Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Predictive mortality models are not like fine wine.

Andrew A Kramer1

  • 1Cerner Corporation, 1953 Gallows Road, Suite 570, Vienna, VA 22182, USA. akramer@cerner.com

Critical Care (London, England)
|December 17, 2005
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Association of Intensive Care Unit Patient-to-Clinician Ratios with Mortality across Two U.S. Health Systems.

Annals of the American Thoracic Society·2025
Same author

Guiding Principles for Data Sharing and Harmonization: Results of a Systematic Review and Modified Delphi From the Society of Critical Care Medicine Data Science Campaign.

Critical care medicine·2025
Same author

ICU Staffing in the United States.

Chest·2024
Same author

Prospective Evaluation of a Dynamic Acuity Score for Regularly Assessing a Critically Ill Patient's Risk of Mortality.

Critical care medicine·2023
Same author

A Palace With a Common Tongue or a Multivariate Tower of Babel?

Critical care medicine·2022
Same author

The authors reply.

Critical care medicine·2021
Same journal

Comparative performance of mitral E/e' versus inferior vena cava variability for predicting fluid responsiveness in mechanically ventilated adults with shock.

Critical care (London, England)·2026
Same journal

Different lungs or different patients? Insights into postoperative ARDS.

Critical care (London, England)·2026
Same journal

Predictive validity of SOFA-2: analytical and contextual considerations.

Critical care (London, England)·2026
Same journal

Interpreting elevated FRAP in critical illness: beyond antioxidant capacity.

Critical care (London, England)·2026
Same journal

Cytokine-based strategies to improve prognostic enrichment of pediatric ARDS.

Critical care (London, England)·2026
Same journal

Genomic dynamics of antimicrobial resistance transmission between bacteria from intensive care unit surfaces and from critically ill patients.

Critical care (London, England)·2026
See all related articles

An updated Simplified Acute Physiology Score (SAPS) II model was developed using 1998-1999 data. However, this updated SAPS II may not accurately benchmark contemporary intensive care unit (ICU) performance due to declining predictive accuracy over time.

Area of Science:

  • Critical Care Medicine
  • Health Services Research

Background:

  • Established hospital mortality models, including Simplified Acute Physiology Score (SAPS) II, Acute Physiology, Age, and Chronic Health Evaluation (APACHE) III, and Mortality Probability Model (MPM)-II, developed in the early 1990s, exhibit declining predictive accuracy over time.
  • This decline is primarily due to calibration drift, where models tend to over-predict mortality when applied to more current patient data.

Discussion:

  • A recent update to the SAPS II model, based on 1998-1999 data, is presented.
  • The authors suggest its utility for benchmarking intensive care unit (ICU) performance.
  • However, concerns exist regarding the model's calibration for contemporary patient data (2005 onwards), potentially limiting its benchmarking applicability.

Key Insights:

  • Older mortality prediction models, including SAPS II, APACHE III, and MPM-II, show reduced accuracy with aging data.

Related Experiment Videos

  • Calibration deterioration leads to over-prediction of mortality in current populations, causing 'grade inflation' in ICU performance benchmarking.
  • The updated SAPS II model's suitability for benchmarking contemporary ICU data is questionable due to potential calibration issues.

    • Outlook:

    • The updated SAPS II model may serve historical research purposes.
    • Further validation is needed to assess its accuracy for benchmarking intensive care unit (ICU) performance with current patient populations.
    • Development of contemporary, well-calibrated mortality models is crucial for accurate ICU performance evaluation.