[The performance of intensive care services].

Area of Science:

• Critical Care Medicine
• Health Services Research
• Biostatistics

Context:

• Intensive care units (ICUs) require robust performance evaluation methods.
• Existing severity scoring systems utilize patient demographics, health status, and diagnoses.
• Predictive models estimate individual patient mortality risk.

Purpose:

• To outline the methodology for evaluating ICU performance using established severity scores.
• To explain the calculation of expected mortality rates based on prediction models.
• To describe the measurement of the Standard Mortality Ratio (SMR) for performance benchmarking.

Summary:

• ICU performance assessment involves measuring severity scores on admission and organ system dysfunctions throughout the stay.
• Severity scores incorporate patient age, prior health, disease severity, and diagnosis.
• Expected mortality is calculated by averaging individual patient risk predictions, then compared to observed mortality to derive the SMR.

Impact:

• Provides a framework for rigorous ICU performance evaluation.
• Enables benchmarking of ICU outcomes against national or international standards.
• Highlights the importance of data quality and population similarity for accurate performance metrics.