“In acute diseases it is not quite safe to prognosticate either death or recovery” Hippocrates

To control the exponential increase in health care costs in the United States, there has been a rapid move towards managed care systems. With this shift has come the public demand for the assessment of the quality of care delivered. Health care organizations are under increasing pressure to compare and contrast more accurately the quality of care provided so that therapy is refined and patient outcome is improved. This need is no more evident than in the intensive care unit (ICU) where high-technology, high-cost care is often provided that may neither improve patient outcome nor enhance quality of life. The quality of care delivered is, however, extremely complex, and like a good wine may be easy to recognize but extremely difficult to quantitate. In 1984, the American Medical Association (AMA) defined high-quality care as care “ which consistently contributes to the improvement or maintenance of quality and/or duration of life.”<sup>1 American Medical Association, Council of Medical Service Quality of Care JAMA 1986 ;  256 : 1032-1034

Cliquez ici pour aller à la section Références</sup> However, different perspectives on the definitions of quality result in different approaches to its measurement.

Monitoring and profiling deaths and complications such as infections, readmissions, reintubations, and iatrogenic events are commonly used as indicators of the quality of hospital care; however, these outcome variables are affected by many factors independent of the quality of care provided. Severity-adjusted outcome variables linked to cost are a scientifically more rigorous measure of quality of care.<sup>40 Marik P.E., Kraus P., Lipman J. Intensive care utilization: The Baragwanath experience Crit Care Med 1994 ;  22 : A41

Cliquez ici pour aller à la section Références</sup> These variables can be used to quantitate the quality of both acute and chronic care. Severity-adjusted death rates are commonly used quality indicators, as death is an unambiguous end point that is easy to measure. Not surprisingly, severity-adjusted death rates are the most frequently used quality indicators in the ICU.

For severity of illness in the ICU to be measured, a number of general multipurpose scoring systems have been developed over the past few decades.<sup>31 Knaus W.A., Draper E.A., Wagner D.P. , et al. APACHE II: A severity of disease classification system Crit Care Med 1985 ;  13 : 818-828  [cross-ref]

Cliquez ici pour aller à la section Références, 32 Knaus W.A., Wagner D.P., Draper E.A. , et al. The APACHE III prognostic system. Risk prediction of hospital mortality for critically ill hospitalized adults Chest 1991 ;  100 : 1619-1636  [cross-ref]

Cliquez ici pour aller à la section Références, 36 Le Gall J.R., Lemeshow S., Saulnier F. A new simplified acute physiology score (SAPS II) based on a European/North American multicenter study JAMA 1993 ;  270 : 2957-2963

Cliquez ici pour aller à la section Références, 38 Lemeshow S., Teres D., Klar J. , et al. Mortality prediction models (MPM II) based on an international cohort of intensive care unit patients JAMA 1993 ;  270 : 2478-2486

Cliquez ici pour aller à la section Références, 51 Pollack M.M., Ruttimann U.E., Getson P.R. Pediatric risk of mortality (PRISM) score Crit Care Med 1988 ;  16 : 1110-1116  [cross-ref]

Cliquez ici pour aller à la section Références</sup> Since physiologic data are readily available in the ICU and a relation is deemed to exist between acute physiologic derangements and the risk of death during acute illness, the degree of derangement of a number of physiologic variables has been used in the development of these scoring systems. General scoring systems have become the cornerstone of quality assessment because they allow the estimation of outcome probabilities that can be related to actual death rates as an indicator of effectiveness of care for groups of critically ill patients. The relation between the actual and the predicted outcome (standardized mortality ratio) is used to compare and monitor rates between hospitals and within hospitals over time.

It has been assumed that although ICUs admit very heterogeneous groups of patients who have large differences in age, previous health status, and acute health status the severity scoring systems can account for most of these characteristics with differences in outcome that are not explained by severity of illness attributed to variation in the quality of care. This article contrasts the various scoring systems and evaluates the validity of using these scoring systems for quality assessment.