weigh the variables, and are able to compute the probability of hospital
mortality for groups of critically ill patients. It has been demonstrated that
they perform better than their old counterparts [11, 12], and they represent
nowadays the state-of-the-art in this field. However, a new generation of gen-
eral outcome prediction models is now being developed, such as the MPM III
developed in the IMPACT database in the United States of America (USA) [13].
In addition there are new models based on computerised analysis by hierar-
chical regression developed by some of the authors of the APACHE systems
[14] or the new version of the SAPS model, developed by hierarchical regres-
sion in a worldwide database (Rui Moreno, personnel communication,
www.saps3.org for more details). Models based on other statistical techniques
such as artificial neural networks and genetic algorithms have been proposed
but besides academics use they have never been widely used [15, 16].
Given the general character of this chapter, we will not present or discuss
instruments developed for particular conditions; for specific issues, the read-
er should consult specific reviews, such as for paediatrics [17], cardiac sur-
gery [18], trauma [19-21] or risk of sepsis [22].
Also, due to the general structure of this chapter, we will not revise scores
designed for the quantification and description of multiple organ dysfunc-
tion failure [23-25] or mixed systems [26]. The reader can find some guide-
lines of their use in Moreno et al. [27] and in Bernard [28].
The Existing Models
APACHE II
APACHE II was developed based on data registered between 1979 and 1982 in
13 hospitals of the USA [6]. The choice of variables and their weights was
selected by a group of experts, using clinical judgment and physiological rela-
tionships as documented in the literature.
The model uses the most deranged value from the first 24 hours in the ICU
of 12 physiological variables (scored from 0 to 4 points), age, surgical status
(emergency surgery, scheduled surgery or non-operative), and previous
health status. A main reason for ICU admission has to be chosen from a list
of 50 operative and non-operative diagnoses, in order to transform the
APACHE II score into a probability of death (in the hospital). The APACHE II
score varies from 0 to 71 points: up to 60 for physiological variables, up to 6
for age and up to 5 for previous health status. This system became the most
widely used of the general outcome prediction systems, and today it is still
used in a large number of ICUs.
118 R. Moreno, P. Metnitz