ICU utilization represents 13.4% of hospital cost, 4.1% of national health expenditures, and 0.66% of gross domestic product. With the high cost and utilization of intensive care, it is important to understand quality of care. Data must be analyzed to understand how we can prevent patient mortality and morbidity, predict patient risk, and make care more efficient overall. Our two objectives were to: 1) Assess if quality metrics and measures accurately reflect the clinical care provided in the ICU, and 2) Examine if publicly reported outcomes (metrics and measures) reflect the quality of care provided in the ICU. We will review the common validated predictive scoring systems, their uses in the ICU, relative advantages, and their comparative efficacy. While predictive scores are valuable to research comparative groups, e.g., comparing performance of two different ICU units, they offer no assistance to the management of individual patients. Hence, we will discuss dynamic scoring systems, which offer real-time data to predict patient status. Examples of typical ICU quality metrics and hospital benchmarks will also be reviewed. Taking into consideration predictive scoring, dynamic scoring, quality and benchmark metrics, measurement combined with public reporting metrics can draw attention to particular areas of concern and stimulate improvement efforts. However, we argue that our current measurement isn’t enough; instead, it is a simplistic approximation of what administrators, regulators, and patients believe represents high quality of care. Artificial intelligence, cost-effectiveness analyses, and a systems approach (i.e., “clinically meaningful measurement into care delivery at appropriate points of interaction with patients combined with specific actions to ensure delivery of optimal care”) could help us best understand the quality and efficiency of our care.

Publication Date


Presented At:

14th Annual BHSF Research Conference

Content Type

Poster Presentation

Open Access

Available to all.